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

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

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

#rockbox log for 2008-04-03

00:02:11 Join Zarggg [0] (
00:03:26 Join corevette [0] (
00:04:28 Quit Galois (
00:04:28 Quit w0rd54 (
00:04:28 Quit Mouser_X (
00:04:28 Quit linuxstb__ (
00:04:28 Quit freqmod_nx (
00:04:28 Quit tvelocity (
00:04:28 Quit Battousai (
00:04:28 Quit ali_as_ (
00:04:28 Quit tedr0ck (
00:04:28 Quit tchan (
00:04:28 Quit Hodapp (
00:04:28 Quit knittl (
00:04:28 Quit ompaul (
00:04:48NJoinMouser_X [0] (
00:05:40 Join Spex [0] (
00:08:21 Quit Rob2222 (Connection timed out)
00:14:01 Part Spex
00:20:46NJoinompaul [0] (n=ompaul@gnewsense/friend/ompaul)
00:21:13bluebrotherBagder and others: what do you think about adding some background colors to headings in the wiki? Like the headings in the menu?
00:22:04bluebrotherI think this could make the wiki appear more structured
00:22:19 Quit Willaken- ()
00:24:24NJoinali_as_ [0] (
00:24:24NJoinfreqmod_nx [0] (
00:24:24NJointedr0ck [0] (
00:24:24 Join linuxstb [0] (
00:24:24NJoinknittl [0] (n=knittl@
00:24:24NJoinw0rd54 [0] (
00:24:24NJoinHodapp [0] (
00:24:24NJoinGalois [0] (
00:24:24NJoinBattousai [0] (n=bryan@gentoo/developer/battousai)
00:24:24NJointchan [0] (n=tchan@lunar-linux/developer/tchan)
00:24:44 Join scriptmunkee [0] (
00:28:52 Quit EspeonEefi ("さよなら")
00:29:30HExso, I've been looking into which player would be useful to attempt some hacking on for gsoc, and so far the sansa v2s look the most interesting
00:30:09HExbut I can't decide whether choosing a target that so little is known about is merely ambitious or actively insane
00:30:37scorche|shHEx: what sort of "hacking on"?
00:30:52bluebrotherHEx: someone posted a task to the tracker regarding the v2
00:31:02bluebrotheri.e. he posted code :)
00:31:15HExthe kind that involves getting rockbox to work. (not that I necessarily think I'll get that far...)
00:31:58 Join robin0800_ [0] (
00:32:03scorche|shHEx: well, the thing about GSoC is that you set a goal and achieve if your goal is to port rockbox to those devices, we are expecting that you will
00:32:45 Quit nplus ("Leaving")
00:35:23 Part scriptmunkee ("Leaving")
00:37:49 Join linuxstb_ [0] (
00:37:57*HEx looks
00:38:08Mouser_X o
00:38:53 Join Jon-Kha [0] (
00:39:10*DerPapst feels that bad things just happend. o.O
00:39:50amiconnHanging build
00:42:44amiconnWhy does svn commit not complain when giving -m twice!
00:43:36 Quit davina^ ("GNU/Linux the free alternative to Windows")
00:43:44 Quit Febs (Read error: 110 (Connection timed out))
00:44:30preglowit isn't usual for any program to complain when giving the same flag twive
00:44:32preglowtwice too
00:47:16amiconn-m isn't a flag, it takes a parameter
00:47:53amiconnI would rather not use -m, but in cygwin, svn doesn't find an editor to fire up if I leave it out
00:48:57 Quit robin0800 (Read error: 110 (Connection timed out))
00:48:57 Nick robin0800_ is now known as robin0800 (
00:48:58 Quit ender` (" Swipe cards don't work because every six weeks or so security sends someone around to clean them with a dirty cleaning card )
00:49:13 Join phinze [0] (
00:49:26Nico_Pamiconn: can cygwin lauch regular windows executables?
00:49:31Nico_Por maybe install an editor
00:49:45 Quit linuxstb (Read error: 113 (No route to host))
00:50:03amiconnIt can, but the gui editors usually run asynchronously
00:52:52preglowtell it to exec vim or something
00:52:56HExbluebrother: interesting
00:53:44bluebrotherexport EDITOR ...
00:54:13 Quit dabujo ("( :: NoNameScript 4.2 :: )")
00:54:17Nico_Pamiconn: they fork and give you your commandline back?
00:54:21kugelThere are some revisions missing in since-4weeks
00:54:30amiconnI now try to avoid accidental tree commits by entering -m "blah" last, but that obviously doesn't help if I accidentally put another -m *before* the filenames...
00:54:35Nico_Pkugel: probably commits to www/
00:54:47kugelah ok
00:55:05amiconnNico_P: Yes, the executable detaches (and for tabbed editors, the problem gets even more complicated....)
00:55:40*Nico_P suggests vim
00:55:47bluebrother.subversion/config can also hold a variable for an editor
00:56:44amiconnYeah, svn even tells how/where to configure, but I rather avoid vi* or similarly user-unfriendly editors unless forced to
00:56:55amiconn(don't even have one installed in cygwin)
00:56:57*bluebrother likes vim :)
00:57:12bluebrotherhow about nano?
00:57:54scorche|shbluebrother: i think last time we spoke about this, any editor that doesnt have a real gui is "user-unfriendly"
00:58:10 Quit HEx (Remote closed the connection)
00:58:45*Nico_P is starting to like vim too
00:58:59Nico_PI even start to press escape in kate-like editors
00:59:12 Join HEx [0] (
00:59:53bluebrotherhehe, happened to me too
01:02:49 Quit bluebrother ("user went to sleep")
01:03:04 Quit ompaul (Client Quit)
01:03:57 Join Nathan__ [0] (
01:08:23 Quit Nathan__ (Client Quit)
01:09:25 Nick Christopher1 is now known as ChristopherW (
01:12:00***Saving seen data "./dancer.seen"
01:14:51 Quit Zarggg ()
01:18:39 Quit arnath (Remote closed the connection)
01:25:21 Join Zarggg [0] (
01:30:38 Quit freqmod ("Konversation terminated!")
01:34:47 Quit Y-Signal ("CGI:IRC (Ping timeout)")
01:35:00 Part toffe82
01:46:26 Join argumentD [0] (
01:46:44 Quit Nico_P (Remote closed the connection)
01:47:40 Quit cool_walking_ (Remote closed the connection)
01:58:02amiconnWow, putting the greyscale timer isr on the cop really pays off...
01:58:40amiconnAlmost no load on the cpu -> smooth playback in mpegplayer without stuttering sound even on PP5002 (!)
01:59:03amiconnLoad on cpu is ~2% according to test_fps (versus 35% before)
01:59:14amiconnI need to solve a few cache coherency problems though
01:59:18 Quit phinze ()
01:59:45amiconnPlugins which put grey_info into iram work, but the others don't (yet)
02:01:05DerPapstyou sir, are a god :-)
02:02:20amiconnHmm, doom is also unstable
02:02:45*amiconn guesses there's a bug in the timer module still
02:03:59amiconnAh, yes
02:04:21*amiconn needs to make some variables NOCACHEBSS_ATTR
02:04:24amiconnjhMikeS: around?
02:12:55 Join FOAD_ [0] (
02:13:26 Join JdGordon|uni [0] (i=82c20d65@rockbox/developer/JdGordon)
02:14:12 Nick fxb is now known as fxb__ (
02:21:04kugelJdGordon|uni: ping :) Look at pls
02:21:46 Quit Thundercloud_ (Remote closed the connection)
02:21:52JdGordon|uniw00t @ fs#8843
02:22:28kugelAre you following me?
02:22:48JdGordon|unikugel: pastebin the whole unchanged file please....
02:22:55JdGordon|unii dont have the code here
02:23:00kugelah ok
02:23:12kugelunchanged? wait
02:23:18JdGordon|uniactually.. svn.rb works so dont worry
02:23:49JdGordon|uniwhat line number?
02:23:54kugelBTW: woot at FS #8843
02:23:58kugelabout 210
02:24:35 Quit roolku ()
02:25:02 Join cool_walking_ [0] (
02:26:39kugelHave you found the file?
02:28:18kugelline rather
02:30:19 Quit FOAD (Read error: 110 (Connection timed out))
02:30:20 Nick FOAD_ is now known as FOAD (
02:30:35 Join solace [0] (n=4b540f84@gateway/web/cgi-irc/
02:31:38solacei just installed rockbox on my ipod, but i dont know how to turn it off now, lol
02:31:44kugelJdGordon|uni: I'm just gonna post the patch, lets see if it works
02:32:03JdGordon|uniit will work but its not correct
02:32:23JdGordon|uniremotes will always show the selection marker even if the icon is set
02:32:29solacehow do i turn off my ipod now that it has rockbox?
02:32:51 Join keanu [0] (n=keanu@unaffiliated/keanu)
02:33:00JdGordon|unisolace: hold play
02:33:12solacei tried
02:33:22*amiconn wonders whether he found another PP quirk
02:33:24 Quit Jon-Kha (Remote closed the connection)
02:33:43JdGordon|unisolace: and what happened? you might need to hold it for 3 or 4 seconds
02:33:57*DerPapst wouldn't be surpised anymore
02:34:11solacethat seems to just pause it or something
02:34:35JdGordon|uniit should stop music, then shutdown after another half second or so
02:34:47solacethanks guys
02:34:48kugelJdGordon|uni: Ah ok
02:35:01solacepatience really is a virtue
02:35:13kugelWhich selection marker do you mean?
02:35:16 Quit solace (Client Quit)
02:36:05kugelJdGordon|uni: The current version just excludes the lines with pointer selector from being scrolled, since in that if the scroll functions are called
02:36:20JdGordon|uniyes, i can see the problem
02:36:24JdGordon|uniyour fix is almost correct
02:36:59kugelJdGordon|uni: Almost?
02:37:00JdGordon|unithe problem with your fix is that the first if () inside that block will be used on remotes with displays even if the icon selection is used
02:37:22JdGordon|uniwhich means it will show the cursor on the main screen, and the cursor AND the selection bar on remote
02:38:22 Quit amiconn (Nick collision from services.)
02:38:28 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
02:39:37*JdGordon|uni thinks about fixing this using putty on his phone because stupid uni doesnt let me ssh out on this wlan :(
02:39:56 Join Febs [0] (
02:40:18kugelWhat would complete my fix?
02:40:33kugeladding a if (!(global_settings.cursor_style)) ... ?
02:40:47JdGordon|uniyeah, but adding that in a nice way
02:43:12kugelI guess the intention should be to only show the pointer
02:43:16 Quit knittl (Read error: 110 (Connection timed out))
02:44:36kugelThat's why asked my drawmode question before
02:46:38kugelBut that will only work for remotes
02:46:51JdGordon|unino it wont
02:46:51kugelwhat will happen with other displays with depth < 16
02:46:55JdGordon|uniit will work for all
02:47:04JdGordon|unithen that happens anyway
02:47:20JdGordon|unithe other options shouldnt be legal if depth < 16
02:47:40JdGordon|unithat should work
02:47:47JdGordon|unitest it
02:48:02kugelYou only check for depth < 16 if remote_lcd is defined
02:48:46kugelAlso, you code still excludes pointer selector from scrolling
02:48:49JdGordon|unitrust me
02:48:51JdGordon|uniit will work
02:49:10JdGordon|uniarg, yeah, remove that global_settings.selectionbar check
02:49:55*JdGordon|uni goes to take a dump and charge his laptop, bbl
02:51:17amiconnHmm, seems the quirk is in pp5002.h, not in the PP itself
02:52:10 Quit JdGordon|uni (" ajax IRC Client")
02:52:56 Part pixelma
02:54:05kugelJdGordon: I don't quite get the #ifdef HAVE_REMOTE_LCD
02:56:00kugelWhat will happen with 8bit LCD_COLOR targets (in future)?
02:56:09 Quit gevaerts ("late enough")
02:57:53 Quit DerPapst ("*agrees with gevaerts* Night all :-)")
02:59:07 Quit ali_as_ ("zzzzzzzzzzzzzzzzzzzzzzzzzzzzzzzZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZZz")
02:59:53kugelJdGordon: It works without that #ifdef too
03:03:10kugelJdGordon: Ah, I see why you want the #ifdef
03:04:03kugelBut I stay at my opinion ;) I guess rockbox is just not prepared for 8bit color displays :) And I doubt that there'll be a need to
03:04:42 Join JdGordon|uni [0] (i=82c20d69@rockbox/developer/JdGordon)
03:05:18 Quit JdGordon|uni (Client Quit)
03:06:47 Quit argumentD (Read error: 110 (Connection timed out))
03:07:34 Join JdGordon|uni [0] (i=82c20d6a@rockbox/developer/JdGordon)
03:08:04 Join midkay_ [0] (
03:09:20JdGordon|uniamiconn: the quirk by chance doesnt have anything to do with DEV_USB0 does it?
03:09:39amiconnNo, but with the interrupt controller
03:09:44amiconn..on PP5002
03:10:04JdGordon|uniah, oh well
03:10:20kugelJdGordon|uni: I've uploaded the patch. With that #ifdef.
03:10:48JdGordon|unidid you test it on a non colour target?
03:11:38kugelNo, I trusted you, as you said ;)
03:12:02***Saving seen data "./dancer.seen"
03:12:30JdGordon|uniSTUPID! :D
03:13:08Hodappwow, Apple's firmware greatly annoys me when I'm trying to get to a particular point in an audio file.
03:13:14 Quit midkay (Nick collision from services.)
03:13:15 Nick midkay_ is now known as midkay (
03:13:34kugelJdGordon|uni: Works in ipodmini 2g sim
03:14:25kugelJdGordon|uni: That's why you wanted that #ifdef, non-color targets can't set anything but pointer and bar anyway
03:14:26JdGordon|unican you check the h300 sim also?
03:14:42kugelJdGordon|uni: Yea, I did.
03:15:24JdGordon|uniremote and main both work properly?
03:15:33kugelJdGordon|uni: Yes
03:15:44kugelI tested that before I trusted you actually
03:15:46JdGordon|unigreat, ill commit when i get home
03:15:54*JdGordon|uni slaps kugel
03:16:48amiconnThere are 7 possible lcd depth combinations
03:18:29amiconnMono main, 2-bit main, colour main; 2-bit main with mono remote (H1x0), 2-bit main with 2-bit remote (M5), colour main with mono remote (H300), colour main with 2-bit remote (X5)
03:18:35amiconnNo colour remotes yet
03:18:49JdGordon|uni... fortunatly... will be a pita to fix when we do
03:19:06JdGordon|unibut yes, for this section only colour and mono remote needs to be worried about
03:19:10amiconnThe handling should be symmetric wherever possible
03:19:16JdGordon|unia fair bit of code assumes a 1/2 bit remote lcd
03:19:22 Join AhrenJB [0] (
03:19:45amiconnI doubt that we'll see colour remotes any time soon though
03:19:49kugelJdGordon|uni: Like your code :)
03:20:00kugelJdGordon|uni: Some if it at least
03:20:07JdGordon|uninot like at all....
03:20:33kugelBut the fix you suggested me would go for bar selector on a 8bit color remote
03:21:07amiconnThe problem is the amount of data to transfer. Remotes are usually wired with a serial link (otherwise the cable would become too thick and too expensive)
03:21:14jhMikeSamiconn: here now
03:21:19JdGordon|uniwell.. the ifdeffing is only there to keep bin size down (probably reduces it by 32 bytes though :p )
03:22:05kugelWell, I go to bed now.
03:22:08amiconnjhMikeS: Just one question for now - do you have an idea how to declare a variable to use UNCACHED_ADDR?
03:22:09kugelgood night
03:22:19 Quit kugel ("ChatZilla 0.9.81 [Firefox]")
03:22:56jhMikeSamiconn: we don't have a way and I'm not sure how to pull it off with the linker script.
03:23:01amiconnI mean, without needing to touch all places where it's accessed. First, this would be an utter mess, and second, UNCACHED_ADDR breaks when applied to an iram variable
03:23:30amiconnAnd since this is a link-time decision, I can't add a compile-time counter-measure
03:23:41amiconn(it's about the greylib and TIMER2 on cop)
03:23:59jhMikeSI think the intent of NOCACHEx_ATTR was to ultimately be that instead of be an IRAM declaration
03:24:02amiconnIt's already sort-of working, and speeds things up nicely
03:24:23jhMikeSIt doesn't hurt video decoding performance?
03:24:40*jhMikeS guess a few FPS lost is better than skipping audio anyway
03:24:51amiconnI also found that pp5002.h is wrong - the pp5002 interrupt controller registers have the same structure as the PP502x ones
03:25:25amiconn(just at 0xcf0010xx instead of 0x600040xx)
03:25:38jhMikeSbasically the same controller then?
03:26:05amiconnPartially. I think the pp5002 doesn't have the HI interrupts
03:26:23amiconnBut registers 0x00..0x3c are identical
03:27:39jhMikeSIt's possible to logically locate a section at the uncached address and have it physically load into the proper ram section?
03:28:08amiconnSure it does hit video performance, but that's way better than choppy audio
03:34:23*AhrenJB Blinks Blankly
03:34:47AhrenJBIs there advice to be had here?
03:38:49 Join midgey [0] (
03:40:34AhrenJBAh, the old IRC waiting game.
03:41:18JdGordon|uniAhrenJB: you didnt ask a question....
03:42:05JdGordon|uniadvice isnt exactly something tangible you can point to and say "look, theres advice sitting right there!"
03:42:50jhMikeSyes, AhrenJB asked, "Is there advice to be had here?" which sounds interrogative enough. :p
03:44:53*jhMikeS notices the presence of advice preceding his previous post
03:48:41AhrenJBWhat is the cause of certain audiofiles not showing up in firmware? RockBox or original.
03:49:06JdGordon|uniflux capacitor mis calibration...
03:49:10JdGordon|unior unsupported filetype
03:49:26AhrenJBAhem, I mean files which are supported not showing up
03:49:53JdGordon|uninot showing up where?
03:50:04JdGordon|unitry being specific and adding more information
03:50:16AhrenJBSay I add 25 MP3's or WMA's, from different sources and whatnot. All play in WMP, but one or two wont be detected in Sansa Firmware music lists, and maybe the same or a couple different in RockBox?
03:50:31AhrenJBIn the database
03:50:58JdGordon|uniwe dotn give support for the sansa firmware.. but them not showing up in rockbox is probably bad tags
03:51:38AhrenJBJust wondering if there was some quirk with the sansa which caused this
03:53:30AhrenJBHave there been any significant changes in the Sansa e-series current release since ermm February last year
03:58:19AhrenJBMajor additions to functionality, mentionable optimization, etc?
04:00:42tatootiani need to figure out how to use flac with an iPod
04:00:46tatootianiTunes doesn't let me >.>
04:01:02AhrenJBPower management... Good enough for me.
04:01:06cool_walking_tatootian: Copy-paste it onto the hard drive.
04:01:45tatootianah, right
04:02:53cool_walking_... Which brings me back to:
04:05:23 Quit AhrenJB ()
04:17:58 Join kreativethinkr [0] (n=485c019c@gateway/web/cgi-irc/
04:18:09kreativethinkr does Sansa e200R/Rhapsody rockbox have to be formatted fat32 ?
04:18:29 Join m0f0x [0] (
04:20:33jhMikeSI thought fat16 was handled on e200 since SD cards can be
04:20:58JdGordon|uniok, fat16 also
04:21:14*JdGordon|uni assumes kreativethinkr was mroe asking if he could format it ntfs or something
04:21:14jhMikeSnot that the main storage should be formatted fat16 from the factory
04:21:24 Join ol_schoola [0] (
04:22:01kreativethinkrmy e200 has to be fat32 to have rockbox to work how about sansa e250 R ?
04:22:26JdGordon|uniyour and everyone elses...
04:23:17advcomp2019kreativethinkr, yes.. it needs to be fat32
04:23:45kreativethinkrTHANKS ALOT
04:28:16kreativethinkris it the same step like in the e250 ?
04:29:40 Join webguest90 [0] (n=47394695@gateway/web/cgi-irc/
04:30:14 Join miepchen^schlaf_ [0] (
04:37:57 Quit miepchen^schlaf (Connection timed out)
04:40:33 Quit kreativethinkr ("CGI:IRC")
04:45:42 Join atm [0] (n=atmb4u@
04:45:52 Quit webguest90 ("CGI:IRC (EOF)")
04:46:27 Join webguest90 [0] (n=47394695@gateway/web/cgi-irc/
04:46:28 Quit webguest90 (Client Quit)
04:49:01 Part HellDragon
04:50:33atmhi every one
04:52:54 Quit JdGordon|uni (" ajax IRC Client")
04:53:34advcomp2019hello atm
04:56:08atmI would like to port Rock box to Android
04:56:41atmbut there is a problem.Android is fully java and RockBox is C
04:57:09atmSo should I convert Rock Box to java or any other option???
04:57:33 Quit BlakeJohnson86 ("Leaving.")
04:58:21advcomp2019atm, you will need wait for someone that will know better
04:59:40 Join BlakeJohnson86 [0] (
04:59:59atmI want your opinion please.............
05:00:47scorcheatm: converting Rockbox to java would be a LOT of work
05:01:05scorchethis isnt the Google Decade of Code ;)
05:02:03atmjust trying
05:02:27atmis there any other way scorche?
05:04:18atmwhat's your opinion?
05:04:56atmscorche, as android is fully java based.......
05:05:35LloreanI thought Android was entirely open source. That would suggest there's a way to write lower level code, rather than only JAVA apps/bytecode.
05:06:43LloreanBut, if it is explicitly JAVA-only, there's really no appropriate solution. Rockbox is C because it's more or less the best language for the job. So you'd have to maintain a C and JAVA version of Rockbox separately if such a conversion were to happen.
05:06:47*HEx opines that running java on embedded devices is silly, and that people who do such things can and should be ignored
05:09:04atmHEex, but android is not silly
05:09:10LloreanHEx: Whether it's a good or bad idea is irrelevant here.
05:09:50 Join |AhIoRoS| [0] (n=ahioros@
05:10:09HExLlorean: true, that was unhelpful of me. I'll apologize and go back to sleep now
05:10:44 Quit m0f0x ("sleep")
05:11:03 Quit S1gn ("leaving")
05:12:04***Saving seen data "./dancer.seen"
05:12:25atmLlorean, if there is a chance for using C in android what shall I prefer?
05:13:53 Quit cool_walking_ (Remote closed the connection)
05:14:07Lloreanatm: Rockbox is written in C. I don't think changing to JAVA is a realistic option, but if you can use C then it's much more possible.
05:20:14 Join cool_walking_ [0] (
05:20:19atmLlorean,But don't you think after all C has to be changed to java ? More easier to debug?(Yes/No)
05:21:57LloreanIt might be easier to debug or work with on Android, but it would essentially split the codebase, as JAVA is not at all suitable for the other Rockbox platforms.
05:22:06 Join homielowe [0] (
05:23:10 Join BlakeJohnson87 [0] (
05:25:06atmwhat about a rock box only for Android?
05:25:13atmA bad idea..
05:26:08LloreanIt's possible, but it would quickly become out of sync. Someone or a group of people would have to take responsibility for porting features between the two.
05:26:13LloreanIt would, in essence, be a parallel fork.
05:36:41 Quit BlakeJohnson86 (Read error: 110 (Connection timed out))
05:37:33 Join Davide-NYC [0] (
05:37:38 Quit csc` (Read error: 110 (Connection timed out))
05:40:37Davide-NYCAny word on the new recording screen? :)
05:41:00 Quit Horscht ("Snak 5.3.3 Unregistered copy. Evaluation period is over. Program will now quit. Thanks for using Snak.")
05:46:44 Join at1 [0] (n=atmb4u@
05:47:47 Part at1
05:48:07 Quit atm (Read error: 110 (Connection timed out))
05:57:13 Join andrew[andrboot] [0] (
06:00:16*Davide-NYC crawls back under his desk
06:00:17 Quit ol_schoola (Read error: 104 (Connection reset by peer))
06:00:22 Quit Davide-NYC ("ChatZilla 0.9.81 [Firefox]")
06:00:25 Part andrew[andrboot] ("RAWR")
06:00:27 Join ol_schoola [0] (
06:00:34 Quit ol_schoola (Client Quit)
06:04:45 Quit Rob2223 ()
06:06:10 Join tatoot [0] (
06:06:10 Quit tatootian (Read error: 104 (Connection reset by peer))
06:09:17 Join Rob2222 [0] (
06:09:42 Quit simonrvn ("Reconnecting")
06:09:42 Join simonrvn_ [0] (i=simon@unaffiliated/simonrvn)
06:09:47 Nick simonrvn_ is now known as simonrvn (i=simon@unaffiliated/simonrvn)
06:10:49 Quit Mouser_X (Nick collision from services.)
06:10:58 Join Mouser_X [0] (
06:11:14 Join robin0800_ [0] (
06:13:17 Join homielowe_ [0] (
06:13:35 Join phinze [0] (
06:13:37 Quit robin0800 (
06:13:37 Quit corevette (
06:13:37 Quit shodanX (
06:13:37 Quit Addicted (
06:13:37 Quit BHSPitMonkey (
06:13:37 Quit gromit` (
06:13:38 Nick robin0800_ is now known as robin0800 (
06:13:49 Join saratoga [0] (n=41becb3b@gateway/web/cgi-irc/
06:14:07NJoincorevette [0] (
06:14:07NJoinBHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
06:14:07NJoingromit` [0] (
06:14:07NJoinshodanX [0] (
06:14:07NJoinAddicted [0] (
06:19:44saratogaif that guy comes back, tell him we'd be interested in the idea on the GSOC page
06:19:53 Quit shodanX (
06:19:53 Quit Addicted (
06:19:53 Quit BHSPitMonkey (
06:19:53 Quit gromit` (
06:19:53 Quit corevette (
06:20:00saratogathat is a way to run rockbox on Andriod, not port it to java
06:20:38 Quit phinze (Remote closed the connection)
06:20:41NJoincorevette [0] (
06:20:41NJoinBHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
06:20:41NJoingromit` [0] (
06:20:41NJoinshodanX [0] (
06:20:41NJoinAddicted [0] (
06:21:14 Join phinze [0] (n=phinze@
06:23:13 Join atm [0] (n=atmb4u@
06:25:03 Quit homielowe (Read error: 110 (Connection timed out))
06:25:36 Join andrew[andrboot] [0] (n=andrboot@unaffiliated/andrewandrboot/x-689432)
06:26:24 Quit andrew[andrboot] (Client Quit)
06:26:39 Join andrew[andrboot] [0] (n=andrboot@unaffiliated/andrewandrboot/x-689432)
06:26:47 Part atm
06:32:05 Quit simonrvn ("Reconnecting")
06:32:13 Join simonrvn [0] (i=simon@unaffiliated/simonrvn)
06:34:08 Quit perrikwp (" ajax IRC Client")
06:36:04 Join perrikwp [0] (i=982163ab@gateway/web/ajax/
06:37:55 Quit Febs (Read error: 110 (Connection timed out))
06:47:19 Quit midgey ()
06:47:44 Quit saratoga ("CGI:IRC (EOF)")
06:48:15 Join midgey [0] (
06:53:25 Join FOAD_ [0] (
06:58:31 Join phinze__ [0] (
06:58:42 Quit midgey ()
07:01:45 Quit homielowe_ (Read error: 110 (Connection timed out))
07:10:33 Quit FOAD (Read error: 110 (Connection timed out))
07:10:33 Nick FOAD_ is now known as FOAD (
07:11:16 Quit phinze (Connection timed out)
07:12:07***Saving seen data "./dancer.seen"
07:19:24 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:27:26 Join WeaponX [0] (
07:27:31WeaponXquick question
07:27:38WeaponXi saw the album art demo in rockbox
07:27:51WeaponXsorry i meant album flow
07:28:03WeaponXdoes it do anything or is there something i have to activate to make it work?
07:29:18 Quit miepchen^schlaf_ ()
07:29:19 Quit phinze__ ()
07:30:03 Join pondlife [0] (
07:30:22 Join phinze [0] (
07:32:50 Join atm [0] (n=atmb4u@
07:33:00 Part atm
07:36:09cool_walking_As far as I know, all it does is scroll through your album covers i.e. it currently can't play music.
07:37:08 Join atm [0] (n=atmb4u@
07:37:40cool_walking_atm: [00:20] <saratoga> if that guy comes back, tell him we'd be interested in the idea on the GSOC page
07:37:52cool_walking_atm: [00:20] <saratoga> that is a way to run rockbox on Andriod, not port it to java
07:37:52 Quit phinze ()
07:37:54 Join JdGordon_ [0] (
07:37:59atmI'm gack
07:38:08atmsorry I'm back
07:39:21 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
07:41:06pondlifeMorning Nico_P, thanks for doing that commit..
07:41:16pondlife..has it broken anything much? ;)
07:41:21Nico_Pthanks for writing a fix ;)
07:41:30Nico_Pnot that I know of
07:41:33*pondlife likes svn blame sometimes
07:43:29Nico_Ppondlife: I've been experimenting with changing ci.new_track in audio_skip instead of having audio_skip posting a message to the audio queue
07:43:53pondlifeHmm, that sounds dangerous. What's the advantage?
07:44:19Nico_Pno more use of wps_offset for manual track skipping
07:44:49Nico_Pand a slightly less contorted skipping process
07:45:00pondlifeYes, but ci.new_track will then be more prone to multi-thread problems, and it's more "important" than wps_offset.
07:45:00Nico_Pbut as you said, I'm not sure it's safe
07:45:38pondlifeI'd suggest we try to scrap track skipping, and replace it with a "play track n" interface
07:46:05pondlifeMake use of buffering a bit more; if the track is buffered then it should just be playable anyway
07:46:08Nico_Pyes, I've been thinking about that... will the playlist API allow it?
07:46:17pondlifei.e. reduce the number of code paths (lots)
07:46:37pondlifePlaylist API needs a shake-up anyway.
07:46:54pondlifeMaybe start with a survey of the playlist calls to see which are used by MASCODEC
07:47:00pondlifeThose are sacrosanct
07:47:12pondlifeThe rest could be reworked as required
07:47:26pondlifeIt's not a small job :/
07:48:24pondlifeI also thought that perhaps a playlist call could be added to prepare the next playlist without doing a skip, and use that in audio_change_new_track()
07:48:39pondlifeI'd really like that auto_change_dir and end_of_playlist died
07:48:46pondlifeThe variables, I mean.
07:48:46 Join at1 [0] (n=atmb4u@
07:48:47Nico_Pme too
07:49:32Nico_Pwhat's a pity is that with auto dir change there is no pre-buffering of the next dir's tracks
07:50:06pondlifeWell, it works ok for Archos
07:50:13Nico_PI wonder if somehow playlist_peek or its eqivalent could return the tracks from the following di
07:50:18pondlifeA delay between dirs is acceptable IMHO
07:50:41pondlifeBetter than instability
07:50:41Nico_Pprobably, yes
07:51:03pondlifeWe never said that we do gapless across directories/playlists
07:52:19Nico_PI think I'll try to study the HWCODEC code this afternoon
07:53:10JdGordon_I know im going to sound like a bloody broken record here... but i do intend to completly rework and clean up playlist so we can use it properly and correctly
07:53:48pondlifeI don't know if you saw, but I suggested that the Archos sim might help
07:53:52 Quit JdGordon (Read error: 110 (Connection timed out))
07:54:10pondlifeNo idea if it calls the playlist code properly, but it might
07:54:27pondlifeThen see if we can use the same strategy in playback.c
07:54:47JdGordon_there is very little mpeg.c->playlist.c happening which is good
07:55:28JdGordon_only _peek() _check(), _next() _amount() and _update_resume_info()
07:55:58JdGordon_need to check what the difference between the first 2 are and what the last is for, but the other two are "correct"
07:56:52JdGordon_crap, ok im going to find some caffeenee then actually start!
07:56:53Nico_PI'm not sure playback.c does much more
07:57:43pondlifeI'd suggest revising playback.c first, to simplify testing of any playlist rework...
07:58:05 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
07:58:40 Quit WeaponX ("If You Can't Be Famous, Be Infamous...")
08:00:33*Nico_P needs to leave for shcool
08:00:56 Join at2 [0] (n=atmb4u@
08:01:23 Quit atm (Read error: 110 (Connection timed out))
08:01:35 Join davina [0] (
08:01:58at2I'm back
08:02:25 Join FOAD_ [0] (
08:05:31 Join atm [0] (n=atmb4u@
08:10:24LinusNJdGordon_: "completely rework and clean up playlist"?
08:10:35 Part atm
08:11:30 Join GodEater [0] (i=c2cbc962@rockbox/staff/GodEater)
08:11:46LinusNiirc, the playlist code is one of the few components in rockbox that has "just worked" for ages
08:12:00BagderrI agree
08:12:15Bagderrand not only has it "just worked" it is a piece of well-thought functionality
08:12:26scorcheBagderr: did you see what i asked earlier about the ads?
08:12:35JdGordon_yeah, well its broken
08:12:53JdGordon_and dont worry.. im nto really going to rework it
08:13:05Bagderrbroken things should be fixed
08:13:12scorcheis the deal just for, or is it a general type of deal that also can apply to the subdomains like forums.r.o themes.r.o, etc
08:13:33Bagderrscorche: zagor's the man for that, but I _think_ the deal is for the front page only
08:14:06LinusNJdGordon_: in which way is it broken?
08:16:06 Nick JdGordon_ is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
08:16:17 Quit at1 (Read error: 110 (Connection timed out))
08:17:39JdGordonarg, you got me... ok, its not broken, imo its a mess though... 15 functions for the current playlist, another 20odd for a generic playlist struct (which imo sthat handling should be removed to trim it a bit)
08:17:45JdGordonthe database loading takes ages
08:18:16pondlifeJdGordon: That last bit needs looking into
08:18:43pondlifeMaybe once playback.c is simplified some more, some of the playlist code could be removed.
08:18:53pondlifee.g. my fix for disk spinup!
08:19:22JdGordonat the very least, dirplay handling should be completly playlist.c
08:19:23 Quit FOAD (Read error: 110 (Connection timed out))
08:19:23 Nick FOAD_ is now known as FOAD (
08:19:56*JdGordon for some reason has the (obviously wrong) philosophy of starting off again with a clean slate is better than bandaids
08:20:36Bagderrrewrites are overrated in general
08:20:55 Join Rob2223 [0] (
08:21:01Bagderr"bandaid" is called evolutionary and iterative development
08:21:01LinusNrefactoring is often a good thing, but not always
08:21:41Bagderrrefactoring when done piece by piece committed is a good approach
08:22:29JdGordonpondlife: I wonder if for DB dirplay we could store the search and reuse it to get tracks as needed
08:22:45LinusNthe playlist code may seem convoluted and complicated, but it is a result of careful optimization for targets with very limited CPU and RAM
08:22:47andrew[andrboot]ppl awake
08:22:48JdGordonwe tend to have quite large peices here though
08:23:13pondlifeJdGordon: DB dirplay?
08:23:28LinusNit would look completely dfferent it we had written it for today's targets
08:23:31JdGordondirplay but from the db instead of file browser
08:23:53 Join petur [50] (n=petur@rockbox/developer/petur)
08:24:01JdGordonLinusN: how so? we still have the same requirements of low ram and speed
08:24:03pondlifeJdGordon: Hmm, I don't understand - use the file browser to make dir-based playlists
08:24:32pondlifeThe browsers are basically playlist generation tools
08:24:35LinusNalso, the database mode is something of a hack on top of the playlist/playback engine
08:24:47JdGordons/dirplay/inram playlist
08:25:09LinusNJdGordon: yes we do, i just tried to explain why it works like it does
08:25:22pondlifeWe should generally think of "per-playlist" and never "per-directory"
08:25:41pondlifeAuto-change dir is an exception to that, of course
08:25:50 Quit at2 (Read error: 110 (Connection timed out))
08:26:07pondlifeAuto-change playlist (i.e. incorporating the database) might be an option I suppose, but not easy
08:26:59JdGordonI also want to allow for auto-change-playlist working with playlist catalogue so you could (randomly if enabled) jump between m3u's as well
08:27:02 Quit BigBambi (Remote closed the connection)
08:28:22JdGordonand the last thing I want changed is the ability to move the insert position so inserting is more obvious
08:29:05JdGordonpossibly allowing us to remove one or more insert options
08:32:02pondlifeI suppose playlist catalogue is another playlist generator...
08:32:19*pondlife is getting a bit OO here.... :/
08:36:13 Quit spiorf (Remote closed the connection)
08:38:17 Quit Rob2222 (Read error: 110 (Connection timed out))
08:39:43 Quit Hodapp (Read error: 110 (Connection timed out))
08:40:40LinusNJdGordon: how would you want insert/queue to work instead?
08:41:11LinusNi agree the the insert position could be a bit confusing
08:41:39JdGordonI'm not sure, its just an easy-ish way to fix that confusion
08:42:12 Quit courtc (Remote closed the connection)
08:42:15 Join courtc [0] (
08:42:32JdGordoninsert, insert last and insert shuffled.. insert next i guess could stay but it wouldnt be needed if the insert position could be moved easily
08:43:21JdGordonactually, if the playlist viewer didnt need plugin ram "insert at..." could open it and let the user place the tracks exatcly where they want it without any fuss
08:44:00 Join spiorf [0] (
08:45:32LinusNwhy does the need for plugin ram prevent that?
08:46:48JdGordonit doesnt, its just not nice when you cant use it because something else is using it (a TSR maybe.. )
08:47:15JdGordonthats another thing.... it shouldnt need extra ram unless its a m3u playlist
08:54:02 Join ender` [0] (
09:09:37 Join krz [0] (
09:12:11***Saving seen data "./dancer.seen"
09:19:40 Quit bnakifhll (Read error: 104 (Connection reset by peer))
09:21:03 Join onakijmeo [0] (i=0@
09:24:27 Join Hodapp [0] (
09:26:18 Join tvelocity [0] (
09:28:58 Join framo [0] (
09:35:12*scorche shakes his head at LinusN for reusing the WMA quote on a curl ML
09:35:34LinusNi couldn't resist
09:36:09 Join DerPapst [0] (
09:39:21 Join atm [0] (n=atmb4u@
09:46:59 Quit fyrestorm (Read error: 113 (No route to host))
10:07:23 Quit atm (Read error: 110 (Connection timed out))
10:07:25 Join at1 [0] (n=atmb4u@
10:09:41 Quit framo ("moo")
10:13:14 Join davina^ [0] (
10:13:18 Quit davina (Read error: 104 (Connection reset by peer))
10:15:27 Join PaulJam [0] (
10:15:42 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
10:17:06 Part andrew[andrboot] ("RAWR")
10:17:23 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
10:21:28 Join Jon-Kha [0] (
10:22:36 Part Bagderr
10:35:49 Quit at1 (Read error: 110 (Connection timed out))
10:37:05 Join atm [0] (n=atmb4u@
10:38:08 Part atm
10:40:35 Join Mathiasdm [0] (
10:49:27 Join gevaerts [0] (
10:59:11 Join moos [0] (
10:59:37 Join [CBR]Unspoken|w [0] (n=cbr@
11:00:16 Quit Mathiasdm ("Yuuw!")
11:02:36 Quit cool_walking_ (Remote closed the connection)
11:03:27 Join nplus [0] (
11:12:14***Saving seen data "./dancer.seen"
11:18:33 Quit courtc (Read error: 113 (No route to host))
11:24:42 Join knittl [0] (n=knittl@
11:27:00 Join courtc [0] (
11:29:30amiconnJdGordon: The (rest of the) plugin buffer is guaranteed to be available in the core. This is because a TSR plugin is not allowed to use plugin_get_buffer() by convention.
11:30:26amiconnJust don't make any assumption about the size of the plugin buffer, but use plugin_get_buffer(). That's what the playist viewer does, btw
11:35:32 Part LinusN
11:39:37 Join mcuelenaere [0] (
11:40:03mcuelenaerehi, I need to locate a freshly formatted FAT32 partition on a hard drive
11:40:10mcuelenaerebut it is hidden in an unknown filesystem
11:40:15mcuelenaerepossibly in a file on it
11:40:42mcuelenaereI thought about doing a grep on NTLDR, but grep doesn't give me the string's position
11:40:51mcuelenaerewhat would be the best method?
11:42:29gevaertsmcuelenaere: maybe can help ?
11:43:26mcuelenaeregevaerts: thanks, I'm going to try that one :)
11:44:35 Join Thundercloud [0] (
11:50:51 Quit GodEater (" ajax IRC Client")
11:51:19 Join GodEater [0] (i=c2cbc962@rockbox/staff/GodEater)
11:58:27 Join ali_as [0] (
12:06:47 Join dabujo [0] (
12:33:55 Join Mathiasdm [0] (
12:40:18 Join roolku [0] (
12:41:32 Quit mcuelenaere (Read error: 104 (Connection reset by peer))
12:41:34 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
12:42:04 Join Febs [0] (
12:47:58 Join tvelocity_ [0] (
12:51:55 Join OlivierBorowski [0] (
13:03:18 Quit tvelocity (Read error: 110 (Connection timed out))
13:08:24 Join atm [0] (n=atmb4u@
13:08:55 Part atm
13:12:15***Saving seen data "./dancer.seen"
13:13:26 Join MethoS- [0] (
13:19:45 Join at2 [0] (n=atmb4u@
13:20:02 Quit markun (Read error: 110 (Connection timed out))
13:21:45at2Please explain the feasibility of porting RockBox to Android
13:22:05Bagderrockbox is an OS and app for running native on a fixed-size LCD
13:22:18BagderAndroid is a java framework
13:22:48Bagdereven though it runs on an ARM arch
13:23:35Bagderso perhaps you should rather tell us how you think it should run
13:24:13at2Android has some capabilities of ARM arch
13:24:26 Join mf0102 [0] (n=michi@
13:24:38at2Utilizing that can't we
13:24:49Bagderwell, for it to be rockbox it needs to be C code that gets compiled
13:25:33Bagderat2: have you checked out the rockbox sources? Have you done any ARM-based android hacking?
13:27:51 Quit DerPapst (Read error: 113 (No route to host))
13:31:13 Quit OlivierBorowski (Remote closed the connection)
13:31:26pondlifeat2: Does Android have a POSIX+SDL-like interface?
13:31:50pondlifeYou might be able to use our simulator configuration as a starting point
13:41:27linuxstbReading the Android FAQ, it specifically says you can only use Java, not C/C++ -
13:42:11pondlifeOK, well that clears things up!
13:43:18linuxstbAh, so Android is based on a Linux kernel?
13:43:39Nico_Pdid they release their patches btw?
13:44:00Bagder too
13:44:02*Bagder runs off
13:44:13linuxstbAre there standard ways to access C libs from a Java app?
13:45:20 Quit at2 (Read error: 110 (Connection timed out))
13:45:31linuxstbIMO, someone needs to do a "proof-of-concept" demo to show that the Rockbox core could be interfaced with an Android UI...
13:46:47 Join Tuplanolla [0] (
13:47:24 Join atm [0] (n=atmb4u@
13:48:01*Nico_P needs a reminder. codecs like vorbis, speex, FLAC, MPC, etc... need the full file to be able to seek, don't they?
13:48:27atm yes
13:48:35atmI think
13:49:41GodEaterin there an android vm you can try that against linuxstb?
13:49:52linuxstbNico_P: What do you mean by "need the full file" ?
13:50:04linuxstbGodEater: I saw reference to some kind of emulator on google's site...
13:50:31Nico_Plinuxstb: I'm thinking of resume in particular... currently we only chop the beginning off for MPA
13:50:32linuxstb"a true device emulator"
13:51:31gevaertsHow much of android is really specified ? Just the java apis or more ?
13:52:06*GodEater wonders how useful a target android really is
13:52:16GodEatersurely a Symbian flavour would be a better audience ?
13:53:25linuxstbNico_P: Some formats (probably all that use a container) need to read the header information from the start of the file.
13:53:40gevaertsI think that Symbian will at least be a lot easier. It has "real" support for native applications, so only the rockbox-specific work has to be done (which IMHO is already a lot)
13:54:30GodEaterAllegedly there's an SDL port for symbian, but the libs are all Windows based
13:54:34linuxstbNico_P: One nice improvement could be for get_metadata() to do this reading, although it would add a little to the core binsize (but not much I think - the existing parsers are probably very close to doing that already)
13:55:01GodEaterat least, the ones I found were
13:55:17gevaertsGodEater: yes, but there _are_ libs... For Android only java actually seems to be documented and supported
13:55:20GodEaterwhich leads to two mutually exclusive dev environments. *nix for rockbox, and windows for Symbian
13:55:31linuxstbWould compiling Rockbox's C code to Java bytecode be feasible? (I've no idea if/how well that works...)
13:55:50GodEaterI would imagine that would be really painful
13:56:09GodEateris that the sort of thing gcj can do ?
13:56:29Nico_PGodEater: I think GCJ is the opposite: compile java to native code
13:56:38Nico_Pinstead of bytecode, that is
13:56:41GodEaterah - think you're right
13:56:47GodEaternow that I dust of the grey cells
13:57:07linuxstbGodEater: I used gnupoc successfully in the past, but I don't know if that's still current...
13:57:11Nico_Plinuxstb: codecs aren't supposed to set id3->elapsed directly, aren't they?
13:57:39linuxstbNico_P: I'm not sure.
13:58:27Nico_PI'm not sure either and I don't see what possible harm there could be in doing it apart from not respecting the PCM latency value
13:59:16Nico_PI'm asking all these questions because I've just moved audio_set_elapsed to mpa.c and although it works I want to make sure I'm not doing anything wrong
13:59:45 Join JdGordon|uni [0] (i=d31c5f5a@rockbox/developer/JdGordon)
14:00:01 Nick JdGordon|uni is now known as JdGordon| (i=d31c5f5a@rockbox/developer/JdGordon)
14:05:37 Join cool_walking_ [0] (
14:06:21Nico_Plinuxstb: to answer the question you asked earlier, to access native code in java there is the KNI
14:06:34Nico_Pdoh. the JNI (java native interface)
14:08:24 Quit atm (Read error: 104 (Connection reset by peer))
14:10:20 Join midgey [0] (
14:10:48 Quit midgey (Client Quit)
14:11:19 Join midgey [0] (
14:13:37 Join Tri [0] (n=fr33l4nc@
14:13:43 Join atm [0] (n=atmb4u@
14:13:57linuxstbNico_P: Thanks - googling for android+jni gives examples. So it seems feasible.
14:14:39linuxstbWhich could work quite nicely - creating a library from the Rockbox apps code would open up its use in lots of places...
14:15:59Nico_Pindeed. eg it could be useful for a WPS designer
14:16:04 Join B4gder [241] (n=daniel@rockbox/developer/bagder)
14:17:03Nico_Pwhy isn't the build system picking up my commit?
14:17:49 Quit Febs (Read error: 110 (Connection timed out))
14:18:58B4gderit happens at times, I haven't figured out why yet
14:19:06 Quit midgey ()
14:23:08 Join TMM [0] (
14:23:08linuxstbNow that we have a lot of student applications (including some duplicates for the same projects), what do people think of asking the applications to perform a task to help us assess their abilities? Probably something related to their project?
14:23:34linuxstbs/applications/students/ (the second use of "applications")
14:24:04*gevaerts agrees, but won't be tricked into thinking of lots of tasks ;)
14:24:44linuxstbe.g. the Android applicant could be asked to implement that "proof-of-concept" demo linking C and Java within Android.
14:25:03 Join at2 [0] (n=atmb4u@
14:26:16linuxstbMaybe the people applying for the theme editor project could be asked to create a Rockbox theme...
14:26:34*linuxstb isn't sure what the usability study applicants could do...
14:26:57gevaertsIm not sure if 'creating a theme' is the right test.
14:27:20at2am I not visible to you?
14:27:22 Quit |AhIoRoS| (Remote closed the connection)
14:27:34B4gderbtw, us being 12 mentors I think we can ask for about 6 projects to get 2 mentors/student.
14:27:34 Part at2
14:27:36linuxstbI'm not either - but at least it would show the serious applicants (people willing to spend time on it)
14:27:39gevaertsat2: you are
14:27:58linuxstbB4gder: Yes, but not 6 usability studies... ;)
14:28:14gevaertsBagder: agreed. That should be manageable
14:28:20B4gdernah, I would personally probably rate them below #6 ...
14:29:04 Join parafin|away [0] (
14:29:08 Join at2 [0] (n=molly@
14:29:21gevaertslinuxstb: that is of course a minimum, but it doesn't really give any idea about skills
14:32:29 Quit Tri ("Leaving")
14:35:07linuxstbat2: Did you read our comments about Android? You seem to be losing your IRC connection a lot...
14:36:10 Quit parafin|away ("So long and thanks for all the fish")
14:36:12 Join parafin|away [0] (
14:36:12 Quit JdGordon| (" ajax IRC Client")
14:36:44 Quit atm (Read error: 110 (Connection timed out))
14:38:02B4gderwe have roughly 9 different ideas, 4 of them exist in multiple versions
14:39:23 Quit parafin (Read error: 113 (No route to host))
14:40:15 Quit B4gder ("It is time to say moo")
14:40:45krzlinuxstb: what tests should be applied to the applicants with the theme editor?
14:41:05linuxstbkrz: I don't know...
14:41:14 Quit Mathiasdm (Read error: 104 (Connection reset by peer))
14:41:19 Join Mathiasdm2 [0] (
14:41:37linuxstbBut we have multiple applications, so need some way to decide who is the "best" applicant.
14:42:12krzhm, may be showing some projets should help?
14:42:51 Quit moos ("Rockbox rules the DAP world")
14:43:03 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
14:43:36krzi can offer some updated themes (which where done for an unofficial build and didn't work on official)
14:44:03krzmay be some other projets written in QT
14:44:53 Quit Thundercloud (Remote closed the connection)
14:45:27krzlinuxstb: may some small task for a day or two?
14:45:52 Join Thundercloud [0] (
14:48:43 Join atm [0] (n=atmb4u@
14:49:23 Join Febs [0] (n=chatzill@
14:51:08 Quit at2 (Read error: 110 (Connection timed out))
14:53:25 Join linuxstb__ [0] (
14:56:25 Join linuxstb___ [0] (
14:56:30 Quit linuxstb (Nick collision from services.)
14:56:34 Quit linuxstb_ (Nick collision from services.)
14:56:44 Quit linuxstb__ (Remote closed the connection)
14:56:52 Nick linuxstb___ is now known as linuxstb (
14:57:13 Join robin0800_ [0] (
14:57:18 Quit robin0800 (Read error: 104 (Connection reset by peer))
14:57:19 Nick robin0800_ is now known as robin0800 (
15:01:20LambdaCalculus37pondlife: Paging Mr. pondlife...
15:01:56 Join Spex [0] (
15:03:10Lloreanlinuxstb: For theme designer authors, we could always just leave "what to do to convince us you're the best man for the job" up to them. GSoC requires a degree of initiative anyway, asking them to take some and come up with their own way of convincing us might not be so bad.
15:04:28linuxstbLlorean: Sounds a good idea to me, assuming at least one shows that initiative...
15:05:01Spexguys is it possible to run java on a musik player?
15:05:07Spexi think yes huh?
15:05:25Spexwhere ever u can install the java runtime it should be possible to use isnt it?
15:05:28pondlifeLambdaCalculus37: pong!
15:05:59krzLlorean: may be some task will show who is the best?
15:06:00linuxstbSpex: Yes, it's possible in theory...
15:06:06krzi'm ready to do it
15:06:31 Join csc` [0] (n=csc@archlinux/user/csc)
15:06:41LambdaCalculus37pondlife: Some more playback oddities rear their ugly heads once again.
15:06:46 Quit Nevtus ("Gone")
15:07:04pondlifeGo on
15:07:05 Join at2 [0] (n=atmb4u@
15:07:09Nico_PLambdaCalculus37: what kind?
15:08:04LambdaCalculus37Running a compiled build against SVN r16927; I applied pixelma's c200 keymap changes (FS #8824) , the MOD plugin (FS #8680), and my own disktidy patch (FS #8595). This is on my c240.
15:08:31LambdaCalculus37Whenever I have a track playing, if I switch to another track without stopping, playback of the new track doesn't start immediately.
15:08:56LambdaCalculus37Either the new track keeps playing, or playback stops. If I pause and resume on the WPS, then it plays the new track.
15:09:13pondlifeWhich format?
15:09:14 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
15:09:36pixelmaMOD codec...
15:09:36Lloreanlinuxstb: If none of them show initiative, that tells us a good deal too. :)
15:09:44LambdaCalculus37pondlife: Everything I've used so far: MP3, Vorbis, FLAC, SPC, and MOD.
15:10:24pondlifeHmm, try plain SVN?
15:10:48LambdaCalculus37pondlife: I almost forgot to mention: this just started happening against this SVN build.
15:10:50 Quit atm (Read error: 110 (Connection timed out))
15:11:07pondlifeHow are switching to a new track? Browsing?
15:11:48LambdaCalculus37pondlife: The switch is where it's giving me strange issues. Like I said, if I have a track playing when I select a new track, the new track doesn't start to play immediately.
15:12:08pondlifeIt goes to the WPS and sits there?
15:12:16 Quit at2 (Read error: 104 (Connection reset by peer))
15:12:19***Saving seen data "./dancer.seen"
15:12:29LambdaCalculus37It either will do that or just keep playing the previous track.
15:12:37LambdaCalculus37When I pause and resume, then the new track plays.
15:13:15pondlifeSeems ok here...
15:13:29 Join wisp_sh [0] (
15:13:31LambdaCalculus37I also saw one other odd quirk: if I set a Repeat to One and play a short track, the track doesn't play completely before repeating.
15:13:47LambdaCalculus37pondlife: Which device?
15:13:55Nico_PLambdaCalculus37: that revision doesn't follow any playback change closely... are you sure it's the right number?
15:14:40pondlifeIt couldn't be keymap confusion, could it?
15:15:25LambdaCalculus37Nico_P: Yes, it's the correct SVN build: r16927.
15:15:42LambdaCalculus37pondlife: Doubt it.
15:15:44pixelmathere's no change in the WPS keymap in my patches
15:15:50pondlifer16945 is current
15:16:15LambdaCalculus37pondlife: Let me back up my build and then I'll use a vanilla SVN build.
15:19:45 Quit XavierGr ()
15:22:07*LambdaCalculus37 updates with a plain SVN build
15:22:15*pondlife does too
15:23:15LambdaCalculus37Still doing it.
15:24:00LambdaCalculus37pondlife: It's still pausing when I select a new track while another is playing.
15:24:27pondlifeI don't have mine here today
15:24:36*pondlife always chooses the wrong one
15:25:33pondlifeOK, so you play an MP3, then go back into the browser without pausing or stopping, then select a second MP3...
15:26:21 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
15:27:06 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
15:27:24LambdaCalculus37pondlife: If I play any music file, then go back to the browser without pausing or stopping, then select a second music file, the second file does not start playing until I either pause and resume, or stop playing and resume.
15:27:36*LambdaCalculus37 thinks it may be a buffering issue.
15:27:48pondlifeEven if both are MP3, specifically?
15:27:55pondlifeThat's all I have to compare with.
15:28:05*pondlife suspects Oggs are a bit broken at the moment
15:28:53LambdaCalculus37pondlife: I don't think that Vorbis files may be broken, because this is happening regardless of what format the first and second tracks are.
15:29:16LambdaCalculus37This happened with MP3, Vorbis, FLAC, SPC, and Musepack files for me.
15:29:20pondlifeOK, and you have explicitly tried 2 MP3s?
15:29:29LambdaCalculus37I've tried 2 MP3.
15:29:32pondlifeJust wanted to be clear
15:29:41pondlifeBinary chop SVN?
15:30:07LambdaCalculus37Both vanilla SVN and my own rolled build.
15:30:22pondlifeYes, but find out which rev first has this problem
15:30:47LambdaCalculus37pondlife: I first noticed it with r16927.
15:31:00LambdaCalculus37Now I'm up to r16944, the most current build.
15:31:02pondlifeOK, so try 16926?
15:31:55LambdaCalculus37Can you roll me a build from r16926? The daily builds don't show it.
15:32:12pondlifeThe previous revision that might affect this would be 16899, I think
15:32:18 Join atm [0] (n=atmb4u@
15:32:52pondlifeJust binchop through the daily builds, until you can be sure which day it first failed. Then we can delve in further, or look closely at the commits for that day
15:33:49atmHi pond life what about porting a software version of rock box to Android
15:34:04 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
15:34:38pondlifeatm: I was just wondering if our simulator would be port-able. But it appears not - check with linuxstb and Bagder maybe?
15:35:04atmwhich simulator?
15:35:11pondlifeThe Roxkbox simulator
15:35:23pondlifeOr however you spell it
15:35:24*LambdaCalculus37 wonders if perhaps other c200 owners see this issue he's having as well
15:35:32atmbut it ports in another way
15:35:40*LambdaCalculus37 stops pointing the Dyslexic ray at pondlife
15:35:41pondlifeLambdaCalculus37: Very likely they will.
15:35:54pondlifeLambdaCalculus37: I've borrowed JdGordon's keyboard
15:35:57atmkeeping the core in C and GUI and others in java
15:36:11pondlifeatm: Rockbox is pure C
15:36:19pondlifeNo Java at all
15:36:19LambdaCalculus37pondlife: I'm also looking through the changelog.
15:36:30atmyes I know
15:36:37LambdaCalculus37pondlife: What about those bits of platform-specific assembler?
15:36:46pondlifeLambdaCalculus37: Which rev?
15:36:51atmbut Android is 100% java
15:37:05pondlifeHence I don't think Rockbox would port to Java
15:37:20atmbut think this way...
15:37:20pondlifeBut I didn't know anything about Android earlier
15:37:32atmits ok
15:37:38atmjust think
15:37:53krazykitatm, if you want to reimplement the simulator in java, go right ahead.
15:38:23atmI don't get it..;)
15:38:30Lloreankrazykit: He's discussing a GSoC application
15:39:01pondlifeIt doesn't sound like a sensible idea to me
15:39:31Lloreanatm: The Rockbox Simulator *is* Rockbox. It is compiled from the same codebase as the on-player version, it's not really a separate program.
15:39:38pondlifeRockbox is a single set of sources for all platforms.
15:39:48pondlifeThat being in C.
15:40:10pondlifeNo C++ or Java allowed
15:40:22atmbut Android is an OS meant for new mobiles by Google and its gonna make a fuzz out there
15:40:48atmthere is a scope of a player in that OS
15:40:48linuxstbatm: It seems possible for an Android application to use JNI to link to native (e.g. C) libraries, so that approach could work.
15:40:48pondlifeFine, but Rockbox is not a suitable project
15:40:58atmyes yes
15:41:13LambdaCalculus37pondlife: There were some buffering changes in r16884:;revision=16884
15:41:22atmyes to linuxstb
15:41:43linuxstbatm: But it wouldn't be ideal... Rockbox is a C app, and life would be much easier if you chose a platform that could run C UI code...
15:41:49Lloreanatm: Rockbox's regular target isn't phones. Any attempt to port it should not interfere with existing workings, this almost requires linuxstb's suggestion.
15:41:50*LambdaCalculus37 still thinks this is buffering-related
15:42:06pondlifeLambdaCalculus37: One way to find out for sure!
15:42:27pondlifeLambdaCalculus37: Try r16884 and r16873
15:42:50pondlifeBe careful to change your build, but not your config
15:42:55atmLlorean:But nowadays mobiles are music players too isn't?
15:43:53atmlinuxstb:Android is completely java and so if we want to use RockBox in Android alone mob's then wat's up????
15:44:09Lloreanatm: But that doesn't mean they're the regular target suddenly. The target is still standalone MP3 players.
15:44:12linuxstbatm: I don't understand your question...
15:44:14LambdaCalculus37pondlife: Yes, sir! :)
15:44:17pondlifeatm: If you want to start a port of Rockbox to Android, go ahead. But you'll need to get compiled C running on there first.
15:44:46linuxstbatm: Rockbox as a pure native application isn't Rockbox any more (IMO). It will be a fork of Rockbox.
15:44:55linuxstb^native java application
15:45:21Lloreanatm: Whatever you do, to be "part of Rockbox" it must be done in such a way that it is still using the Rockbox sources, not a rewrite in another language.
15:46:24LloreanOtherwise any time someone adds a feature, the two versions will be different. Right now Rockbox is _one_ program, that can be compiled for many places. Not many different programs.
15:46:34linuxstbatm: I know nothing about Android - are there actually any phones in existence that can run it/
15:46:43atmI concentrate it as a player for Android..just adding something which is required to run on Android
15:47:11atmIt's about to released commercially;)
15:47:20pondlifeatm: You may be better off looking for an existing Java app. A full rewrite would take some effort, I suspect.
15:47:45atmis it necessary to rewrite Rock BOX??
15:48:00linuxstbatm: If you want to convert it to a Java app, then yes...
15:48:02pondlifeYes, unless you can get C code running under Android
15:48:11LambdaCalculus37pondlife: Now testing r16873.
15:48:12atmI think just API interfacing is ok
15:48:23pondlifeMaybe Google will add C support at some point?
15:48:29linuxstbatm: Have you done any Android programming in the past?
15:48:37atmyes a bit
15:48:46atmIt's almost easy as java
15:48:53linuxstbHave you linked it with C code?
15:49:06LambdaCalculus37pondlife: It's happening in r16873.
15:49:07pondlifeYes, a Java bootloader....
15:49:31atmnot yet but I searched for it and got some links which says JNI can run C
15:49:35pondlifeLambdaCalculus37: The best way is to find an old one that works ok, then binary chop between that and 16873
15:49:40Lloreanlinuxstb: Ideally only /firmware code would have JAVA counterparts, right? (or mostly, at least)
15:49:43 Join kugel [0] (n=kugel@unaffiliated/kugel)
15:50:29atm>:o to pondlife
15:50:32linuxstbatm: If I was you, I would produce a "proof-of-concept" demo, showing that what you want to do (writing the UI in Java, and linking to Rockbox as a C library) will work (assuming that's what you are planning...)
15:50:34*LambdaCalculus37 needs someone to help verify this problem and make sure it isn't just him
15:50:49LambdaCalculus37pixelma: Have you ever had this problem on your c240 before?
15:51:09linuxstbLlorean: I see the firmware/ code being the native OS of the platform (in Android's case, a Linux kernel), and the Java parts would replace the UI widgets in apps/
15:51:28pondlifeLambdaCalculus37: I didn't see it yesterday on my c240... I'm pretty sure I would have done that. I don't normally ever stop playback.
15:52:15LambdaCalculus37Wonder if there's a problem with mine...
15:52:18pondlifePerhaps it's setting related in some way?
15:52:23pondlifeHave you changed any config?
15:52:38pondlifeMaybe make a backup of your current config.cfg, then reset settings?
15:52:59linuxstbatm: Another possible problem is that for security reasons, you may not be able to install C code on a real device, just Java apps...
15:53:15LambdaCalculus37pondlife: I have the config.cfg from my rolled build... want to take a look at it?
15:53:19atmlinuxstb:what's that?
15:53:22Lloreanlinuxstb: See (to me at least) I think as much as possible of the UI code should still be the code we've already got, with just the actual drawing bits being "native" to the platform. That way there's not a separate JAVA UI for people to actually be maintaining when new screens are added, as long as they use the existing pixel/line/etc drawing.
15:53:45pondlifeLambdaCalculus37: Not much use without being able to repro the problem. Better if you could try resetting first.
15:53:55linuxstbLlorean: I see it working at a higher level than pixel/line drawing - otherwise it can't be considered a native app (which IMO should be the aim)
15:54:20 Quit cool_walking_ (Read error: 110 (Connection timed out))
15:54:23LambdaCalculus37pondlife: Okay.
15:54:54 Quit nplus (Read error: 104 (Connection reset by peer))
15:55:36linuxstbatm: What's what?
15:55:46 Join nplus [0] (
15:55:48atmlinuxstb:Nothing I got it.
15:56:58atmlinuxstb:But as far as I know Android will support Rock Box(after some mods) as a player inside.
15:57:39linuxstbatm: How?
15:57:43pixelmaLambdaCalculus37: it takes a while (maybe a second or two) until the new track starts but I don't have to pause or stop. But I'm running a quite old build currently... r16899)
15:58:04atmusing JNI...Can't he?
15:58:27LambdaCalculus37pixelma: Do you want to try my customized build and see if it causes issues?
15:58:45 Join MethoS-- [0] (
15:59:09linuxstbatm: Yes, _assuming_ JNI isn't disabled on real hardware.
15:59:52linuxstbWhich I suspect is likely to be the case - given that google seem to be telling everyone they can't use C/C++ code.
16:00:01LambdaCalculus37pixelma: When I reset my settings, now track playback is normal.
16:00:14pixelmaLambdaCalculus37: mine's quite similar (also includes the MOD codec, keymap changes, not your disktidy patch but sokoban (both are plugins though))
16:01:01pondlifeLambdaCalculus37: Now the fun part - work out which of your config settings breaks Rockbox!
16:01:06LambdaCalculus37pixelma: Now I actually feel quite silly... it WAS taking a second to load! :P
16:01:31*LambdaCalculus37 puts on his dunce cap and sits in the corner like a good boy
16:01:35*pixelma should go ahead with the sokoban changes (but got distracted by bitmapstippifying (or so) WPSs) :\
16:01:53pixelmaor strippifying
16:02:34*pondlife is just having a "repeat"
16:02:55LambdaCalculus37pondlife: It turns out that my Sansa was just buffering the file and was taking a couple of seconds longer than normal to do so.
16:02:55pondlifeRebuffering just happened, I'm sure that's always the case.
16:03:11pondlifeLambdaCalculus37: Wow, very slow?
16:03:21pixelmait feels a bit slow here too
16:03:35LambdaCalculus37pondlife: Yep. Maybe I should check to make sure there's no errors on the filesystem.
16:03:40pixelmabut then I'm used to hwcodec playback :P
16:03:46pondlifeThe priority change you pointed out perhaps?
16:03:58LambdaCalculus37Quite possible.
16:04:19*LambdaCalculus37 wants his rolled build back! :)
16:04:45*pondlife apologises to pixelma for still not trying her keymap patches
16:04:56pondlifeWon't get time until Monday now either
16:05:31linuxstbatm: Does Android have any native audio/video libraries, or a native media player?
16:06:22LinusNamiconn: there?
16:06:31pixelmapondlife: no reason to apologise... still haven't tried your beeep fix ;)
16:07:20LinusNamiconn: Could you check out FS #8840? Could that be caused by your LCD optimizations in r16832?
16:07:38atmlinuxstb:It favours me.have any native audio/video libraries, but no sophisticated media player.
16:08:03LambdaCalculus37pondlife: Which FS entry is your beep fix?
16:08:11*LambdaCalculus37 wouldn't mind giving it a go
16:08:15*pondlife looks
16:09:12pondlifeIt's not a fix yet - well, it stops the beep of death, but pretty much stops the keyclick completely
16:10:48pondlifeOnly on PortalPlayer (of course)...
16:11:30pondlifeSomething in there is very touchy about memory caching, I think. But I am only vaguely aware of such concepts.;)
16:12:10*LambdaCalculus37 has three PortalPlayer targets, sooooo...
16:12:26*pondlife has 1 PP and 2 non-PP, plus sims
16:14:39LambdaCalculus37I didn't sign myself on here for nothing :) :
16:18:18 Quit MethoS- (Read error: 113 (No route to host))
16:19:17 Quit ali_as ("router testing.")
16:19:37 Join Chipsaru [0] (
16:21:27pondlifeI think I'll commit that beep patch anyway, it's simpler code and seems to have no downside on non-PP
16:22:11 Join phinze [0] (
16:22:35 Quit wisp_sh (Read error: 110 (Connection timed out))
16:24:10LambdaCalculus37pondlife: Sounds good.
16:24:52amiconnLinusN: Very possible, although I didn't experience any problems on both H180 and H340 from it (when boosted)
16:25:23amiconnMaybe I should slow down the driver a bit when boosted
16:25:43LinusNworth a try
16:25:49 Quit Mathiasdm2 ("Yuuw!")
16:26:42 Part LinusN
16:27:12Chipsaruhi, i want to create a plugin for rockbox, which will produce sound on a user-specified frequency, from where could i start? (from programming point of view)
16:27:13Chipsarup.s. i know c, c++ programming for windows, and have already installed debian on vmware with nedded compiller version and latest sources of rockbox
16:28:45 Quit phinze ()
16:29:06LambdaCalculus37Chipsaru: Not bad! :)
16:29:22LambdaCalculus37Chipsaru: Here, read these two pages: and
16:29:27 Join xxemoxx [0] (n=5c503684@gateway/web/cgi-irc/
16:29:30pondlifeWow, LinusN's last change gave plenty of green delta on SWCODEC for some reason
16:29:54 Quit xxemoxx (Client Quit)
16:30:03LambdaCalculus37Chipsaru: More pages here for your reading pleasure:
16:30:30*LambdaCalculus37 goes for that sweet dark nectar of the Gods known as coffee
16:30:31linuxstbChipsaru: Maybe look at the metronome plugin.
16:32:46pondlifeLambdaCalculus37: You meant beer, surely. Coffee is strictly off-topic.
16:32:50pixelmapondlife: the green delta is probably from NicoP's commit before which wasn't picked up from the build system (I think he said so)
16:33:31*Bagder confirms
16:34:24 Quit MethoS-- (Connection timed out)
16:34:38*Nico_P smiles at the delta
16:35:34ChipsaruLambdaCalculus37: thanks, is that all i need?
16:35:34Chipsarulinuxstb: does metronome plugin produce that sound or only plays some embedded file?
16:37:27pixelmaNico_P: sorry for misspelling your nick (should have used tab completion) :)
16:37:51Nico_Pheh, no problem at all
16:40:02Nico_Phmm rebuffering seems to be happeniing too often
16:40:45pondlifeNico_P: Perhaps it's too quick, thus causing the repeat track?
16:40:51linuxstbChipsaru: It plays data from an embedded array - but from a programming point of view, that's the same thing (at least in terms of doing things in Rockbox0
16:41:20Nico_Ppondlife: I don't know but it happens when the number of tracks in buffer is shown as 2
16:41:25Nico_PI'll investigate
16:41:26linuxstbChipsaru: i.e. you need to generate some PCM data in a buffer, and then play that PCM data
16:41:35pondlifeNico_P: That would probably be the case
16:41:38 Quit Rob2223 (Read error: 104 (Connection reset by peer))
16:41:56 Join Rob2222 [0] (
16:42:21Nico_PI'll investigate
16:43:04LambdaCalculus37pondlife: Beer is for after work. :)
16:43:18pondlifeBefore, after, during, whatever
16:44:59 Quit atm (Read error: 110 (Connection timed out))
16:45:03 Join phinze [0] (
16:45:28 Join atm [0] (n=atmb4u@
16:46:05 Join toffe82 [0] (
16:47:36Nico_Phow safe is rockbox's high speed on the c200?
16:48:05Chipsarulinuxstb: ok, i understand your idea... now i need to read about pcm data sound representation...
16:48:10 Quit corevette (Remote closed the connection)
16:51:18LambdaCalculus37Nico_P: I haven't tried it yet. I'm still using full speed on my c240.
16:57:56 Join Mathiasdm [0] (
16:59:43linuxstbLambdaCalculus37: full speed is working reliably for you?
17:00:23pondlifeWith 1GB, full-speed is fine here...
17:00:54 Part pondlife
17:01:07 Join markun [50] (n=markun@rockbox/developer/markun)
17:01:31LambdaCalculus37linuxstb: Full speed's working very well; no complaints.
17:03:29 Join simonrvn_ [0] (i=simon@unaffiliated/simonrvn)
17:03:51 Quit simonrvn (Nick collision from services.)
17:03:56 Nick simonrvn_ is now known as simonrvn (i=simon@unaffiliated/simonrvn)
17:10:04 Part atm
17:12:23***Saving seen data "./dancer.seen"
17:13:27 Quit TMM (Client Quit)
17:14:58 Quit XavierGr (Nick collision from services.)
17:15:09 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
17:22:00 Nick parafin|away is now known as parafin (
17:24:22 Join Rondom [0] (
17:27:43 Nick fxb__ is now known as fxb (
17:28:17 Join wisp_sh [0] (
17:28:19 Nick fxb is now known as fxb__ (
17:29:22 Join mcflow [0] (
17:31:07Nico_PLambdaCalculus37: I think the commit I just made might alleviate your issue
17:31:32Nico_Prebuffering would happend one track too early, which could mean much more often with rather large tracks
17:31:54Nico_PSlasheri: her?
17:33:03 Join gregzx [0] (
17:35:02LambdaCalculus37Nico_P: Is that r16945?
17:35:26Nico_Pno, r16948
17:36:05*LambdaCalculus37 just saw the commit appear on the front page
17:37:34 Join OlivierBorowski [0] (
17:39:57 Join at1 [0] (n=atmb4u@
17:45:49Nico_Phmm it seems high speed got my sansa's FS corrupted
17:46:15LambdaCalculus37Nico_P: Yes, gevaerts had warned about that before.
17:48:36Nico_Prunning dosfsck seems to have fixed it
17:48:38 Quit mcflow (Read error: 104 (Connection reset by peer))
17:48:57Nico_Pis rolo broken?
17:49:01Nico_Pon the c200 I mean
17:49:13Nico_Pno it isn't
17:50:26 Join mcflow [0] (
17:50:29Nico_PLambdaCalculus37: do you know how to hard reset the c200? I could use that :)
17:50:46LambdaCalculus37Nico_P: Hold POWER until it shuts off (about 15-20 seconds).
17:51:00Nico_Pah, thanks
17:51:14 Part at1
17:53:59 Quit GodEater (" ajax IRC Client")
17:54:40 Quit preglow ("leaving")
17:56:17Nico_Pmy sansa doesn't boot anymore :(
17:56:21 Quit Febs ("ChatZilla 0.9.81 [Firefox]")
17:56:55LambdaCalculus37Nico_P: Neither into Rockbox nor the OF?
17:56:57 Quit ChristopherW (Read error: 104 (Connection reset by peer))
17:57:03Nico_POF is OK, not rockbox
17:57:19Nico_Pit gets stuck on the white lilmonsta screen
17:57:41 Join ol_schoola [0] (
17:57:49 Join ChristopherW [0] (
17:58:17 Join preglow [0] (i=thomj@rockbox/developer/preglow)
17:58:34LambdaCalculus37Nico_P: I don't know if it'll help, but see what happens when you remove the bootloader and the build, then reinstall. It's a long shot, but hey... it may work.
17:59:02Nico_PI'll try
18:03:14 Join low_light [0] (i=c730190a@gateway/web/ajax/
18:06:15 Quit toffe82 (Remote closed the connection)
18:08:20Nico_Pah, working again :)
18:08:35Nico_Preinstalling the BL did the trick
18:09:37 Quit petur ("work->home")
18:09:54 Join fyrestorm [0] (
18:11:07LambdaCalculus37Nico_P: Do I get a beer? :)
18:11:20Nico_PLambdaCalculus37: you do, thanks ;)
18:15:33markunBagder: what do you think?
18:16:00 Join piga [0] (n=leonardo@
18:16:22markunwe might need to split some of them per model
18:19:44 Join atm [0] (n=molly@
18:19:45 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
18:21:18 Join miepchen^schlaf [0] (
18:22:26 Part atm
18:30:13 Join waldo [0] (
18:31:35 Quit Thundercloud (Remote closed the connection)
18:32:36 Join Thundercloud [0] (
18:32:38 Quit Thundercloud (Remote closed the connection)
18:35:33gevaertsNico_P: try the (first) patch in FS #8663. It makes the c200 flash driver much more reliable for me.
18:35:50Nico_Pgevaerts: even at high speed?
18:36:16gevaertsI haven't seen any corruption since I removed that udelay
18:36:49gevaertsHigh speed gives me lots of usb level errors, resulting in retransmitions, but no silent corruption
18:36:55markunBagder: any comments?
18:39:59 Quit gevaerts ("work->home")
18:44:58 Quit davina^ (Read error: 104 (Connection reset by peer))
18:45:29 Join davina [0] (
18:47:17 Quit Mouser_X (Nick collision from services.)
18:47:26 Join Mouser_X [0] (
18:47:40 Quit kugel ("ChatZilla 0.9.81 [Firefox]")
18:48:34 Join kugel [0] (n=diablo@unaffiliated/kugel)
18:48:39 Part kugel
18:48:48 Join kugel [0] (n=diablo@unaffiliated/kugel)
18:49:53 Join petur [50] (n=petur@rockbox/developer/petur)
18:52:10 Quit Chipsaru (Read error: 110 (Connection timed out))
18:52:34 Quit tvelocity_ ("Αποχώρησε")
18:56:07 Quit miepchen^schlaf ()
18:56:08 Quit fyrestorm (Read error: 104 (Connection reset by peer))
19:00:16 Join bughunter2 [0] (
19:02:16 Join Christopher1 [0] (
19:02:18 Join lee-qid [0] (
19:02:39kugelWhy is elephantsdream download so slow?
19:05:50 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
19:08:20 Join gevaerts [0] (
19:09:15linuxstbkugel: I'm getting 815KB/s...
19:10:59gevaertsSo that's why it is slow for kugel :) You are hogging all the bandwitdh
19:11:07pixelmaisn't it on different mirrors?
19:11:17*gevaerts swaps some letters
19:11:41linuxstbpixelma: Yes. Maybe if kugel stopped and restarted, he'll download faster...
19:11:55linuxstbAlthough I've never noticed any of the mirrors being that slow
19:12:05 Join DerPapst [0] (
19:12:26***Saving seen data "./dancer.seen"
19:15:04kugelmy internet is fine
19:15:26kugelbut download is still slow
19:15:42 Join ryran [0] (
19:17:05linuxstbkugel: Have you tried stopping the download and restarting?
19:17:20kugelYea, several times now
19:18:37 Join Hyper-magiK [0] (n=meh@
19:19:09 Quit ChristopherW (Read error: 110 (Connection timed out))
19:20:57 Join gevaerts_ [0] (
19:21:15 Quit gevaerts (Nick collision from services.)
19:21:20 Nick gevaerts_ is now known as gevaerts (
19:22:53 Quit mcflow (Read error: 104 (Connection reset by peer))
19:23:15preglowgevaerts: how's the status on usb these days?
19:24:06gevaertspreglow: The actual status is the same as usual. However, we have a theory on why it doesn't work properly
19:24:46 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:25:04gevaertsWhat basically happens is that if the CPU accesses RAM during USB transfers, the USB controller gets delays in its RAM access
19:25:53gevaertsIf those delays happen at the wrong moment, there is a short interruption in outgoing packets, which can make the packet unreadable for the other side
19:26:27 Join mcflow [0] (
19:26:50preglowis this the signal quality problem?
19:26:50gevaertsThere is probably a register somewhere on PP that controls RAM access priority. Finding this would solve the problem
19:27:07gevaertsYes. As far as I can see this is the only problem we have
19:27:09 Quit Horscht (Client Quit)
19:27:32gevaertsAccording to amiconn, this register would most probably be located somewhere in 0x700000xx, or maybe in 0x600060xx
19:28:10*gevaerts goes to have dinner now
19:28:44 Join miepchen^schlaf [0] (
19:29:16 Quit phinze ()
19:29:47 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:29:56 Quit robin0800 (" Try HydraIRC -> <-")
19:30:07 Join robin0800 [0] (
19:30:15preglowgevaerts: certainly does sound plausible
19:30:36preglowso i guess having a go at retailos or the bootloader with ida pro is next...
19:31:10ryranhey guys. I'm pretty new to doing anything with the code other than reconciling patch errors but I was wondering if someone could point me in the right direction for how to disable the recording menu for an ipod.
19:31:32 Quit bughunter2 (Remote closed the connection)
19:32:59DerPapstryran: firmware/export open config-ipodXXXX.h and disable #define HAVE_RECORDING
19:33:02ryranI assume it'd be pretty simple... something to do with changing ifdefs for have_recording or something.. but I've only just now started looking. anyway, I ...
19:34:03DerPapstlots of (unwanted) features can be disabled there
19:34:27ryranawesome possum. that's exactly what I figured must exist.
19:35:53LambdaCalculus37ryran: Did you just say "awesome possum" out loud? :)
19:36:03ryranhehe. no way. just typed it. ;P
19:36:07 Quit krz (Read error: 110 (Connection timed out))
19:37:28ryranDerPapst, so ... this is gonna sound like a stupid question, but in the rb codebase.. '#' doesn't make a line be a comment? I have to use '/* */'
19:38:09ryranahh. that explains a lot.
19:38:24DerPapst /* */ are comments and # is for pre-processor stuff
19:38:34 Quit Horscht (Read error: 104 (Connection reset by peer))
19:39:27ryranI allllways wondered about that.
19:41:36 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:41:47 Join DerDome [0] (
19:42:23Nico_Pgevaerts: I keep getting "ROLO error: error reading file" after transferring a new binary to my c200
19:42:50Nico_Pand then it fails to boot... that's with the udelay patch applied
19:43:46Nico_Pand with USB_HIGH_SPEED defined
19:43:53LambdaCalculus37Nico_P: I don't think ROLO works on the PP devices yet, IIRC.
19:44:14Nico_Pit does on at least some AFAIK
19:44:20LambdaCalculus37I did manage to ROLO my friend's iriver H120 before, and my girlfriend's H320.
19:44:23Nico_Pand the failing to boot indicates a problem
19:46:49Nico_Phow do I exit recording?
19:47:05LambdaCalculus37Short press on POWER, I believe.
19:47:28 Join mossimo2k3 [0] (n=18eda03b@gateway/web/cgi-irc/
19:47:31LambdaCalculus37I'm using pixelma's c200 keymap changes from FS #8824, so I don't 100% remember the original keymappings.
19:47:57Nico_Pthat gives me the menu... I tried a long press of POWER but had a disk full warning so I'm not sure what made it quit
19:48:01pixelmathe original is long power IIRC
19:48:32Nico_Pthat's a bit counterintuitive
19:48:38*LambdaCalculus37 decides to refer to our fine manual :)
19:49:08mossimo2k3I have an issue with my database can anyone help me?
19:49:30LambdaCalculus37Nico_P: pixelma's right; long press on POWER is the original keymap.
19:49:49DerPapstmossimo2k3: no, because our crystal balls are broken
19:49:56pixelmaNico_P: which is why we're trying to improve it...
19:50:20Nico_Ppixelma: I'll check out your patch :)
19:50:53 Join superman [0] (n=a306870a@gateway/web/cgi-irc/
19:51:14 Join toffe82 [0] (
19:51:50 Quit Nevtus ("Gone")
19:52:00 Join tvelocity [0] (
19:52:03supermandoes rockbox play videos or do you need to switch to the original firmware to watch them?
19:52:23pixelmaNico_P: and if you have better ideas, please tell. I meant to put the patches there because I wasn't a 100% satisfied with it :)
19:52:42 Quit linuxstb (Remote closed the connection)
19:52:46scorche|shsuperman: see the PluginMpegplayer wiki page
19:53:02 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:53:03Nico_Ppixelma: I'll try to think about it. I can see how it's really not easy with so few buttons
19:53:05pixelmasuperman: also, what player do you have?
19:53:17supermanipod 5G
19:53:26 Join FOAD_ [0] (
19:53:40*DerPapst notes that pixelma continues deleting almost entire svn...
19:54:04pixelmaNico_P: the fun thing is that the buttons aren't few, just the layout is a bit weird so reduces the choice for "intuitive" a lot in some places
19:54:06DerPapstsuperman: i'd recommend to use the OF for video in 5G iPods-.
19:54:09*Nico_P hopes his commit will be picked up as he hopes for another nice green delta
19:54:14supermanalso does rockbox delete the songs you have when you install it?
19:54:21supermanthanks DerPapst
19:54:29DerPapstsuperman: no.
19:54:46supermancool, i didn't know cuz i had a clean pod when i stalled RB
19:55:13supermanthanks guys for your help
19:55:21 Quit superman (Client Quit)
19:55:29mossimo2k3does RB have a music loader (ipod 80gb)
19:55:36DerPapstsuperman: but you will have to use rockbox' database because the OF stores the songs... well different (too lazy to explain :-P)
19:55:36*pixelma notices that the frontpage makes a link to r2 in the "engineer2" part of the commit message...
19:55:38scorche|sha music loader?
19:55:49 Join fyrestorm [0] (
19:55:58pixelmaI meant the "frontpage script" of course
19:56:04DerPapstalways too slow -.-
19:56:08scorche|shwhy do you need a "music loader"?
19:56:38DerPapstmossimo2k3: just copy&paste your music wherever you want
19:56:51mossimo2k3what else should i used to upload music to the ipoop
19:56:52linuxstbmossimo2k3: You can just copy the files to your ipod as if it was a normal disk (which it is...), and Rockbox will let you play them
19:57:04mossimo2k3oh ok
19:57:15*gevaerts decides to grab his c250 to test RoLo
19:57:15mossimo2k3anywhere on the drive?
19:57:34Nico_Pmeh, it seems the build system hates me
19:57:56scorche|shmossimo2k3: yes...anywhere...most people make a folder named "Music" and put it all in there
19:58:01pixelmaDerPapst: there are also some new bitmaps :P
19:58:01Nico_P...or I'm just not patient enough :)
19:58:26mossimo2k3will the oe system see the music 2?
19:58:40scorche|shthe apple firmware?
19:58:47scorche|shno it will not
19:58:51 Join superman [0] (n=a306870a@gateway/web/cgi-irc/
19:59:18DerPapstif (committer == "Nico_P") do_nothing(); else ...
19:59:51*amiconn hrmpfs @PP5002 timers
19:59:57supermanactually i have another question, how do you view movies you put on your ipod?
20:00:18 Join stripwax [0] (
20:00:21amiconnDerPapst: Btw, do you have an idea where went? The PP5020 page is there...
20:00:57scorche|shsuperman: are these movies that were made for the apple firmware, or the rockbox firmware?
20:01:11DerPapstamiconn: i can still see it. we have some server problems but all pages can be accessed if you're logged in. soory.
20:01:15mossimo2k3my music database loaded once and after a reset the boot commited 9 of 9 and was working fine then after another reset i pulled the battery and then restarted it now RB cant see anything cept the last album i played any suggestions?? IPOD 80GB
20:01:15supermanhow do i tell?
20:01:25amiconnDerPapst: logged in??
20:01:39DerPapstamiconn: yes. want a "dump" of the page?
20:01:47scorche|shsuperman: well, if you dont know, then i would assume it wasnt encoded for rockbox
20:01:52amiconnIs that new?
20:01:58scorche|shdid you look at the page i told you to look at?
20:02:01amiconn(the login stuff I mean)
20:02:11DerPapstamiconn: we have that for a while now. seems to affect random pages.
20:02:19supermanscorche|sh: how do i encode them for rockbox?
20:02:31scorche|sh< scorche|sh> did you look at the page i told you to look at?
20:03:05LambdaCalculus37superman: Read the page scorche|sh told you to look at.
20:03:11*LambdaCalculus37 points up to the link
20:03:12supermanscorche|sh: the mpegplayerplugin one? ya i read that one
20:03:16advcomp2019superman, the PluginMpegplayer
20:03:22scorche|shsuperman: well, that one tells you
20:03:39supermanscorche|sh: i just read that mpegplayer was included i din't read in depth
20:03:48supermanscorche|sh: sry
20:03:49amiconnAll I get is a completely blank page. Not even an error message...
20:03:50scorche|shwell, do :)
20:03:50LambdaCalculus37Read it in depth, then.
20:03:50DerPapstsuperman: you should
20:04:09 Join Chipsaru [0] (n=chips_ru@
20:04:10supermanscorche|sh: k thanks
20:04:14 Quit superman ("CGI:IRC")
20:04:28DerPapstamiconn: yes. i know. maybe an error in the html code. i'll check.
20:04:51*Nico_P managed to avoid red :)
20:06:57gevaertsNico_P: I just tried copying a new rockbox.mi4 to my c250, and it correctly RoLo-ed
20:07:05 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
20:07:48Nico_Pgevaerts: then maybe I shouldn't be using high speed. do you use it?
20:08:26gevaertsI did first reformat the filesystem. Maybe that makes a difference ?
20:08:53*gevaerts had too much cruft on his c250 and decided that that was the easiest way to get rid of it
20:09:28gevaertsI mean, I formatted, installed a fresh build, rebooted, copied rockbox.mi4, and then RoLo-ed
20:09:42mossimo2k3can RB display album icons like ipoop does
20:09:53DerPapstamiconn: for now.
20:10:15 Quit FOAD (Read error: 110 (Connection timed out))
20:10:15 Nick FOAD_ is now known as FOAD (
20:11:27scorche|shmossimo2k3: yes, and stop saying ipoop...
20:12:00 Join mcuelenaere [0] (
20:13:11amiconnDerPapst: thanks
20:14:13DerPapstamiconn: no problem :-)
20:17:00 Part Spex
20:17:05 Join wumbus [0] (n=zachdunn@
20:17:46 Part wumbus
20:20:01 Quit PaulJam (".")
20:22:21 Nick fxb__ is now known as fxb (
20:22:57 Quit fyrestorm (Read error: 113 (No route to host))
20:24:21 Quit ryran ("Leaving")
20:26:21 Join fyrestorm [0] (
20:32:37kugelroolku: Go ahead, the game is nice, and I didn't experience bugs yet
20:32:50 Nick Christopher1 is now known as ChristopherW (
20:34:19kugelroolku: But it would be nice if you would add scroll wheel support, as in the scroll wheel does the same as up and down
20:34:29roolkukugel: I have had it in my build for ages and only just noticed it by the missing keymaps for m:robe
20:34:38ChristopherWSomebody earlier was saying that was very slow. If you do a "host" you get three IP addresses. I can connect to two of them ( and, but I can't connect to
20:34:57roolkukugel: I don't have a target with a scrollwheel, so I wouldn't be able to test it
20:35:14kugelBut I do ;)
20:35:36LambdaCalculus37kugel: Need someone with an iPod?
20:36:31kugelah wait, I just see that scroll wheel is allready in use
20:37:44kugelroolku: Weird, for the e200 the scrollwheel is in use, but not for ipods
20:37:47*linuxstb wonders what kugel and roolku are talking about
20:38:19kugellinuxstb: pegbox game, fs5725
20:38:34kugelroolku: There are also keymaps missing for older ipods
20:38:59roolkulet me get a clean tree...
20:39:23kugelthe ipod keymap could need a rewrite before committing
20:39:52roolkuI am open for suggestions
20:41:21*Llorean thinks it could simply be named "Pegs"
20:41:28*Llorean is slightly tired of FooBox names.
20:41:45*gevaerts suggests rockpegs instead :P
20:41:50*pixelma guesses the Ondio keymap will have some problems (you'll probably "select" when you only want the othe options that are on menu+x button combos)
20:42:24*LambdaCalculus37 thinks Pegboard is a good name
20:42:30*Llorean looks for things to throw at gevaerts
20:42:56*gevaerts hands Llorean some rocks and some pegs
20:42:58*pixelma guesses this applies to the Ipod ones too
20:43:22*LambdaCalculus37 hands Llorean a lightsaber
20:43:39*gevaerts runs away
20:43:50*pixelma further notices that the button combos defined for the X5M5 pad won't work at all
20:44:11kugelroolku: this looks a bit better to me
20:44:37*LambdaCalculus37 further further notices that button combos need to be defined for the c200 and m:robe
20:45:43LambdaCalculus37And there's still #define's for the iFP-7xx.
20:45:50roolkuthe bitmaps for mrobe are also a little small
20:46:06pixelmaLambdaCalculus37: why not?
20:46:40LambdaCalculus37pixelma: Meh... good point. :) Maybe one day the iFP-7xx will join the family.
20:47:45 Join Spex [0] (
20:48:44pixelmaroolku: and if the bitmap size is the same as the item size (or something) or a multiple if in a strip, you wouldn't need to #define for each screen, you can get them from the used bitmaps (so only have to check in bitmap SOURCES)
20:49:48kugelroolku: Don't forget to add BUTTON_SCROLL_FWD|BUTTON_REPEAT (and _BACK) for ipods and e200 (including PEGBOX_LVL_UP_REPEAT action)
20:50:19roolkupixelma: yes, I'll have to check that out - at the moment I am trying to get it to work for m:robe
20:50:54 Join disorganizer [0] (
20:51:13kugeldisorganizer: hey
20:51:29disorganizerhi kugel, hi all the others
20:51:31kugeldisorganizer: I probably know what makes mpegplaye crashing with multifont
20:51:53disorganizerso what is it?
20:52:17 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
20:54:07kugeldisorganizer: multifont alters several functions, like font_get_width. They require an argument more, but the plugins and the api still uses the SVN function
20:55:14disorganizerso the fix would be rather huge? i wonder if fixing that makes sense. we urgently need to find a glyphcache algorithm and work on a commitable version :-)
20:55:27kugeldisorganizer: I think taking the plugin parts of an older version (with replacing FONT_PLUGIN with FONT_UI before) will work
20:55:46kugelI'd rather have a working version first
20:56:40disorganizerwell, the problem is that the patch is completely f*ed up. it was changed to not use font_ui, then to completely get rid of font_ui, then back to only use font_ui. so imho there are just too many changes and nobody really took care of the patch
20:57:00mossimo2k3how long has RB been around for the ipods?
20:57:01kugelDid you guys allready see ?
20:57:34 Join TMM [0] (
20:57:41LambdaCalculus37mossimo2k3: Since 2006, I believe.
20:58:02amiconnDepends on the ipod generation
20:58:04disorganizerkugel: if we have a caching algoriythm, implementing the font directives would be easy even afterwards because wps-vp is already prepared for multifont
20:58:26mossimo2k3i have 80gb
20:58:36LambdaCalculus37kugel: I think Bagder was the one who first saw it:
20:58:53 Quit mf0102 ("Verlassend")
20:59:28kugeldisorganizer: And how do you want to test the caching without a working version?
20:59:40DerPapstamiconn: if you refresh your browsers cache can you access now?
21:00:03kugeldisorganizer: The fix isn't huge as I allready stated, so it should be no problem to fix it asap and focus on the caching
21:00:10amiconnDerPapst: Yes, works now :)
21:00:57DerPapstamiconn: good :-)
21:01:55disorganizerkugel: an algorythm can also be without any code :-) if we have a good caching algorythm then implementation would be easy for any dev (we just need to put together what we have then). but at the moment we dont even have the slightest idea of how to implement the caching.
21:02:06 Quit midkay ("Leaving")
21:03:46amiconndisorganizer: An algorithm does have nothing to do with rhythm (i.e. music) ;)
21:03:46 Quit toffe82 (Read error: 104 (Connection reset by peer))
21:03:58 Join toffe82 [0] (
21:04:27*disorganizer smacks amiconn with some y's
21:04:41disorganizerwho cares?
21:04:53ChristopherWalgorithm has nothing to do with Al Gore either
21:05:06ChristopherWor his (in)ability to dance
21:05:07*gevaerts thinks that if amiconn believes that he is in the wrong project :)
21:05:55*disorganizer wonders what this sentence meant to say
21:07:06bluebrotherhow does bootloader installation work on m3? The same way as on m5 / x5?
21:07:51domonokybluebrother: i have done it according to the instructions given to me, i hope it works..
21:08:05domonokyits like m5/x5 with a different filename
21:08:18bluebrotherdomonoky: I'm currently adding a table to list the current mountpoint resolvation status
21:08:22 Join bughunter2 [0] (
21:10:27 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
21:12:29domonokybluebrother: sounds nice..
21:12:30***Saving seen data "./dancer.seen"
21:13:08 Part Spex
21:14:59 Quit fyrestorm (Read error: 104 (Connection reset by peer))
21:15:10 Join joris [0] (
21:15:13bluebrotherinterestingly it looks much better for non-windows os
21:15:27 Quit spiorf (Remote closed the connection)
21:18:42*domonoky just sees that Autodetection:detectUsb still uses the configuration directly, should perhaps also use the rbsettings object ?
21:19:01 Join juudu [0] (
21:19:21 Nick juudu is now known as appleipodder (
21:19:46bluebrothersure, why not?
21:21:15*bluebrother would like to clean up quite a bunch of rbutil source files
21:21:20 Join krz [0] (
21:21:38domonokybluebrother: feel free to do it, when you find time for it.. :-)
21:22:11 Join bughunter3 [0] (
21:22:11 Quit bughunter2 (Read error: 104 (Connection reset by peer))
21:23:39bluebrothersure. First I want to address the bootloaders. That file got quite chaotic ;-)
21:23:51bluebrotherdo we have something like a red X icon in the wiki?
21:23:57 Join argumentD [0] (
21:24:13 Join fyrestorm [0] (
21:24:13bluebrotheror a question mark icon?
21:24:13amiconnbluebrother: yes
21:24:17domonokysure, this file has grown a lot over time.. :-)
21:24:45amiconnSee e.g.
21:25:25*amiconn also points bluebrother towards the "More formatting help" link at the bootom of each wiki edit page :)
21:25:52*bluebrother found the link (it's somewhat hidden below that page) :)
21:26:23amiconnAnyway, that window in turn links to
21:28:07 Join corevette [0] (
21:28:20 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
21:28:35 Join midkay [0] (n=midkay@rockbox/developer/midkay)
21:28:59mossimo2k3are there programmers/builders in here?
21:29:19bluebrother"hidden" in the sense that I usually fail to find the link to the DocGraphics page the first try ;)
21:29:58*gevaerts looks in corners to see if he can spot a programmer somewhere
21:30:01Bagdermossimo2k3: yes
21:30:10*Horscht runs away from gevaerts
21:30:29Bagdershotofadds: hey you have some yellows to squash! ;-)
21:30:34mossimo2k3can i make suggestions?
21:30:44 Join Spex [0] (
21:30:53bluebrotherdomonoky: please add / correct what's missing:
21:30:54shotofaddsone, or two. I've been busy searching out the ultimate question..#
21:31:56shotofaddsBagder: I spotted some D2 talk in the logs the last few days, so it's having the desired effect (though no-one's fixed the warnings for me...) :-)
21:31:59 Quit fyrestorm (Read error: 113 (No route to host))
21:32:30amiconnbluebrother: Would it be sufficient for this mountpoint detection to work if the OF is installed?
21:33:00linuxstbmossimo2k3: Just go ahead...
21:33:01shotofaddsanyway, what I would like to ask is:
21:33:02*shotofadds would like to create a generic firmware/drivers/pcf50606.c that uses standard i2c_write() calls (for the D2 port) and update pcf50606.h with register definitions. any objections?
21:33:19mossimo2k3in the now playing screen i cant find how to change the graphs to numerical for battery or volume, just thinking out loud here but it'd be cool to have a numerical value inside the graphical ie* best of both worlds
21:33:31bluebrotheramiconn: depends on the player −− for example on Iaudios I don't know of any file we could rely on
21:33:36domonokyamiconn: yes, as when rockbox is installed we can find the mountpoint with the rockbox-info.txt
21:34:01amiconnThe cowon OFs write a whole bunch of (annoying) folders and files when they start
21:34:08LambdaCalculus37mossimo2k3: Different WPS screens can display either graphical or numerical. It depends on the WPS author.
21:34:15amiconnSo these folders & files should be present when the OF is installed on an iaudio
21:34:28bluebrotherBagder: what do you think about making h1 and h2 tags in the wiki having background colors similar to the headings in the left-side menu?
21:34:50bluebrotheramiconn: feel free to fill such possibilities in −− we already rely on OF files on Archos
21:34:57domonokyamiconn: sounds good..
21:35:02bluebrothers/feel free/please do/
21:35:09mossimo2k3so cnage the wps screen and then the ability to change those will appear in settings?
21:35:11amiconnOkay, need to flash back an OF and dig up the folder names
21:35:30amiconnOne of the files is in the root, and iirc called NED_LIC.SYS
21:36:06mossimo2k3sweet i just changed it thanks!
21:36:25*amiconn solved a part of the PP5002 timer mystery
21:36:54*bluebrother wants to make the bootloader installation class smart enought to distinguish between OF and RB bootloader
21:37:05bluebrotheramiconn: nice, congrats!
21:37:19amiconnOur (guessed) TIMER2_MASK seems to be incorrect. At least that interrupt bit gets set *both* when timer2 fires *and* when a GPIO interrupt fires...
21:37:25LambdaCalculus37amiconn: Nice work! :)
21:37:44bluebrotherthat way we could reuse it for more robust detection of the players mountpoint
21:37:45amiconnVery odd, actually
21:39:04*amiconn found that by putting a backlight toggler (toggling every 20th interrupt) into the isr
21:39:16Nico_Pamiconn: what were the side effects?
21:39:51 Quit toffe82 (No route to host)
21:40:00gevaertsNico_P: epilepsy of course
21:40:11amiconnThe side effect was that the greylib froze/unfroze with the ISR on COP when using the wheel or buttons during display
21:40:30amiconnButtons on 1st..3rd Gen use GPIO interrupts
21:41:57*Nico_P has managed to get "his" ipod to do weird stuff on track skip by forcing a rebuffer on every skip
21:41:58 Join toffe82 [0] (
21:42:14Nico_Pmaybe I'll finally be able to fix it now I can repro
21:42:58domonokybluebrother: what UNC path is this, we know of the sansa ? do you have a example how such a UNC path looks ?
21:44:06 Join ali_as [0] (
21:44:12Nico_Pwow I even got it on the sim! \o/
21:45:11bluebrotherdomonoky: something like \\.\PhysicalDrive1
21:45:17 Join fyrestorm [0] (
21:46:01domonokyhm, doesnt look really helpfull..
21:46:24bluebrotherit's the same as for Ipods (unless my memory serves me wrong −− it's definitely this way on Ipods)
21:46:51amiconnThat's in no way a fixed path
21:47:33amiconnThat's just the second drive that windows knows about (first would be \\.\PhysicalDrive0)
21:48:00*domonoky thinks sansapatcher gives this, so it should be valid after running sansapatcher.. :-)
21:48:00bluebrotheryep. But ipodpatcher figures the correct one.
21:48:12 Quit petur ("switching...")
21:48:21 Join petur [50] (n=petur@rockbox/developer/petur)
21:48:34domonokythe problem is mapping this to a drive name ie "D:"
21:48:38bluebrotherunfortunately I haven't found of a way to figure the drive letters that got assigned to the partitions on that drive
21:53:00 Quit perrikwp (" ajax IRC Client")
21:54:50amiconnbluebrother: You can do that via WMI. See here:
21:55:02 Part Hyper-magiK
21:55:12 Quit LambdaCalculus37 ("CGI:IRC 0.5.9 (2006/06/06)")
21:55:50 Join _abc_ [0] (n=Peter@
21:58:57bluebrotherinteresting. I searched msdn quite a bit but haven't found this myself. Will have a look.
21:59:28*Nico_P is closing in on FS #8513
22:01:26 Part _abc_ ("Leaving")
22:01:42 Join phinze [0] (
22:04:19*gevaerts wonders if phinze is a USB guru _today_ ;-)
22:04:35phinzegevaerts: guru no, but getting closer :)
22:04:46gevaertsGood :)
22:05:08phinzedown on the basics... tonight i'll be getting to USB audio and fleshing out my application
22:05:32phinzegotta run to class now though
22:05:35 Quit phinze (Client Quit)
22:08:52 Join midgey [0] (
22:13:15Nico_Pvictory \o/
22:13:37Nico_Pnow to write a proper fix that (hopefully) won't break anything else
22:14:39kugelCan I make svn diff ignore changes to svn:keywords (those happen to change on my sources for some reason I don't know)
22:15:04kugelNico_P: Nice!
22:15:42Nico_Pit was much easier to fix with a reliable repro recipe :)
22:16:39*kugel eagerly awaits the commit
22:16:45*gevaerts thinks that 30 minutes to find this is pretty good
22:19:05*amiconn fooled himself :/
22:20:51kugeldisorganizer: There?
22:21:13disorganizerkugel ?
22:22:00kugelI fixed multifont (hopefully)
22:22:45disorganizerhip hip hurray! :-)
22:23:16kugelmeh, wrong file name
22:23:30kugelcorrect file though
22:23:32disorganizerNico_P: what did you fix?
22:23:51Nico_Pdisorganizer: FS #8513
22:23:53kugelCould anyone delete my last post in FS #4733? My filename is a bit missleading
22:25:10disorganizerkudos @ Nico_P
22:25:39bluebrotherkugel: the one with the attachment?
22:25:40Nico_Pwell I'm the one who introduced it :)
22:26:05*disorganizer slaps nico_p with some hard and old kudos
22:26:05bluebrotherdon't we always introduce bugs to fix them later? ;-)
22:26:06peturhaha 20080304 ?
22:26:50Nico_Pbluebrother: hehe, yeah we do :)
22:27:00Nico_Pplayback.c is particularly prone to it too IMHO
22:27:48Nico_Pwow, now a segfault
22:27:52peturkugel: last post deleted
22:28:22 Quit keanu (Remote closed the connection)
22:29:00kugelpetur: thanks
22:29:48*bluebrother feels ignored
22:31:12 Join mikus [0] (n=5316bdf7@gateway/web/cgi-irc/
22:32:13*disorganizer waits for kugels next post with the right filename
22:32:24 Quit low_light (" ajax IRC Client")
22:33:34 Join qwedsa [0] (
22:36:14 Part Spex
22:36:48 Quit piga (Remote closed the connection)
22:36:56kugeldisorganizer: Yea, sorry, gotta upload to versions this post
22:38:02disorganizeranother question: what would be needed to commit the list-vp patch?
22:41:21 Quit mikus ("CGI:IRC")
22:42:06kugeldisorganizer: My patch?
22:42:50kugelit has a little design flaw, it repareses the setting at every redraw of the list (multiple times even). This should be fixed
22:43:08kugelSo that it's only parsed once, i.e. at settings load
22:44:10kugelThough it's running decently without noticable performance backdraws (at least in my opinion and on my e200)
22:44:55 Join perrikwp [0] (i=98213942@gateway/web/ajax/
22:46:40disorganizerok. so its not finished yet :-)
22:49:21kugelI never said that
22:50:54disorganizerif it uses unnecessary cpu then it wont pass our QA tests regarding memory-cpu and binsize savings :-)
22:51:19 Join Rincewind [0] (
22:51:21*disorganizer gets the impression we should rename rockbox to greenbox
22:51:37*Horscht thinks we should honor Horscht
22:52:46*Horscht proposes Horschtbox
22:54:18 Quit ColinT (Remote closed the connection)
22:54:18 Quit qwedsa_ (Read error: 110 (Connection timed out))
22:54:31 Join ColinT [0] (
22:54:31 Quit corevette (Remote closed the connection)
22:54:48kugeldisorganizer: It doesn't waste cpu. The sansa won't get boosted while parsing
22:55:03kugelthe parsing needs about nothing CPU time
22:55:47kugelBut (very) little more than reading variables directly
22:56:39 Part mcflow
22:58:27peturamiconn: what triggers an I03:AddrErr on h300? I'm resuming bootloader debugging on my h380 and a sleep before usb_init seems to cause this. Address = 0x31800000 which is .ibss thread.o - the previous crashes I observed were also near thread code so I suspect some issue there :/
22:59:31amiconnAre interrupts enabled in iriver bootloaders?
22:59:45amiconnIf not, sleep() won't work
22:59:53 Quit OlivierBorowski (Remote closed the connection)
23:00:31petursleep is used in other parts near my code
23:00:47peturlike the charge screen
23:01:14peturyup, enable_irq() is called in the beginning
23:02:30markunI don't really get this guys point about donating to rockbox:
23:02:51 Nick parafin is now known as parafin|away (
23:03:02n17ikhdoes rockbox have a cash fund to buy players, test equipment, etc with?
23:03:22kugeldisorganizer: does the latest multifont version work?
23:03:32markunn17ikh: the paypal button on the frontpage
23:03:57n17ikhwell, I know you can donate to them
23:04:07n17ikhjust wondering how the money is distributed
23:04:16n17ikhis rockbox a nonprofit?
23:04:32disorganizerkugel: seems to work, just i dont find mpegplayer to test *g*
23:04:49kugelyou need to start a video from the filebrowser
23:04:49amiconnpetur: MCF5249UM.pdf section 3.5.2
23:04:59kugelit's a viewer plugin
23:05:05markunn17ikh: rockbox is currently not an organization
23:05:06scorche|shn17ikh: rockbox isnt anything, really
23:05:38n17ikhI see.
23:06:18*petur scratches head
23:06:18domonokyhow do we handle our new "big money for Ads" in repeskt to tax and else ?
23:06:38domonokyare they also "donated" ?
23:06:54 Join n1s [0] (
23:06:54 Join kushal_12_27_200 [0] (n=kushal@
23:06:56kugeldisorganizer: I just noticed, text viewer wasn't working either (my version makes it working again)
23:07:20disorganizerkugel: how do you set a font with your list-patch?
23:07:28Bagderdomonoky: I'm leaving that worry to zagor ;-)
23:07:54*petur reads donate page and remembers to bring beer bill next devcon ;P
23:08:30scorche|shwhenever that will be...
23:08:52scorche|shargh...another night passed with me forgetting to write up a mail..
23:09:04Bagderscorche: ssssh, don't tell petur he's bringing a bill! ;-)
23:09:09*scorche|sh put it in his palm so he wont forget
23:09:56kugeldisorganizer: How are you doing that in a wps viewport? :)
23:10:16Nico_Paudio_check_new_track() is a noghtmare :(
23:10:23Nico_Pnightmare even
23:10:27disorganizerok. then it doesnt work. "list viewport: 16,14,145,145,2" should use userfont 2...
23:10:34*gevaerts wonders if anyone else than petur and Bagder already has a date for DevCon
23:11:12scorche|shgevaerts: i dont get these subtle hints...if you are going to ask me, just come out and say it!
23:11:17kugeldisorganizer: you are missing some parameters
23:11:31disorganizerkugel: are all of them needed?
23:11:47gevaertsscorche|sh: ok. Do you know when DevCon will be ?
23:11:50*Bagder does the rare subtle announcement dance
23:11:54kugeldisorganizer: Yes, or does the wps allow to leave them blank?
23:11:58*gevaerts sometimes misunderstands things :-P
23:12:08*scorche|sh runs away crying
23:12:31***Saving seen data "./dancer.seen"
23:12:52kugeldisorganizer: See, my list viewport setting and the wps %V tags use the same function for getting parsed
23:13:02*gevaerts gives scorche|sh a beer
23:13:08disorganizerkugel: i just got confused again by the 2=userfont1 thing *sigh*
23:13:29kugeloh, but you still need to define colors
23:14:17 Quit kushal_12_27_200 (Client Quit)
23:14:18*scorche|sh drowns his sorrows
23:14:36gevaertsThat's better :)
23:14:46 Quit dabujo ("( :: NoNameScript 4.2 :: )")
23:15:40disorganizerkugel: i noticed. took the config from an old version of my theme *sigh*
23:15:50*amiconn hates dependency problems
23:16:34amiconnThe subdir-plugins aren't relinked when the pluginlib changes
23:17:16pixelmamarkun: maybe send one of "our swedes" over to the guy... ;)
23:18:39disorganizerkugel: works on sim. couldnt test on real device though
23:19:00kugelbertrik is a committer now?
23:19:02linuxstbkugel, disorganizer: Why does the multifont patch have a UI font, then 7 "user fonts", rather than simply 8 fonts?
23:19:28disorganizeri dont know. as i said its completely crippled
23:19:43kugellinuxstb: Don't ask me, I just keep it sync'd and working from time to time. I'm not too into that patch
23:20:00kugellinuxstb: I guess it's a backwards compability thing again ;)
23:20:11disorganizerthe mf patch kugel just synced is not commitable. needs complete reprogramming imho
23:20:44disorganizerbut the patch is a good sync to keep the patch-users pleased :-) for reprogramming we first need a caching algorIthm
23:21:04Bagdergrep -c '^[a-z]' COMMITTERS
23:21:30*kugel wanna be number 80 ;)
23:21:36 Quit midgey ()
23:21:47kugelno, the 100th is cooler
23:21:49 Quit Rondom ("Ex-Chat")
23:23:09kugelBagder: COMMITTERS isn't utf-8
23:23:26BagderI know
23:23:36*amiconn is an idiot :(
23:23:40markunpixelma: I just noticed he's from Stokholm as well :)
23:24:05 Quit davina (Remote closed the connection)
23:24:06disorganizerkugel: i think you should think about further polish for custom list and ask for a commit so you can concentrate on the next wpsification task :-)
23:24:08 Quit bughunter3 (Read error: 104 (Connection reset by peer))
23:24:15kugelBagder: Isn't it meant to be utf-8?
23:24:19BagderI'm just not sure if everything survives it simply being switched to utf8
23:24:52amiconnThe problem with the hanging greylib when running the isr on cop is rather obvious...
23:24:54markunamiconn: do you have a utf-8 aware editor now?
23:25:21pixelmamarkun: ?
23:25:42amiconnOn 1st/2nd Gen, backlight switching writes to the lcd controller (because of the inversion).....
23:25:54markunpixelma: weren't you talking about Err0r from Meizume?
23:25:58kugeldisorganizer: As soon as I get an enlightement on the issue
23:26:02amiconnAnd when that happens on CPU *while* the cop is also writing to the lcd....
23:26:29pixelmamarkun: yes, my "?" was about the question to amiconn...
23:26:33amiconnTouching the wheel or buttons makes the backlight thread set backlight (repeatedly)
23:27:12markunpixelma: I thought he was against the switch to utf-8 in rockbox files because he editor couldn't cope with it
23:27:24markun2nd he -> his
23:27:59kugeldisorganizer: I asked linuxstb allready, he said doesn't know how to do that too
23:28:33disorganizerkugel: on which issue?
23:28:35kugelAcutally I have an idea, but I'm afraid of giving draw_list etc another argument
23:28:43kugeldisorganizer: The one about customlist
23:28:55 Part domonoky
23:28:58kugelwhich I told you a few minutes ago
23:29:20 Quit Rincewind ("bye")
23:31:12disorganizerthe parsing. i assumed so but wanted to be sure. so how did the "old" non custom version work?
23:32:44disorganizeroh. now i get it. the config file is read on every redraw?
23:34:00 Quit krz (Read error: 110 (Connection timed out))
23:34:32*Nico_P decided to revert most of the track skipping changes of 2008
23:36:52 Quit jhMikeS (Nick collision from services.)
23:36:58 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
23:37:46kugeldisorganizer: you mean the SVN version?
23:38:36disorganizerkugel: yes. anyways, did you mean it reloads the cfg on every reload?
23:38:45 Join stripwax [0] (
23:40:03 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
23:40:43kugeldisorganizer: No, the cfg's are only loaded when you run then in the filebrowser
23:40:58 Join miepchen^schlaf [0] (
23:41:03kugelAll the time, the settings are stored in the RAM
23:41:10disorganizerso what do you mean by parsing on every reload?
23:41:15kugelAnd my patch parses it from the ram at every redraw
23:41:36kugelredraw, everytime the list is redrawn, i.e. when you scroll
23:41:36 Quit mossimo2k3 ("CGI:IRC (EOF)")
23:42:30disorganizerso i dont think thats a real problem. did the svn-version also do this (just with the defined constants for lcd-size)?
23:43:27kugelNot exactly. It doesn't need to parse anything, since it uses the default values
23:44:18disorganizerso the only thing is that your version gets the vars and the svn version uses consts for the redraw.
23:44:51disorganizeri dont believe this has any measurable impact
23:46:35kugelFrom my experience you are right, but I agree that there's a better solution
23:46:54kugelLook, the wps isn't re-parsed constantly too
23:48:10*bluebrother spots the word "Devcon" while scrolling back and looks around interested
23:48:30bluebrotherdo we have a date / location now?
23:48:43*Nico_P just made his third green-delta commit of the day :)
23:48:48disorganizerkugel: any idea how to do it though?
23:49:03*gevaerts checks if there is any code left
23:49:41kugeldisorganizer: No, unfortunately
23:49:50 Quit appleipodder ()
23:50:12 Join druidbartek [0] (n=9c11f0a9@gateway/web/cgi-irc/
23:50:36disorganizerkugel: i dont see one either. anyone having a idea? did jdgordon look at it?
23:51:53amiconnThe desynchronized track index in the wps is nasty
23:52:18 Quit tvelocity (Remote closed the connection)
23:52:42Nico_Pamiconn: I agree, but it's better than unreliable playback
23:53:31Nico_PI tried to find a solution that allowed keeping the synchronized index but it would've been even more convoluted than it was already
23:54:01pixelmabluebrother: not that I know of...
23:54:27Nico_PI think I'll just rewrite audio_check_new_track(), but not this evening
23:54:40bluebrotherwe should really try to find a time / location soon
23:54:56*gevaerts agrees
23:54:58bluebrotherscorche|sh: didn't you want to write an email regarding devcon? Or did I miss it?
23:55:11 Quit toffe82 (Read error: 104 (Connection reset by peer))
23:55:16 Quit druidbartek ("CGI:IRC (Ping timeout)")
23:55:59 Join toffe82 [0] (
23:56:35DerPapstbluebrother: he forgot it :-)
23:58:32mcuelenaereI'm having trouble accessing a 1,8" hdd in windows; after about the first 1000 sectors it complains about a I/O error
23:58:39mcuelenaerealthough it perfectly works under linux

Previous day | Next day