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

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

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

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

#rockbox log for 2008-12-23

00:02:19Dekkardi think its time to go next door and see if they have itunes
00:02:28Dekkard<−−- quitter
00:02:29 Quit itcheg (" ajax IRC Client")
00:02:30MarcGuayonlysoaa: If you're asking about the rockbox bootloader and what it does, you can check out the source code.
00:04:18XavierGrtoo bad that the new build system is quite slower for cygwin usage.... :(
00:09:12XavierGrI don't understand why it wastes so much time into "generating dependencies" it must not be that way in linux but only a cygwin hindrance.
00:09:41amiconnNot for me...
00:10:00 Join ZincAlloy [0] (n=d9eeeddd@gateway/web/cgi-irc/
00:10:24ZagorXavierGr: dependency generation is *much* faster in linux
00:10:24amiconnIt needs about the same overall time for a full build as before. It might seem slower due to the dependency generation
00:14:31XavierGractually it isn't that much slower, 10 secons slower than a previous test (for a target build), I got confused because I built a simulator build that lasted 1 minute more
00:15:01XavierGrbut yeah the dependencies step must be the bottleneck for cygwin if it wasn't for that it would be much faster
00:16:43gevaertsIs %Vl|a|0|0|240|42|-|-|-| a correct viewport pre-declaration for later use with %Vda?
00:18:20onlysoaa@MarcGuay: It's not about the rb bootloader. After disassembling my Samsung P2's original firmware, it seems it accesses registers that aren't set yet, so perhaps they were set by the ROM in the TCC7801 SoC.
00:19:40 Quit bertrik ("Leaving")
00:19:43pixelmagevaerts: I'm not sure if the sysfont/userfont definition can be replaced by - (I usually fill out everything) other than that, it looks ok
00:21:16 Part captainkwel
00:21:23pixelmaCustomWPS says it should be possible, defaulting to user font
00:22:23pixelmabut: you need to define the display viewport %Vd before the load %Vl
00:23:16XavierGrwow compiling the simulator on cygwin with make -j just gave me a segfault!
00:23:20gevaertsYes, I think that's where I'm going wrong
00:29:03*gevaerts achieves enlightenment
00:29:47 Quit ZincAlloy ("CGI:IRC (EOF)")
00:30:10 Quit casainho ("ChatZilla 0.9.84 [Firefox 3.0.5/2008121622]")
00:36:55XavierGrI see in the language files that there are quickscreen options, how does one access them? Is it documented?
00:38:05Zagorisn't it in the manual?
00:38:20XavierGrnot under the quick screen chapter
00:38:41XavierGrand by a quick look on the settings menu I can't find them
00:39:08pixelmashould be in the context menu on the settings, and it's not yet in the manual
00:39:25Zagorit is under each screen. "file browser key controls", "wps key controls" etc.
00:39:45Zagoroh, quickscreen _options_. doh!
00:39:49pixelmaAFAIK it doesn't work for all settings
00:40:31pixelmawell, for some settings it wouldn't make much sense anyways
00:41:00XavierGrpixelma: my problem is that I am trying to translate the entries but can't find them on the rockbox ui
00:41:27XavierGror do you mean that I can hold the context menu button for the setting I would like to send on the quickscreen?
00:41:41XavierGrthere it is
00:42:00pixelmareminds me...
00:42:04 Join jeroen- [0] (n=jeroen-@wikipedia/jeroenvrp)
00:42:04XavierGrsorry but I didn't know how to access them, it seems logical that way but I didn't know about it at all until now
00:42:41***Saving seen data "./dancer.seen"
00:43:31jeroen-hello; I have updated rockbox to the current build - according the Releasenotes of 3.1. I should be able to "configure which items are shown in the quick menu." - ...but where do I configure this?
00:43:58Lloreanjeroen-: 3.1 isn't out.
00:44:05 Quit {phoenix} (Remote closed the connection)
00:44:24LloreanAre you running one of the development builds, or the previous release build?
00:44:25jeroen-Llorean: I know, tomorrow
00:44:40Unhelpfulbut do we have things in the releasenotes that aren't in svn?
00:44:59gevaertsI hope not :)
00:45:02LloreanUnhelpful: We have things in the 3.1 release notes that won't be in a "Release" version someone downloads.
00:45:07jeroen-I'm running rev.19555
00:45:12LloreanSince the 3.1 notes don't apply to the release rbutil would give you.
00:45:15 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
00:45:34Unhelpful3.0 is older than 19xxx, i thought, though
00:45:34Lloreanjeroen-: Just use the context menu on menu items, those that can be added to the quick screen will have the option to
00:45:52 Quit miepchen^schlaf ()
00:45:53LloreanUnhelpful: Yes, but I didn't know which version he had when I first asked, which is why I asked...
00:46:23 Quit ender` (" Thats right, babe, Cochrane chortled. Were not possessors, were just like dimensionally disadvantaged. -- Peter F. H")
00:49:59jeroen-Llorean: I see options to add to the left, bottom or right quickmenu, but that doesnt have any effect on my Ipod 30G
00:50:19 Quit Dekkard (Remote closed the connection)
00:50:34 Quit jhulst (Read error: 113 (No route to host))
00:51:46 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
00:52:28Lloreanjeroen-: The left, bottom, or right options stay the same after you select that?
00:52:57jeroen-what you mean the same?
00:53:17jeroen-I can select those and I return to the item-settings
00:53:52Lloreanyes, and then when you go to the quick screen, the items on the left, bottom, and right are what they were originally still?
00:53:53XavierGrjeroen: after you make your changes (when holding the context button on the setting you want) you should check again the quickscreen and see if that setting was put where you selected
00:54:18jeroen-Llorean: there is no left, bottom or right on my ipod
00:54:26Lloreanjeroen-: Then you're not looking at the quickscreen.
00:54:27jeroen-or is this theme-specific
00:54:44Lloreanjeroen-: have you read the manual, are you familiar with the terms?
00:54:45MarcGuayjeroen-: How are you accessing the quickscreen?
00:55:03jeroen-thats the mainmenu, right?\
00:55:36jeroen-so wheres the quickmenu
00:55:40*MarcGuay groans at the double-titled Quick Screen/Quick Menu
00:55:41LloreanIn the manual...
00:55:52LloreanMarcGuay: Is it called the "quick menu" anywhere offiical?
00:55:58MarcGuayOr triple, even "quickscreen"
00:55:58jeroen-ok ok ok
00:56:19MarcGuayLlorean: I believe the manual references it as the quick menu...
00:56:22jeroen-I never saw it and I use rockbox for 5 months
00:56:30jeroen-but I will look
00:56:37blkhawklater guys
00:56:41LloreanMarcGuay: it's not a menu, though... it shouldn't be that in the manual.
00:56:44 Part blkhawk
00:57:56LloreanMarcGuay: No, the manual only ever calls it "Quick Screen"
00:57:58MarcGuayMaybe not the manual... Search came up dry...
00:58:23LloreanSo in terms of official nomenclature at least, we're consistent.
00:58:48MarcGuayFound one in the FAQ.
00:58:54MarcGuayI'll get it.
00:59:05MarcGuay1 word or two.
00:59:14MarcGuayErr...That's a question.
00:59:16jeroen-I've got it
00:59:23LloreanMarcGuay: Two, the way the manual has it.
00:59:30jeroen-I thought it was possible to change the main menu
00:59:36MarcGuayjeroen-: Nope.
00:59:37jeroen-but I had it wrong
01:04:45gevaertsIs there a WPS conditional that indicates whether or not there is a next track?
01:05:22 Part jeroen- ("Leaving.")
01:06:02 Join toffe82_ [0] (
01:06:08Lloreangevaerts: I don't believe so.
01:06:23*gevaerts smells a feature request
01:06:24pixelmaI usually use %?Fn for that (checks for the next track's filename
01:06:30gevaertsThat works
01:07:33 Quit toffe82 (Read error: 110 (Connection timed out))
01:07:44 Nick toffe82_ is now known as toffe82 (
01:11:31gevaertsWould there be any objections to searching for album art in parent directories up to the root instead of just one level?
01:12:01LloreanIt makes sense to me.
01:12:13LloreanThen someone who wanted to could just have, for example, genre logos or something
01:12:30gevaertsThat's what I was thinking of
01:12:48*Llorean seems to recall discussing this at some point in the past.
01:13:05*gevaerts may work on a patch for this
01:13:16 Quit mirak ("Ex-Chat")
01:13:37gevaertsIf I manage to keep existing behaviour and no real delta, I can't imagine much objection
01:14:18LloreanNor I.
01:14:30 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
01:14:48kugelwell, I think non-aa users will suffer from the extended search time (even more on hdd targets). But it probably won't hurt much
01:15:20gevaertsAh yes. I hadn't thought of that one
01:15:27 Join miepchen^schlaf [0] (
01:16:27LloreanIt probably won't be very many levels deep, and the search happens during a period where the HD is already spun up anyway. I'd be interested to see how much longer it makes buffering take per song per extra folder level, at least.
01:17:21kugeldoes the playback start before the search is finished?
01:18:03kugelthen it shouldn't be a big problem I imagine
01:18:06LloreanOr at least, before the first scale finishes, so I imagine it's not dependent on the presence of the AA.
01:18:12MarcGuayThe manuals use a file called 'utf8.def', but the recommended latex-ucs package has renamed this file utf8x.def, causing manual building errors. Any thoughts on the proper solution?
01:18:17LloreanI don't know if it's the search of the scale that delays AA display currently
01:18:59kugelI've only noticed that the playback starts before the AA is displayed, but I have no idea about the other states
01:19:10LloreanOther states?
01:19:15MarcGuayEither the compiler looks for utf8x.def instead or... I guess the image should have a compatible unicode package installed.
01:19:18Zagorwhy is track_widx and track_ridx not the same when a single file is filling the entire file buffer?
01:19:36gevaertsMarcGuay: didn't bluebrother fix that? a few days ago?
01:19:39kugelI mean "searching", "scaling" and "displaying"
01:20:52Unhelpfulscaling is interleaved with loading. unless we know of a TSC on one of the targets, it's not going to be easy to say how much each takes.
01:21:41MarcGuaygevaerts: Unsure. I just re-installed the VMware image, checked out the source, and tried to build a manual and received errors.
01:22:44Unhelpfulmost recent x86 chips have a cycle counter that you can read. it's often not much good if the CPU speed changes, but it measures clocks.
01:23:20pixelmaisn't the VMware image old?
01:23:41DogBoyimages are always old
01:23:53Unhelpfulthe tick counter does not have sufficient resolution to "subtract" the time spent loading chunks from the total time spent on scale+load, because the whole thing, with possibly hundreds of chunk loads, takes 1-20 ticks in many cases.
01:24:26MarcGuaypixelma: But the latex-ucs package is a fresh apt-get.
01:25:04Unhelpfulmaybe a test build with a very high HZ?
01:25:20pixelmaMarcGuay: missed that in your explanation
01:25:49gevaertsMarcGuay: fresh from what source?
01:26:33MarcGuaygevaerts: co svn svn:// etc.
01:26:46MarcGuayerr. no face, patch.
01:27:26gevaertsI mean the apt-get
01:29:36MarcGuaynon-linux user scratches head. "apt-get install latex-ucs". I presumed it pulled it from whoever distributes the package.
01:31:28Unhelpfulat least some arms seem to have a cycle counter, per google. i don't know if any of ours do.
01:31:53gevaertsMarcGuay: that comes from the debian repository for whatever release you're running, so no new things
01:32:35kugelah great. my c200v2 into rockbox tree patch works!
01:40:01 Join __lifeless [0] (n=lifeless@
01:42:30 Join SUfrEr [0] (
01:42:38 Nick SUfrEr is now known as soap (
01:48:33 Join itcheg [0] (i=62db4767@gateway/web/ajax/
01:49:39 Quit DerDome ("Leaving.")
01:49:41 Join mud-rb [0] (
01:51:55 Quit Photoguy (Read error: 110 (Connection timed out))
01:57:06 Quit _lifeless (Read error: 113 (No route to host))
01:59:00 Join Lynx_ [0] (
01:59:50MarcGuaygevaerts: Alrighty. Any suggestions on where to go from here?
02:21:08XavierGrrasher: I hope it isn't too late: FS #9699
02:21:24 Quit MethoS_ (Remote closed the connection)
02:21:34XavierGr(or anyone else on that matter, it is just an updated translation)
02:22:27 Quit Zagor ("Clint excited")
02:27:36 Quit linuxstb (Remote closed the connection)
02:42:44***Saving seen data "./dancer.seen"
03:13:09 Quit MarcGuay (" ajax IRC Client")
03:17:19 Quit suom1 (Read error: 104 (Connection reset by peer))
03:17:43 Join suom1 [0] (n=suom1@2001:4c40:1:0:0:0:0:11)
03:19:06 Quit Thundercloud (Remote closed the connection)
03:25:43 Join toffe82_ [0] (n=chatzill@
03:26:11 Quit jhulst_ (Read error: 113 (No route to host))
03:31:14 Join webguest71 [0] (n=62c6e5c9@gateway/web/cgi-irc/
03:31:27 Quit webguest71 (Client Quit)
03:34:34 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.4/2008111319]")
03:42:48onlysoaaHello again, anyone familiar with the TCC7801 SoC who can give me a hand at a small task?
03:52:07 Join k-man [0] (n=jason@unaffiliated/k-man)
03:53:32 Quit toffe82 (Read error: 110 (Connection timed out))
03:59:21 Join toffe82 [0] (
03:59:56 Join mellery [0] (
04:01:52 Quit Schmogel (Read error: 104 (Connection reset by peer))
04:17:32 Quit toffe82_ (Read error: 110 (Connection timed out))
04:23:33 Quit jhMikeS (Nick collision from services.)
04:23:39 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
04:28:19 Quit JdGordon (Remote closed the connection)
04:30:18 Join JdGordon [0] (
04:34:55 Quit Lynx_ (Read error: 110 (Connection timed out))
04:38:19 Nick fxb is now known as fxb__ (
04:42:00 Join Minthe [0] (
04:42:47***Saving seen data "./dancer.seen"
04:44:31 Quit Minthe (Client Quit)
04:54:47 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:59:52 Quit mellery ("Ex-Chat")
05:02:22onlysoaaOnce again, hello everyone... Is there anyone who is experienced with the TCC7801 SoC who is willing to share his or her knowledge? :P
05:03:26Lloreanonlysoaa: You are probably better off posting to the mailing list and waiting, rather than repeating that every few hours.
05:04:29onlysoaaI suppose. :D Thanks for the tip.
05:06:38 Join JdGordon|uni [0] (i=7bf38c1f@rockbox/developer/JdGordon)
05:07:24 Nick JdGordon|uni is now known as JdGordon_ (i=7bf38c1f@rockbox/developer/JdGordon)
05:26:39*JdGordon_ wishes there was an easier way to get trnalsations into svn
05:27:13XavierGrJdGordon: Updates or new ones?
05:34:36JdGordon_downloading the patch, applyiong then commiting is annoying
05:35:54JdGordon_ideally it would be nice to setup a web interface so known people can submit patches, maybe with it running through a checker to make sure it doesnt break anything first
05:38:03onlysoaaWow, in my quest to unlock custom fonts for my Samsung P2, I think I've found information that many TCC7801 devs would like, including lots of stuff on the NAND and other things.
05:38:34onlysoaaI even have the complete TCC78xx SDK for whoever's interested.
05:42:45 Join Makuseru [0] (
05:48:23Lloreanonlysoaa: Look, the TCC targets aren't really worked on by very many people. You really ought to try using the mailing list so it can be seen by people who aren't always around.
05:49:08onlysoaaAlright, I'll do that then.
05:49:28onlysoaaWell first I'll see what people need exactly in the forums.
05:52:33 Join Darksair [0] (n=user@
05:57:35 Join Minthe [0] (
06:00:12MintheRockbox r19559 plays wma very well... Thank you.
06:03:03 Quit amiconn (Read error: 110 (Connection timed out))
06:03:20 Join pixelma_ [0] (n=pixelma@rockbox/staff/pixelma)
06:04:03 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
06:04:05 Join Photoguy [0] (n=Photoguy@
06:05:22 Quit pixelma (Read error: 110 (Connection timed out))
06:06:57 Quit faemir ("Leaving")
06:22:56 Quit itcheg (" ajax IRC Client")
06:23:22 Quit Makuseru (Read error: 104 (Connection reset by peer))
06:42:49***Saving seen data "./dancer.seen"
06:45:08 Quit DogBoy (Read error: 110 (Connection timed out))
06:45:51 Quit JdGordon (Read error: 104 (Connection reset by peer))
06:48:26mud-rbis there a coding style guide somewhere?
06:49:13 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
06:50:19mud-rboh, found most of it in CONTRIBUTING, woops
06:55:02 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
06:56:42 Part k-man
06:59:37 Quit Aurix_Lexico (Read error: 110 (Connection timed out))
07:20:12 Join axionix [0] (
07:22:40 Quit lucent ("leaving")
07:24:17 Join lucent [0] (i=lucent@
07:42:11 Join feisar [0] (
07:51:31 Quit JdGordon (Remote closed the connection)
07:51:48 Quit JdGordon_ (" ajax IRC Client")
07:53:27 Quit jhulst (Remote closed the connection)
07:55:07 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
08:06:13 Part toffe82
08:07:15 Join webguest83rockda [0] (n=ad21216f@gateway/web/cgi-irc/
08:07:54 Join Rob2222 [0] (
08:10:40 Quit webguest83rockda (Client Quit)
08:10:58 Join rockdawg [0] (n=ad21216f@gateway/web/cgi-irc/
08:12:54 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
08:14:34 Quit Rob2223 (Read error: 60 (Operation timed out))
08:14:58 Quit jhulst (Read error: 148 (No route to host))
08:15:29rockdawgIs somebody here able to grant me wiki write access?
08:23:51Bagderrwhat's your user name in the wiki?
08:24:56rockdawgits EricCoolman
08:24:57 Join kachna [0] (
08:26:20Bagderryou've been added, enjoy!
08:29:44 Quit rockdawg ("CGI:IRC 0.5.9 (2006/06/06)")
08:38:47 Quit XavierGr ()
08:42:54***Saving seen data "./dancer.seen"
08:47:59 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
08:48:30JdGordonBagderr: hey, got a min to fiddle with build servers?
08:49:50JdGordonive got a static ip now, so change to just, port 22... and if 100KB/s is fast enough then it can do all builds :)
08:50:09JdGordonand get rid of i wont bother setting it up again anyway
08:50:24 Join ender` [0] (
08:51:12Bagderrok, so no ?
08:51:35Bagderrcool, I'm enabling all builds and we'll see how it looks
08:51:54JdGordonjust doing a build to make sure rbclient has the paths all good
08:52:52amiconnBagder: For the release, if we want new flash packages for archos, some things need to be coordinated.
08:52:58 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
08:54:05Bagderr"if" you say. Do we?
08:54:35amiconnThe release .ucl files are needed for building the packages, and the flash image version needs to be bumped in the flash plugins and the manual
08:55:33 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
08:56:19amiconnImo we do. There's that faster & smaller ucl decompressor now.
09:05:55 Join gregzx [0] (
09:09:28pixelmalooking at the german translation updates - midgey's edit reminds me of that non-exception for the "end of playlist" _voice_ string for the Player that's unnecessarily there in english.lang (deutsch.lang doesn't have it and works flawlessly there)
09:11:41pixelmaI hope that change makes it into english.lang one day (probably not for the release) but...
09:17:38 Quit Minthe ("Leaving...")
09:19:42 Quit Darksair (Read error: 104 (Connection reset by peer))
09:20:28 Quit Photoguy (Read error: 104 (Connection reset by peer))
09:24:30JdGordonthere was talk last night about renaming apps/recorder and apps/player... while we are at it can some of the files in apps be moved around a bit also? a folder for playback would be nice...
09:33:55 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
09:36:53*pixelma wonders how "recording,h100,h120,h300,m5,x5,ipod4g,ipodcolor,ipodnano,ipodvideo,e200*,c200:" made it into english.lang
09:37:47JdGordonwhats that for?
09:37:55pixelmait's only in a deprecated string but that list is somewhat useless as all the the targets listed seperately already are covered by "recording"
09:38:20pixelmain the sysfont "disk full" string
09:39:11 Join stoffel_ [0] (
09:39:15pixelmavoice part
09:40:17JdGordonLANG_DISK_FULL has almost the same and is a good indication of what probably happened
09:40:42JdGordonthese strings with buttons in them are annoying
09:41:41pixelmaah, but this difference is not needed in the voice field even if it's there for the language string
09:42:39pixelma(as long as the button name is not part of the voice string)
09:51:16JdGordonhow would people feel about moving all the resume point stuff into the .playlist_control file?
09:52:41JdGordonthe change would be adding 1 more command to that file and removing some ints from global_status (and nvram[.bin])
09:52:58 Quit daurn| (Remote closed the connection)
09:53:07B4gderbut will require disk use on the archos for example, instead of nvram use...
09:53:24JdGordon.playlist_control is written to disk anyway
09:53:38 Quit stoffel_ ("leaving")
09:53:49pixelmaquick brainstorming question for German speakers around: any suggestions on how to translate "quickscreen" - not too long but so that people at least have an idea what it is?
09:53:50B4gderisn't that a bug?
09:54:16B4gderI _never_ fiddle with my playlist(s) on my target, saving the control file seems totally pointless here
09:54:28JdGordonwell, its flushed when the disk is spun up, so not really... it would force a write on shutdown/stop though
09:54:56B4gderbut "flushed" implies that it has contents, doesn't it?
09:55:05amiconnJdGordon: If it works correctly, .playlist_control is only be written if the playlist changes, i.e. not during normal playback. The resume position, however, is written quite often
09:55:51amiconnHence the resume position should stay in nvram, meaning it won't hit the disk if there's real nvram
09:56:27 Join bertrik [0] (
09:56:36JdGordonyeah, hmm...
09:56:49 Quit daurnimator (Read error: 113 (No route to host))
09:59:15 Quit freddy__ (Remote closed the connection)
09:59:58 Join fredddy [0] (
10:01:20 Join lasser [0] (
10:11:34*pixelma is getting annoyed by the "Quickscreen"
10:11:47 Join Genre9mp3 [0] (n=yngwiejo@rockbox/contributor/Genre9mp3)
10:11:47 Join stoffel_ [0] (
10:12:02 Quit Genre9mp3 (Client Quit)
10:22:44 Join daurnimator [0] (n=quae@unaffiliated/daurnimator)
10:42:57***Saving seen data "./dancer.seen"
10:54:57B4gderok, I plan to do my "svn up" for the 3.1 release in ~11 hours from now
10:55:26 Part B4gder
11:01:46 Join n1s [0] (n=nils@rockbox/developer/n1s)
11:02:26 Nick blippe_ is now known as blippe (n=none_of_@
11:05:30 Quit markun (Remote closed the connection)
11:10:58 Join {phoenix} [0] (
11:19:09 Quit lasser ("ChatZilla 0.9.84 [Iceweasel 3.0.4/2008112309]")
11:30:49 Quit stoffel_ (Read error: 113 (No route to host))
11:32:56daurnimatoranyone care to give me the status of rockbox on the fuze?
11:33:50pixelmathe forum might
11:35:47bertrikdaurnimator, the SansaV2 page is up-to-date. Buttons have been identified, as well as the wheel inputs, but they don't quite work yet.
11:37:47_Auron_it's progress though
11:38:08*_Auron_ can't wait :)
11:40:06daurnimatorI know I'm getting a fuze for xmas, and I'd like to get coding on it :)
11:45:00 Join markun [50] (n=markun@rockbox/developer/markun)
11:49:26 Quit {phoenix} (Remote closed the connection)
11:56:36 Join labradorable [0] (
11:58:30 Quit labradorable (Client Quit)
12:02:34 Join Thundercloud [0] (
12:08:17 Join stoffel_ [0] (
12:16:54 Join robin0800 [0] (
12:19:04 Join moos [0] (n=Mustapha@rockbox/staff/moos)
12:20:08 Quit gevaerts (Nick collision from services.)
12:20:20 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
12:21:20 Quit daurnimator (Read error: 60 (Operation timed out))
12:21:35 Join daurnimator [0] (
12:23:44 Quit robin0800 (Remote closed the connection)
12:26:58 Join AndyI [0] (i=AndyI@
12:34:31 Join robin0800 [0] (
12:36:30 Quit robin0800 (Remote closed the connection)
12:36:41 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
12:40:32 Join robin0800 [0] (
12:40:53 Quit robin0800 (Remote closed the connection)
12:43:00***Saving seen data "./dancer.seen"
12:47:09 Nick fxb__ is now known as fxb (
12:56:21 Quit stoffel_ ("Lost terminal")
12:57:06 Join stoffel_ [0] (
12:59:02 Join nplus [0] (n=nplus@141.25.Globcom.Net)
13:01:11 Quit dionoea (Read error: 131 (Connection reset by peer))
13:01:14 Join dionoea [0] (
13:06:23 Join robin0800 [0] (
13:12:08 Join {phoenix} [0] (
13:16:33 Join vertic39 [0] (
13:16:45vertic39is scorche available?
13:17:41 Quit robin0800 (Remote closed the connection)
13:22:39 Quit __lifeless (Remote closed the connection)
13:22:40 Quit AndyI (Read error: 60 (Operation timed out))
13:23:34scorchevertic39: what did you need?
13:23:43pixelmais a flyspray account name possibly restricted to ascii? I'm wondering if the German translator's name is spelled with an ß or not (flyspray account has "Kaspar Rothefusser" but I can also find mails by "Kaspar Rothenfußer" in the Rockbox ML archive). Someone has a suggestion how to deal with that (CREDITS etc.)?
13:24:19vertic39scorche: you told me once you knew better "players" than the ipod 5.5
13:24:24 Quit BigBambi (Read error: 113 (No route to host))
13:24:56 Join AndyI [0] (i=AndyI@
13:24:56scorchevertic39: well, it depends upon your desired purpose, but the gigabeat F devices are what are usually recommended
13:26:41vertic39well, my standard was the iaudio X5V ...let's see if the gigabeat can hold up with that
13:29:08 Join DerDome [0] (
13:29:25vertic39phew this one is even harder to get, right
13:33:41n1spixelma: if he spells the name with ß, it should be like that in credits, we already have non-ascii chars in there anyway
13:36:00 Join _lifeless [0] (n=lifeless@
13:38:24pixelmayes, I think so too. Just wasn't sure which is the correct spelling but since I suspected ß when I first read the name and searching the Rockbox site turned the mails up (and also a wiki account - which seems to be delated now but google still knows about it), I'm more an more convinced
13:39:29linuxstbpixelma: You could email him and ask how he would like his name in the credits.
13:41:45pixelmaindeed... and hope for a quick answer to get it into the release :)
13:47:09 Join Schmogel [0] (
13:47:29pixelmahmm, his ML address looks like a temporary one
13:50:54n1sthe same "addressforads" as he used in FS? :)
13:53:49 Join culture [0] (
13:54:28pixelmayes. I should just try (and also add a comment to the tracker task)
13:56:33n1sis there a reason KBD_MODES is not used on all bitmap targets, it seems to be defined for every btmap target except recorders and the mrobe
13:56:44n1seh s/,/?/
14:00:23 Join __lifeless [0] (n=lifeless@
14:01:03 Quit _lifeless (Remote closed the connection)
14:02:37 Join desowin [0] (n=desowin@atheme/member/desowin)
14:05:23pixelmawhat is KBD_MODES used for, again?
14:08:09n1si think it's for the line edit mode, so you can select the edited line and move around in it
14:08:40n1sand delet/insert chars in other positions than the end without dedicated cursor buttons
14:13:26pixelmaah yes, then it's not used for the c200 yet too. But I already changed it locally...
14:14:44 Join ZincAlloy [0] (n=d9eeceb2@gateway/web/cgi-irc/
14:17:18 Quit moos ("Rockbox rules the DAP world")
14:18:14 Quit Thundercloud (Remote closed the connection)
14:23:28 Quit vertic39 ()
14:23:52 Join gregzx_ [0] (
14:25:15n1sit seems like a nice feature and i see no reason to exclude it for some targets except maybe added binsize on recorders
14:25:16 Join faemir [0] (
14:30:17 Quit gregzx (Read error: 145 (Connection timed out))
14:36:55 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
14:38:33 Join LambdaCalculus37 [0] (i=44a04303@rockbox/staff/LambdaCalculus37)
14:39:16amiconnn1s: There's really no need for keyboard modes if the target in question has dedicated cursor and delete buttons for the input line
14:39:38 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
14:40:02 Quit stoffel_ (Read error: 113 (No route to host))
14:40:31 Join itcheg [0] (i=41d59de2@gateway/web/ajax/
14:40:55n1samiconn: I think it's a bit easier to use for new users ans the only reason not to have it is binsize imho
14:43:01***Saving seen data "./dancer.seen"
14:45:34 Quit Schmogel (Read error: 54 (Connection reset by peer))
14:47:44 Join Ali87sa [0] (n=chatzill@
14:51:50 Join Horschti [0] (
14:52:06 Quit Horscht (Nick collision from services.)
14:57:36 Nick gregzx_ is now known as gregzx (
15:06:09 Quit Ali87sa ("ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]")
15:17:57bertrikn1s, you mentioned your beast possibly has a RDS capable radio chip, right?
15:18:40n1sbertrik: yes, afaict it's the si4701 which has RDS
15:21:50 Quit CaptainKewl (Remote closed the connection)
15:26:15bertrikunfortunately my clip (and probably also the other ams sansas) just has a si4702 which does not contain RDS decoding
15:30:24n1si want to have a go at it soemtime but proper scanning will probably come first
15:31:45 Join stoffel_ [0] (
15:33:26 Quit n1s ()
15:43:12 Join evilnick [0] (i=0c140464@gateway/web/ajax/
15:49:43 Join Strife89 [0] (n=michael@
15:50:51 Join ertzbio [0] (
15:51:57*bertrik wonders whats so special about korea w.r.t. FM settings (band, spacing and de-emphasis)
15:52:00ertzbiohi, i orderd a ipod 5. gen and got a ipod 6gen ;(.... should have been a christmas present.... are you guys working on a rockboxx version for the classic?
15:52:29scorchewe as a group dont work on any port, but as far as i know, no individuals are working on such a port
15:53:00bertrikactually for europe, the korean settings seem to be more correct than the europe settings
15:53:12 Quit synergist (Remote closed the connection)
15:53:29ertzbiook the way i am very happy with rockbox on my iriver h120 ^^ thx :)
15:55:34 Quit itcheg (" ajax IRC Client")
15:55:52 Join itcheg [0] (i=41d59de2@gateway/web/ajax/
16:06:28 Quit stoffel_ (Read error: 113 (No route to host))
16:15:03 Join Aurix_Lexico [0] (
16:15:44 Join BigBambi [0] (
16:36:45 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
16:38:49 Join Galaxor [0] (
16:40:05GalaxorHi. The version of rockbox I've got had a problem. I'm wondering if it's fixed yet. I'm on a sansa c250, and it doesn't have the ability to write to the disk through the usb. I have to reboot into the original firmware to put music on. Is that fixed and I should upgrade?
16:40:48 Quit mud-rb (Read error: 104 (Connection reset by peer))
16:43:02***Saving seen data "./dancer.seen"
16:43:43 Quit Horscht ("electromagnetic radiation from satellite debris")
16:45:54 Quit desowin (Read error: 113 (No route to host))
16:48:01LambdaCalculus37Galaxor: No, that's intended right now. The USB stack isn't 100% finished yet.
16:49:24 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
16:50:33 Quit Strife89 ("Bye, guys!")
16:53:01 Quit Horschti (Read error: 110 (Connection timed out))
16:55:00 Quit DataGhost ()
16:59:05 Join kugel [0] (n=chatzill@unaffiliated/kugel)
17:00:58 Join tyfoo [0] (
17:01:17 Quit tyfoo (Read error: 104 (Connection reset by peer))
17:04:07 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
17:05:54 Part onlysoaa
17:06:05 Quit daurnimator ("Cyas later...")
17:06:06 Join bxcracer [0] (
17:06:06 Join stoffel_ [0] (
17:11:09 Join Vixus [0] (n=quackzil@
17:11:38VixusHey guys
17:11:56VixusI was wondering why rockbox won't display .txt files in the file browser
17:12:30VixusAnyone know why it might not be?
17:12:39Galaxorlambda: Still not finished? Okay. Thanks.
17:13:08kugelLlorean: hey, I was wondering what we do about the quickscreen and the fact that top button changes bottom item
17:14:13VixusShould I be asking in -community? :)
17:14:31kugelVixus: no, but wait until someone answers
17:14:49ertzbiosorry maybe a stupid question...: if you run rockbock on an ipod, there is no need for itunes, right?
17:15:01kugelertzbio: that's right
17:15:02Vixusnope, that's why i switched
17:15:15 Join toffe82 [0] (
17:15:22ertzbiooki thx! :)
17:15:33Vixussorry about the impatience, i'm catching a train soon and my family's in packing turmoil :D
17:15:49evilnickVixus: have you got the browser set to Supported?
17:15:51kugeldoes anyone else feel strong about that? I've put a small patch here:
17:16:10Vixusevilnick: I'm not sure, is it by default?
17:16:25LambdaCalculus37It should be by default. What player have you got?
17:16:29kugelyes, but you might have accidentely changed it
17:16:37evilnickGo to the quickscreen and check
17:16:38Vixusmini 2nd gen
17:16:40 Quit bxcracer (Remote closed the connection)
17:16:40pixelmaVixus: did you check your "flie view" settings? Maybe you accidentally changed it to "music" or so, for txt you would need "supported" or "all"
17:16:54*Vixus checks
17:17:23*kugel assumes he accidentally changed it in the quickscreen by pressung up ;)
17:17:40Vixushow do I access the quickscreen again? :P
17:17:47LambdaCalculus37Hold MENU for a second.
17:17:53kugeldepends on the target
17:18:04*LambdaCalculus37 hands Vixus a copy of the manual for future reference
17:18:51Vixusah you had me down guys
17:18:52kugelpixelma: re: translation of quickscreen: The best one that came into my mind is "Kurz[aus]wahl", not sure if it's good enough though
17:18:53Vixusthanks a bunch
17:19:38VixusI also need to say 3.0 is amazing. (yeah yeah, only just upgraded)
17:20:00Vixusfew mins ago
17:20:10kugelhehe, check back in some hours ;)
17:20:48Vixuswhat are you up to? :}
17:21:02LambdaCalculus373.1 will be releasing very soon. :)
17:21:06kugeloh, it's a secret!
17:21:11*kugel slaps LambdaCalculus37!
17:21:31Vixusno worries, I was guessing as such
17:21:38kugeldon't announce our christmas present :(
17:22:07Vixusman, I totally forgot about christmas. all this travelling
17:22:19VixusI'd love to get in on rockbox development at some point.. maybe start with some plugins
17:22:25pixelmakugel: thanks. I saw it translated for a voice string as "Schnelleinstellungen" and went for that so far ("linke Schnelleinstellung" etc. for the item then). Not that sure about it either but would be consistent at least, it's not easy to translate if you want short and descriptive :\ )
17:23:23VixusI never know how to contribute to an open source project though
17:24:01kugeli think kurzwahl is short and descriptive imho, the major flaw is only that you'd rather connect the term with mobiles phones ;)
17:24:10LambdaCalculus37Vixus: Check out our wiki. We have a ton of pages full of information.
17:24:21LambdaCalculus37And you're also free to download the source and study it.
17:24:35kugelor phones in general, not necessarily mobile ones
17:24:52VixusI think I'll do that, LambdaCalculus37.
17:25:29kugelVixus: so was my assumption right? I mean that you accidentally changed it in the quickscreen?
17:25:50Vixusyeah probably
17:25:57Vixusit was set to Music
17:26:17kugeland you didn't browse the settings yet, did you?
17:26:39VixusI've played around in there (2.5)
17:26:42pixelmawant to commit the language update but didn't receive an answer about the spelling of the translator's name yet.. hopefully Bagder gives a warning so that I could still get it into the release (plan to commit anyways with the IMO most likely spelling if I don't get an answer but want to give it as much time as possible)
17:26:53VixusI don't use playlists and stuff much so I never used the quickscreen except accidentally
17:27:06Vixusdon't use them at all actually :|
17:27:18kugelpixelma: he said he'll svn up his release tree in some hours
17:27:43pixelmakugel: what's a "quickscreen item" then? And btw. I had a similar idea and someone said "but there are shorter menus"...
17:28:48kugelkurzwahltaste :) I.e. "Lege diese Einstellung auf die linke Kurzwahltaste"
17:29:01Vixusman I can't believe I didn't know that you could use the player while charging :(
17:30:14pixelmakugel: I don't want to write novels
17:31:01 Nick __lifeless is now known as _lifeless (n=lifeless@
17:38:23Vixusalright, cheers guys
17:38:35 Quit Vixus ("Returning to packing")
17:42:06kugelpixelma: "Als linke Kurzwahl" (or Schnelleinstellung or whatever)?
17:45:41kugeldoes anyone have an opinion on the "up button changes bottom item in the quickscreen" issue? imho we should find a consensus before the release as it's confusing new users
17:46:11amiconnThat behaviour should be kept for the archos recorders. On all other targets it makes no sense.
17:47:06pixelmaI have to stop thinking about the quickscreen translation, it already feels like the whole day consists of it :\
17:47:42kugelamiconn: I think does that
17:48:14 Quit Galaxor ("Leaving")
17:49:10kugelthat's just a quick hack, proper would be to purge the keymaps from that ACTION_QS_DOWNINV action I suppose
17:50:05amiconnkugel: It looks like it does, but in a non-optimal way
17:50:47amiconnIf we're not going to use ACTION_QS_DOWNINV on anything but recorder, the respective keymaps should have it removed as well
17:50:59kugelyep, that's what I thought too
17:51:05amiconn->saves binsize and unnecessary events
17:52:07kugelamiconn: so if I would do that properly would it go in the release?
17:52:23kugelgevaerts: haha
17:53:34kugelmeh, I'm just gonna do it :)
17:55:23*amiconn thinks that a 4th item can be introduced on target with LCD_HEIGHT >= 96
17:56:18 Join webguest38 [0] (n=479013c8@gateway/web/cgi-irc/
17:56:20kugelthe quickscreen shows on remotes, right?
17:56:33ertzbiois there any other chance if you will not use itunes with an ipod of the 6.gen?
17:56:34 Quit webguest38 (Client Quit)
18:03:09 Join DataGhost [0] (i=dataghos@unaffiliated/dataghost)
18:08:23 Quit Lss ()
18:13:45 Quit ertzbio ()
18:17:40kugelamiconn: I updated, this should be better now
18:23:02 Join GoodGuy [0] (n=GoodGuy@
18:23:26kugelreally, this should be committed before release imo
18:23:44 Join karashata [0] (n=karashat@
18:28:59 Join robin0800 [0] (
18:29:51*kugel sings the can we decide on that before release song
18:30:24 Quit GoodGuy ("The true measure of a man is how he treats someone who can do him absolutely no good. - Samuel Johnson")
18:31:02 Join tyfoo [0] (
18:31:51rasherIs this outright stupid: - allowing the user to turn off hold to continue booting, rather than always shutting down in the bootloader?
18:35:00kugelbertrik: is there any reason you used #defines instead of a enum in r19559?
18:35:41*kugel is just curious about #define vs enum as always
18:36:09bertrikkugel, yes, it's not a random enumeration, the registers indexes have to match exactly with the datasheet
18:37:54 Quit Horscht ("I got raided by the FBI and all i got is this lousy quit message")
18:38:34kugeland you can't get exact matches with an enum?
18:38:52 Join rockdawg [0] (n=ad21216f@gateway/web/cgi-irc/
18:40:02 Quit rockdawg (Client Quit)
18:40:33 Join rockdawg [0] (n=ad21216f@gateway/web/cgi-irc/
18:40:38 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
18:41:11 Quit evilnick (" ajax IRC Client")
18:42:19 Quit rockdawg (Client Quit)
18:42:36bertrikIMO, an enum is used when you need a bunch of different numbers but don't really care what those numbers are, for example to enumerate states for a state machine
18:42:50bertrik#defines are used when you want specific numbers
18:43:05***Saving seen data "./dancer.seen"
18:44:28bertrikan enum is stronger typed than a #define if you care about that
18:45:25kugelyea, that's why I'd prefer enums over defines
18:45:47kugelbesides the fact that it's less typing (which is error prone)
18:50:12bertrikI don't mind typing
18:59:18bertrikI could have used an enum, but in this case I need the values to match up exactly with the datasheet. An enum could be made to match up also, but what if the datasheet skipped a value? would you insert a dummy member, or assign an explicit value? I think both of those defeat the purpose of an enum a bit.
19:07:04 Join Slack_ [0] (
19:08:10kugelbertrik: wasn't the case here, hence I wondered
19:08:28kugelbut I see your point
19:09:59 Join einhirn [0] (
19:11:20 Quit stoffel_ (Read error: 60 (Operation timed out))
19:12:24bertrikand I agree it's not always very clear what to choose, maybe people in here disagree with me
19:17:10*linuxstb doesn't disagree
19:21:40 Join mc2739 [0] (
19:21:43 Join Kitti [0] (n=himka_co@
19:23:59 Quit {phoenix} (Remote closed the connection)
19:24:54mc2739kugel: did you find anything in the Fuze OF concerning gpio A1, or did you find it by trial and error?
19:25:41 Quit Nico_P (Remote closed the connection)
19:25:51kugelmc2739: there's a function which doesn't do anything but setting/clearing that pin. and it's called in dbob_init (and I believe in a function which calls the button read function too)
19:26:39kugeland, fdinel told me to try setting it before reading :P
19:27:24 Part Kitti
19:27:46kugelmc2739: are you familiar with disassembling a bit? if not I could have a look in the e200v2 firmware
19:27:58mc2739do you know the address in the OF where it is located? I'm trying to find something similar in the e20o firmware
19:28:12kugelmc2739: but actually, unless we find a way to trigger an IRQ from the scrollwheel, this is all useless
19:29:37kugelmc2739: it's right before lcd_init_device here
19:30:45mc2739kugel: is that in version 1.01.15a?
19:31:03mc2739ok, thanks
19:31:24kugeladdress is 4AEC (including OF header)
19:31:53kugeluhm wait
19:32:04kugelmc2739: oh wait
19:32:42kugelmc2739: I just see the function sets/clears GPIOA_3_
19:35:09kugelmc2739: but the function that calls the dbop read sets in anyways
19:35:19 Quit robin0800 ("When two people dream the same dream, it ceases to be an illusion. KVIrc 3.4.2 Shiny")
19:35:49fredddyI think someone could commit FS #9617 it compile all plungins now for Fuze
19:35:50 Join MethoS- [0] (
19:36:16kugelfredddy: it's uses until all buttons work
19:37:03fredddyyes I know but I doesn`t hurt if it is in svn or ?
19:39:04mc2739kugel: I didn't see anything that sets gpio A1 in the OF code, unless they are loading the register with the base and then add/sub to address different gpio pins
19:39:37kugelmc2739: do you have found the function which reads DBOP_DIN?
19:40:32mc2739is that the code you sent me? (from fdinel)
19:41:20 Join grndslm [0] (
19:41:33mc2739I have located the same code in the e200v2 OF
19:41:50amiconnbertrik: You know that you can assign explicit values to enum members?
19:42:12 Join onlysoaa [0] (
19:42:14kugelmc2739: that look out for a function which calls that
19:42:22lucentkugel: FS #9617 is a pain to re-sync, should we ignore plugins?
19:42:33amiconn(not that I'd prefer enums for register values)
19:42:39kugelmc2739: i see such a function here even before the lcd_write_* ones
19:43:30kugellucent: I'm ignoring plugins since the first day :)
19:43:44mc2739ok, I'll have another look when I get off work
19:43:48kugelbut how's it a pain to resync?
19:44:19 Quit mc2739 ("ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]")
19:44:32kugelamiconn: any opinion regarding ?
19:46:21lucentkugel: it is common to include the same button behavior as other targets by adding "|| (FOO == BAR_TARGET_PAD)" in defines, but the defines need trailing slashes, and it gets unusual for diff
19:47:07kugelI don't understand
19:47:17lucentso if another target gets plugin support keymaps added, it would disrupt this patch .... maybe this is a small problem only
19:47:50kugelso you mean it should be committed before another target gets keymaps in plugins?
19:48:02kugelsince it's a mess to resync in such a case?
19:48:04 Join ameyer [0] (
19:48:18 Quit einhirn (Read error: 104 (Connection reset by peer))
19:48:52lucentand if another target wants to commit first, we should know to ignore this for some time and work on it after them
19:48:56 Join robin0800 [0] (
19:49:59bertrikamiconn, yes I know
19:50:02gevaertskugel: I agree that FS #9661 is a good idea, but I'm a bit doubtful if it's a good idea to introduce changes just hours before a release
19:50:14lucentE200 and FUZE are very similar with the keymap, the only difference I think is that E200 has "REC" and FUZE has "HOME" buttons
19:52:58kugelgevaerts: it's on the tracker for a few days now...
19:53:33kugelgevaerts: I think this should be in the release, since it's causing major confusion
19:53:39Lloreankugel: The cleaner version wasn't in the tracker for days.
19:53:43ameyergevaerts: there's always 3.1.1 or 3.2/4.0
19:54:04LloreanThere's no significant harm in it waiting for after 3.1
19:54:28LloreanWe've had this "issue" for years now, an extra three months before it makes it into a "release" build won't be a big deal.
19:54:37kugelexcept for the hundreds that accidentally change the file filter again and again
19:54:53*ameyer isn't sure what the heck gevaerts is talking about and doesn't have time to figure it out right now
19:55:14Lloreankugel: The whole point with releases is *not* to introduce changes in intentional behaviour during the freeze.
19:55:27LloreanThis may be a UI improvement, but the current behaviour *is* intentional, this isn't fixing a "bug".
19:56:03kugelI count this as a bug, but ok, not a big deal
19:56:08gevaertskugel: I fully agree that this is wanted, but we decided on time-based releases, and that means that you basically lock the repository a while before release, except for critical bugfixes.
19:56:26gevaertskugel: I count it as a bug too, but it's not a critical one
19:57:33*kugel finds every bug that leads to unexpected behavior and clueless users critical
19:57:41kugelok, not critical, but major
19:57:44Lloreankugel: Then we NEVER have a release.
19:58:02gevaertsNot major either
19:58:06Bagderrelease-plan: ~2 hours until my svn up
19:58:08bertrikclueless users probably won't even be aware of the quickscreen :)
19:58:18gevaertsMajor means the user lost his data
19:58:23kugelyou don't need to be aware of it to get to it
19:58:28kugelas the one today has shown
19:58:32LloreanI don't think it's a "bug" if it's described in the manual.
19:58:56LloreanOr, rather, if it's expected and *should* be described in the manual, as I don't know if this is but it should've been.
19:59:04kugelI count this as a bug, since the quickscreen doesn't indicate at all that the up button is doing anything
19:59:08LloreanRight now the behaviour was intentional, it's just able to confuse people.
19:59:31Lloreankugel: It does indicate.
19:59:33 Join miepchen^schlaf [0] (
19:59:38LloreanWhen you press the up button, the on-screen value visually changes.
20:00:01kugellol, I wouldn't call that indicating
20:00:16LloreanSo visual feedback doesn't indicate the button has done something?
20:00:17kugelindicate means to me the make the user aware of it before he does anything
20:00:28LloreanYou said "is doing anything"
20:00:30LloreanNot "will do something"
20:00:40LloreanI can't read your mind, I'm sad to say.
20:01:16LloreanVery few screens in Rockbox actually display indications as to what buttons do what in those screens, anyway.
20:02:17lucentwhat about "Press PLAY to confirm" and the user has to press select, PLAY cancels?
20:02:30lucenti.e. to rebuild the database
20:02:35LloreanThat's a bug.
20:02:46LloreanAt least, if it's happening on a supported player.
20:02:55 Part grndslm ("Leaving")
20:03:06bertrikI know this happens on some AMS sansas, e.g. clip
20:03:11LloreanIf it's happening on one of the development ports, it's probably just related to keymaps being incomplete.
20:03:13gevaertsI'm pretty confident that this patch has not been tested on most of the targets it touches. *Any* code can contain bugs. That means the question is "Is this serious enough to postpone a release?", and I think the answer is no, because if it is we probably won't release again anytime soon
20:03:19LloreanThe languages will need to be updated once the keymap is settled.
20:03:54lucentI haven't heard confirmation that it does the right thing on a supported target, so that's why I ask
20:04:09kugelgevaerts: the "unclean" version doesn't introduce bugs. That's why I put that up instead of the one which touches multiple files
20:04:10Lloreanlucent: You can test in the sim.
20:04:31kugeli.e. apply the safe and unclean to the 3.1 branch, the cleaner one to the trunk
20:05:49kugelgevaerts: and again re few hours before the release: The task is open for a while, enough time to comment on (no matter of the patch attached)
20:07:09kugelLlorean: and the manual mention of the quickscreen (besides that it's outdated and doesn't contain the customizablity) doesn't indicate that up does something.
20:07:20*amiconn thinks 9661 should be applied in full, also to the 3.1 branch
20:07:23Lloreankugel: I already mentioned that it may not...
20:08:50kugelamiconn: I agree (obviously)
20:08:53gevaertskugel: don't bring comments into it, or I'll be tempted to point out what the consensus from the current comments apart from yours is ;)
20:09:15Lloreankugel: There's a -dev mailing list you can use when you think your patch warrants committing to ask people to look at it.
20:10:13 Join aurix_lexic1 [0] (
20:10:38amiconnSpeaking about the tracker - imo it would be a good idea to have a "flyspray" section as part of the "recent activity" on the frontpage
20:11:31kugelsounds like a good idea to me too
20:12:04bertrikamiconn, something like the 10 newest tasks, or the 10 most recently edited tasks?
20:12:38Bagdereither way would be great
20:12:39*gevaerts decides to test-build 3.1 with 9661 applied, and test this on a variety of players
20:12:45amiconnMost recently edited, and just 5 items like in the other sections
20:13:01lucentsome tasks get edited like they are a forum though
20:13:27amiconnCould be more on recent.html
20:13:32bertrikanyway sounds like a nice idea to me too
20:13:49 Join Zagor [242] (
20:13:57Lloreanlucent: Lots of activity is good, usually.
20:14:02kugelBagder, amiconn, bertrik: and while we're at it, provide RSS feed for each recent activities?
20:14:02lucenthow to include this useful view of the tracker but discourage Forum Idols from posting needlessly to bump their attention to the front page of something?
20:14:10amiconnBtw, am I the only one who doesn't understand why the sections in recent.html have a different order than on the frontpage?
20:14:22Bagderkugel: well, that's usually more work but sure
20:14:38*amiconn never used RSS so has no opinion on that one
20:14:52kugelI regulary use the opened tasks one
20:15:21kugelgevaerts: that's cool!
20:15:50 Quit fredddy (Read error: 110 (Connection timed out))
20:16:14 Join fredddy [0] (
20:17:27kugelBagder: you're svn up'ing your release tree in 3 hours?
20:17:27Lloreanlucent: If that becomes a real problem, we can address it once it is. We don't worry about people editing the Wiki or posting to the ML
20:18:11lucentLlorean: not everyone who visits the wiki gets access to it, they have to ask for access. Okay though.
20:18:27Lloreanlucent: Yeah, but everyone who asks gets access as long as they're polite about it.
20:18:31LloreanAnd anyone can sign up for the mailing list.
20:20:20ZagorI keep missing nico_p :-(
20:20:27 Quit ZincAlloy ("CGI:IRC (Ping timeout)")
20:23:59gevaertsAnyone with a Player around?
20:24:33Zagorgevaerts: yes
20:25:15gevaertsMy Player doesn't seem to work properly with the 3.1 branch today. It stops playing after a few seconds to a few tens of seconds
20:25:48Zagorouch. I'll dig mine out and check.
20:25:58gevaertsBut as I don't use it very often, I have no idea if this is hardware or software
20:26:25ZagorI haven't used mine in years either.
20:26:34 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
20:28:22 Join tyfoo2 [0] (
20:28:56Zagordo we have anything written about the playback/buffering code, other than MetadataOnBuffer? does anyone other than nico_p know the philosophy/design idea behind the code?
20:31:23 Quit aurix_lexic1 ("Leaving.")
20:32:07Zagorfor example I find some of the thread signalling unnecessary complex. what is the point of setting the watermark in a signal rather than a straight function call?
20:38:16Zagornot to mention: what is the advantage in splitting this up in three different threads?
20:38:22 Join akur [0] (
20:38:37 Part akur
20:41:44kugelZagor: It's known that playback code needs some love, so I wouldn't necessary assume "that's intended/needed"
20:42:48 Quit tyfoo (Connection timed out)
20:42:56Zagorsince the current code is the result of a 2007 gsoc project, I think there could be more than just "natural evolution" behind it.
20:43:08***Saving seen data "./dancer.seen"
20:44:02Zagorbut primarily, I don't want to poke too much in it until I've established what is and is not intentional
20:45:23kugelZagor: it is? you mean playback or mob?
20:45:36kugelI was talking about playback, not mob
20:47:04linuxstbZagor: I think there are some wiki pages about playback/buffering but I don't know their names. IIRC, pondlife has been involved in it a little, as well as Nico_P
20:47:21Zagorplayback and buffering are rather intimately tied together, and share many design elements
20:49:26linuxstbZagor: There is this page -
20:51:44Zagoryay, my player has cvs-041209
20:52:06gevaertsI hope you don't want support for that ;)
20:53:06Zagorlinuxstb: thanks
20:54:55 Quit Aurix_Lexico (Remote closed the connection)
20:56:55Zagordo we have 3.1rc builds somewhere?
20:58:01gevaertsThe one I tested is home-built. I'll test one of Bagder's builds soon
21:01:36 Quit TouchOdeath (Read error: 60 (Operation timed out))
21:01:55Zagorplayer works fine after 60 seconds
21:08:48Zagorthree tracks in, not a glitch. looks fine to me.
21:09:58*gevaerts tries again
21:15:41 Join jhulst_ [0] (n=jhulst@unaffiliated/jhulst)
21:15:50kugelgevaerts: thanks for testing
21:16:23 Quit jhulst (Read error: 113 (No route to host))
21:16:56gevaertsZagor: it seems fine here now as well. I reformatted the filesystem in the meantime (I accidentally unplugged without unmounting, so the old one wasn't very good anymore), so maybe my issues were due to FAT corruption
21:17:58lucentinteresting, the Player is a new thing to me
21:18:16lucentI guess the BeagleBoard wasn't around when they started? (
21:18:22 Quit onlysoaa (Read error: 131 (Connection reset by peer))
21:19:28kugelBut I guess it still won't be committed in the release branch?
21:19:35 Quit jhMikeS (Read error: 110 (Connection timed out))
21:20:46 Quit jhulst_ (Read error: 60 (Operation timed out))
21:21:37Zagorlucent: the Player came in 2000.
21:21:46 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]")
21:21:47gevaertskugel: I'd like to see a few more opinions from other devs
21:21:54lucentthanks Zagor
21:22:24lucentZagor: but where is Player on that page?
21:23:05kugelgevaerts: what opinions? whether this is a bug or not, or whether it should be in the release?
21:23:06Zagorlucent: heh, it's the third one. Archos Jukebox 6000.
21:23:52lucentZagor: oh I'm confused, I thought "Player" as referenced by developers here today, I thought that this meant the open-hardware design?
21:23:52kugelZagor: you may comment on too :)
21:24:00gevaertskugel: whether it should be in 3.1. Whether it's a bug or not is not very interesting IMHO
21:24:39*kugel doesn't really see why anyone would object a tested bugfix
21:24:50kugelbut lets see ;)
21:25:21Lloreankugel: Well, it's not a "bugfix" it's a "behaviour change"
21:25:32LloreanBluebrother at least suggests he has a use for the existing behaviour.
21:25:43LloreanA bugfix would be simply making the arrow point up and down so people know either button works.
21:26:19kugelor my patch, since it's already been brought up introduce a 4th item
21:26:39 Join Jaykay [0] (
21:26:40LloreanYour patch doesn't introduce a 4th item.
21:26:45Lloreanit simply removes an existing, used behaviour.
21:26:48kugelI know
21:27:01LloreanSo, from some perspectives, it's a "break", not a "fix"
21:27:51 Quit LambdaCalculus37 (" ajax IRC Client")
21:27:52Zagorkugel: I hardly ever use the quickscreen so I don't have an opinion
21:28:14LloreanIt seems like this could wait until we actually know what we want to do with the quickscreen.
21:28:38Jaykaycan anybody fix my wiki account JohannesLinke? i cant login..... and i dont have a wiki page....
21:28:57Jaykaywhen i try to login i get a blank page with this address:;def=topic_access;param1=change;param2=access%20not%20allowed%20on%20web
21:29:01kugelLlorean: if existing behavior is confusing and leads to unexpected behavior for new users, then I don't mind breaking it
21:29:19Lloreankugel: Fix the manual, then.
21:29:24kugelespecially since the behavior is already perfectly covered by the down button, which does exactly the same
21:29:29LloreanThat way new, confused users can be helped without breaking existing use patterns.
21:30:16 Quit robin0800 (Read error: 104 (Connection reset by peer))
21:31:12LloreanMy point is, there are several ways to reduce user confusion without removing existing behaviour, I don't understand the rush to remove an existing method of using the screen, especially since the complaint simply seems to be "the screen doesn't tell them the button press will do something" and that can be solved without making the button press do nothing.
21:31:24 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
21:31:55 Join robin0800 [0] (
21:32:28Unhelpfula "fix" might be to indicate what it does on the quickscreen?
21:33:32Lloreankugel: Your patch looks worrisome for the H100 remote
21:33:53kugelLlorean: I'm open to other approaches, but I picked the one which is most unlikely to introduce bugs (given that it should be in the release)
21:33:58Lloreankugel: on the quickscreen the three options are set to three bidirectional toggles, where either direction changes the value. I think INV still needs to work for the H100 remote, to preserve that behaviour
21:34:07LloreanSo, yours does have bugs still.
21:34:15kugeland "fixing the manual" is another task, since it doesn't even contain the customizability, see here
21:34:47 Join thegeek [0] (
21:34:52pixelmaand that patch only helps a bit
21:34:54 Quit faemir ("Leaving")
21:34:56kugelLlorean: I don't understand
21:35:14Lloreankugel: On the HXXX-series remotes, INV needs to not be removed.
21:35:19LloreanThere's no directional control
21:35:24kugelLlorean: this INV thing is because you can't flip the buttons as you flip the screen
21:35:39LloreanInstead each of the three options is associated with one "wheel" that can be turned either direction, and both directions change that one option
21:35:50kugelthat's not given on other targets, so the buttom button is flipped
21:36:16kugelI mean, you cannot do that on recorder, but on every other targert
21:36:38Lloreankugel: You need it on the H100 LCD remote, too, at least
21:36:54LloreanBUTTON_RC_VOL_UP needs to work the same as BUTTON_RC_VOL_DOWN in the quickscreen to mirror the behaviour of the other two quickscreen items.
21:37:08LloreanYou'll notice both FF and REW do LEFT
21:37:16LloreanAnd SOURCE and BITRATE do RIGHT
21:37:45LloreanYou can just set it to _DOWN instead of _DOWNINV of course
21:37:55LloreanBut the removal of it entirely is bad.
21:38:14kugelI was going by amiconn who said this INV is only needed for recorder and not for the other targets, and he agrees with my patch
21:38:24Lloreankugel: This is why testing is needed.
21:38:32LloreanINV was used there because it was around.
21:38:32lucentJaykay: it's possible your login info does not exist, try registering again
21:38:38LloreanINV isn't necessary, but the two buttons doing the same thing are.
21:38:45LloreanThis is why patches should have a little time testing, instead of rushing them in. It's entirely possible there are other unique cases like this that haven't been noticed yet.
21:38:57Jaykaylucent: i could already login a few days ago, but illl try
21:38:58LloreanWe don't want to _break_ things in the name of fixing something that at least _works_ right now.
21:39:00gevaertsJaykay: try pinging Zagor (if he isn't pinged by this)
21:41:14kugelLlorean: I didn't get your h100 remote example I think
21:41:34Lloreankugel: If you want it to work like the left and right options, it needs to be the way it was before.
21:41:48LloreanIt's possible to substituted _DOWNINV with _DOWN too, I believe
21:41:50kugelso VOL_UP and VOL_DOWN both change the bottom item? with my patch it would only be one of these
21:41:57Jaykaylucent: i registered again, i now have a wiki page, but when editing my page i get this blank page again
21:42:00LloreanBut if you remove it entirely, the bottom option will work completely differently than the two side options
21:42:07Lloreankugel: On the REMOTE it should be both of them still
21:42:21lucentJaykay: :(
21:42:37kugeldoes the screen on the remote indicate that both buttons change that one item?
21:42:42LloreanFF/REW both change the left item on the remote, and SOURCE/BITRATE both change the right item. VOL_UP/DOWN should both change the bottom option.
21:42:47kugelif not, my patch perfectly applies to them too
21:42:53LloreanThe screen on the remove doesn't match the buttons available at all anyway
21:43:33Jaykaylucent: if the smiley should get the account working, it didnt work ;)
21:43:38Jaykayzagor: ping
21:43:40pixelma <- the upper left one I think
21:43:40Lloreankugel: If you want to go that way, you need to remove either _FF or _REW from doing the left option, and either SOURCE or BITRATE from doing the right one, and then it makes the screen vastly less usable overall.
21:43:49lucentyeah ping Zagor ;)
21:43:50ZagorJaykay: yep, look at it now
21:43:55Lloreanpixelma: Yes, the upper left one.
21:44:09ameyerhmm, I take back what I said about 3.1.1. If fs#9661 doesn't make it into 3.1, it should be pushed back to 3.2 or 4.0.
21:44:14Lloreankugel: Each of those tabs (the two above, and the one below) can go left or right. Both directions change one of the quickscreen items.
21:44:25Lloreanameyer: 4.0 isn't what comes after 3.2
21:44:25kugelLlorean: why? the buttons are just flipped as the screen is flipped
21:44:33kugelthey're still doing both
21:44:36Lloreankugel: It has nothing to do with screen flip.
21:44:47LloreanThey're all horizontal buttons.
21:45:23ameyerLlorean: I figured 3.2, but I haven't heard anything official. The next major release after 3.1, expected to be released in late March.
21:45:36kugelok I understand, the situation is even worse on the remote
21:45:42Lloreankugel: Think like a mousewheel that can only move a little bit. You can turn it either way, and it changes the value.
21:45:45LloreanThe situation is BETTER on the remote
21:45:49LloreanIt actually makes perfect sense right now.
21:45:54LloreanIn an odd sort of way.
21:45:58kugelmutliple buttons doing the same thing is something I'm strongly against
21:45:58ameyerit shouldn't make it into 3.1.x, if there even is a 3.1.x
21:46:08LloreanEach of them points at a "tab" thingy, and that tab thingy changes the value no matter which way you move it.
21:46:21LloreanYour suggestion makes one of the tabs only work in one direction, while the other two tabs still work in both directions
21:46:24ameyer(oh, and it's too late to touch 3.1 unless there's something major, IMO)
21:46:58 Quit karashata ("G'bye everyone!")
21:46:59Lloreankugel: Multiple buttons doing the same thing makes perfect sense on that screen, with the remote. I suspect you won't understand until you actually hold one.
21:47:33LloreanMy point is, though, this behaviour is made much more inconsistent by your patch.
21:47:34ZagorJaykay: could you try simply restarting your browser?
21:47:39LloreanSince it makes one control work differently than two other controls
21:48:01kugelLlorean: oh yes, "much more inconsistent"
21:48:05Jaykayzagor: ill do
21:48:08 Quit Jaykay ("ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]")
21:48:11Lloreankugel: Was that sarcasm?
21:48:30kugelyes, sort of
21:48:39kugelbut I see your point with the remotes
21:48:47 Join Jaykay [0] (
21:48:48Lloreankugel: It really does. Think that you have three light switches. Two turn the lights on and off. One though does nothing when you flip it up, but turns the lights on and off when you flip it down.
21:48:52LloreanYou've just made that third light switch.
21:49:14LloreanThey used to all change the lights every time they switch, but now one only works in one direction.
21:49:23Llorean"the lights" being "whatever option they controlled"
21:49:47Jaykayzagor: didnt work
21:50:05kugelLlorean: but if a 4th item is added you need to break remotes too, or am I wrong?
21:50:07ZagorJaykay: ok
21:50:27Lloreankugel: The remotes either have no free buttons, or three free buttons that do not fit the quickscreen anyways.
21:50:28 Quit fredddy (Connection timed out)
21:50:33 Join fredddy [0] (
21:50:55LloreanEven if you add a 4th item, the button you've freed up would be *more* confusing for that 4th item.
21:51:14LloreanThe remote doesn't have Up/Down. What it has is three "left/right" buttons, two on top and one on bottom.
21:51:39LloreanThe top left left/right button controls the left quickscreen item on left or right. The top right one controls the right quickscreen item. The bottom one controls the bottom quickscreen item.
21:51:52LloreanNow the bottom one only works on left, not on right, while the top two can still toggle with left or right presses.
21:52:00kugelI mean, are the remotes a reason do reject a 4th item too, just because the button situation is odd?
21:52:34ZagorJaykay: how are you testing? trying to edit your own page?
21:52:38Lloreankugel: They'd be more confusing with a 4th item than the current screen is now.
21:52:53LloreanSo, by your standards, they'd absolutely reject a 4th item for fear of user confusion.
21:53:12Jaykayi tried editing my own page, and editing and attach files to some other pages like faqs and sansaruntime
21:53:57ZagorJaykay: any edit you do gives you that empty accessdenied page?
21:54:22kugelLlorean: now there (random, but generally fitting numbers) 90% confused, with my patch the other 10% are confused
21:54:25Jaykayyes, everytime i type my account informations in and klick ok
21:54:41kugelbut with the difference that the 10% don't accidentally change an option due to that
21:54:44Lloreankugel: You think 90% of our user are confused right now/
21:55:25ZagorJaykay: try now
21:55:28Lloreankugel: Right now, your patch breaks remotes on at least two targets. Why not fix those, test it, and then commit it, instead of trying to rush in improperly tested fix and hope it doesn't break anything you haven't thought of yet?
21:55:52kugelI'm already doing this while we're discussin here :)
21:55:58Jaykayzagor: it works now......
21:56:19ZagorJaykay: you simply didn't have edit rights.
21:56:23Lloreankugel: You don't have the hardware to actually test it...
21:56:48Jaykayzagor: then someone "deleted" them, i had them some time ago.....
21:56:53Lloreankugel: As we've clearly established, you don't care about the H100 remote because you don't understand the physical layout, and so don't hold it as an important problem with the patch (or don't believe it is, because you disbelieve I'm acting in good faith)
21:57:18LloreanI find it hard to believe, at this point, you'll make good calls on other hardware you haven't handled either.
21:57:45LloreanWhy not request testers in the -dev mailing list, get people to actually report back whether QS is broken or apparently working on each of the remotes and targets, etc.
21:57:55ZagorJaykay: ah. I did a cleanup run a month or so ago, and removed all accounts which had regisered >x months ago and had never made an edit. perhaps that matched you?
21:58:53kugelLlorean: sure I don't care about remotes, you got it right, you rumbled me
21:59:10gevaertsZagor: IIRC Jaykay registered a few weeks ago at a time when a wiki issue caused new people to not get their user page. My guess is that somebody "cleaned up" the access page
21:59:18Jaykayzagor: i think my registration was not so long ago....
21:59:21Lloreankugel: This is the problem. Your goal is to get "quick fixes" in rather than taking your time and solving the whole problem.
21:59:22Zagorgevaerts: aha
21:59:28Lloreankugel: Why is patience such a bad thing to you?
22:01:15kugelsolving the whole problem cannot be made until there's a decision on a 4th item. And I searched for a quick fix for the release, to stop people being from being confused and helpless
22:01:43LloreanA quick fix is "adding an up arrow"
22:01:44linuxstbkugel: Hasn't Rockbox behaved like this for years though?
22:01:52LloreanThen you've solved it without potential bugs.
22:02:17LloreanBoth your old quick fix and your new quick fix break remotes, this is not a good way to "fix" it.
22:02:35kugelLlorean: No, certainly not, since it requires rearranging the viewports and stuff in the quickscreen, which is imho not suited as a commit just before the release
22:02:37Jaykaybetter than nobody fixes it in any way
22:02:44Jaykayor in a few months
22:03:00Lloreankugel: You could put the V right next to the ^ on the same line, couldn't you?
22:03:27LloreanOr is the V in its own tiny viewport?
22:04:04LloreanWell then, describe it in the manual, and find an icon to replace V with that indicates up and down.
22:04:08kugelbut ok, we'll do it all right after the release, your remote point is certainly valid
22:04:24LloreanA single character substitution should be pretty darn safe.
22:04:47kugelthere's no icon compiled in
22:04:58 Join massiveH [0] (
22:05:08Lloreankugel: Yes, but you can make one and still do it more safely.
22:05:11 Quit massiveH (Client Quit)
22:05:18kugelit uses the stock ff/rew/down arrows that are compiled in (like the pitchscreen)
22:05:32LloreanMy real point though is that you haven't done full testing on this patch, but you're trying to rush it in.
22:05:40kugelLlorean: sure, that's something that shouldn't go in just before the release
22:05:45LloreanYou're not interested in testing.
22:08:00kugelyep, that's true
22:08:09kugelreal men don't test, that's my opinion
22:08:46Lloreankugel: Drop the sarcasm.
22:09:33kugelthen don't claim that I don't care about abc or that I'm not interested in xyz.
22:09:57LloreanYou're the one who claimed "(1:04:05 PM) kugel: gevaerts: the "unclean" version doesn't introduce bugs. That's why I put that up instead of the one which touches multiple files"
22:10:01LloreanBoth versions introduce bugs
22:10:14LloreanYou wanted to commit either version without further testing, that indicates to me you're not really interested in it.
22:10:18*linuxstb isn't sure this discussion is still helpful...
22:10:31Jaykayis it right that the list of supportet players in is outdated? or did i understand something wrong?
22:10:44*Jaykay nods at linuxstb
22:10:53kugelLlorean: I don't think the remote stuff is a bug, as opposed to the behavior on non-remotes
22:11:13Lloreankugel: It's most certainly a bug.
22:11:14kugelor if it's a bug, it's one that hurts less to the majority of users
22:11:28LloreanThat's not the point. You didn't bother finding out if the bug existed.
22:11:32LloreanWe don't know if there are other bugs.
22:11:49LloreanYou're basically saying "I don't know if there are other bugs, but I hope it'll help so it should be committed"
22:12:03LloreanIn what is supposed to be a _stable_ release.
22:12:43 Quit jhulst (Read error: 60 (Operation timed out))
22:14:13gevaertsJaykay: I may be nitpicking, but that line doesn't really have a list of supported players. It has a list of FAQs, but I think that's still complete (although I could be wrong)
22:15:04Jaykaygevaerts: im sorry, i meant "what is rockbox"
22:15:08Jaykaythe next paragraph
22:15:34LloreanJaykay: Which one's it missing?
22:15:52gevaertsJaykay: ah ok. seems complete to me
22:16:08Jaykayllorean: it says rrockbox is for the archos line, for the rest its under developement
22:16:38Jaykayall the mentioned targets are fully supported and others are under development
22:16:47gevaertsJaykay: you know, you need to spell out everything for us :)
22:16:53*gevaerts agrees now
22:17:38Jaykayill fix it.... i just wanted to be sure im doing it right
22:18:24kugelLlorean: so, what's your proposal? I guess now it's to late anyway to do something about the remotes
22:19:29 Part linuxstb ("Leaving")
22:19:43 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:21:05Lloreankugel: I told you how to fix it quite some time ago.
22:21:15Jaykayare X5, M5, and M3 including all the targets mentioned on the main page?
22:21:19LloreanYou went on to argue about the remotes not really mattering instead of just fixing it.
22:22:09kugelLlorean: ok, but those other fix need to wait for post-release
22:22:26LloreanI told you how to fix _your patch_ too.
22:23:00LloreanYou either need to pick a direction and make all the rockers work in the same direction, or you need to keep them both omnidirectional by replacing the DOWNINV with DOWN for the remote.
22:23:09 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
22:23:11gevaertsJaykay: there are some minor differences, but they are mostly the same hardware. X5V has no FM, and the L ones have a bigger battery
22:23:15LloreanThe latter is much, much better for the way the hardware works / makes sense.
22:23:31Jaykaygevaerts: should i write them all down in the faq?
22:23:44gevaertsI don't think that's really needed
22:23:55Lloreankugel: you'll need to find someone who actually has experience with the other two remotes (non-LCD and H300 LCD) to find out which behaviours make sense for them.
22:24:05LloreanI don't think they have the same rocker situation the H100 LCD does
22:24:19kugelit's too late for that I think
22:24:29LloreanThe patch shouldn't go in without further testing anyway
22:24:33LloreanWho knows what else is broken, right?
22:24:48kugelI mean, I could test on sims, but that's apparently not enough
22:25:13kugelI tested on sims and the quickscreen still worked, so it wasn't an apparent bug to me
22:25:38gevaertsYou won't find button issues in the sim
22:26:20LloreanEspecially when you have unique inputs like the rocker switches, that aren't well simulated by two (well, three) keys instead.
22:27:55kugelI see
22:28:29LloreanThe "Left, Right, Down" layout of the quickscreen doesn't match the controls on the remote at all, because there is no normal directional control.
22:28:54LloreanBut there is a "left" a "right" and a "bottom" rocker, physically, so it actually is pretty intuitive anyway.
22:29:56*Bagder arrives and does an svn up
22:30:06kugelso I'm wondering how to integrate the 4th item
22:30:28kugelI mean the task can be closed now, as it was aimed to get the confusion out of the release
22:30:40gevaertsBagder: tools/release/ still has version 3.0, but you're probably aware of that
22:30:44LloreanThe 4th item either needs to be controlled separately (there's still a couple face buttons) or uncontrollable from the remote.
22:30:52Bagdergevaerts: yeps, I'll commit a 3.1 change first
22:31:01kugeland there's no real fix, as I think there should be a 4th item, not just thee (where suitable of course)
22:31:05Lloreanif you tried to use what's currently DOWNINV, it'd be impossible to guess especially if you'd ever used either the left or the right one already.
22:31:45kugelI assume a 4th item is kinda difficult to display on the quickscreen anyway, especially with statusbar enabled?
22:32:09LloreanOn the remote, almost certainly
22:32:19kugelyes, I meant on the remote
22:32:19LloreanIt's kinda cramped for space.
22:33:54kugeldropping the quickscreen from the remote isn't liked I guess? It's not going to get easy, without multifont
22:33:56LloreanI'd say if there is a 4th option patch, it's reasonable to only have 3 controllable from some remotes if that makes more sense.
22:34:31LloreanWhy would you drop the whole quickscreen, rather than just one option?
22:34:35kugelthat would mean to have 4 items on the main screen, but only 3 on the remote. That's confusing too imho
22:34:59kugelbut probably the only way
22:35:10LloreanYou don't look at both screens at once anyway, and it's less confusing that having the remote go blank when you bring up the quickscreen.
22:35:10*Zagor enjoys glitch-free mp3 on clip
22:35:16Bagderr19569 is 3.1
22:35:18LloreanZagor: Got the buffering mostly worked out?
22:35:19*gevaerts thinks outside the box a bit
22:35:37gevaertsWho says remote quickscreen and main quickscreen need to have the same options on them?
22:36:01ZagorLlorean: most of it, yes. I still have an odd bug where sometimes you have to pause/unpause to get the music to start though.
22:36:02bertrikZagor, a remark like that is nothing without a patch! :P
22:36:17Zagorbertrik: coming up
22:36:27LloreanZagor: I remember running into that one too.
22:36:44Lloreangevaerts: I'd say it'd make more sense if they didn't, if that's doable
22:36:59kugelI suppose if the quickscreen is going to look differently on the remote, it could be more optimised for the buttons in that run
22:37:13Jaykaygevaerts: could you look through the mentioned faq-paragraph? im still not sure i did it right....
22:37:25kugellike showing the options not in directions but above each other
22:37:29gevaertsLlorean: I mean, if you can have three options on the remote and four on the main screen, you potentially have seven quickly accessible options
22:38:10Lloreangevaerts: That's what I was thinking, but more along the lines of, if you want to quickscreen from the remote, you probably want playback settings, if you want them from the main unit, you may want file view to still be around, for example.
22:38:27gevaertsexactly, yes
22:38:33 Quit jhulst (Remote closed the connection)
22:38:46gevaertsJaykay: maybe add a link to the WhyRockbox page there as well?
22:38:46 Join faemir [0] (
22:38:59Zagorouch, still glitches on 320kbit
22:39:02kugelso a totally seperate quickscreen? that sounds like a good idea to mee
22:39:19gevaertsIt's at least worth thinking about I think
22:39:40Zagorhmm, I need better test tracks
22:39:56LloreanZagor: Does FLAC still work?
22:40:06kugelyea, it's the track's fault if the clip is glitching :)
22:40:24 Join Thundercloud [0] (
22:40:28Zagorkugel: hehe, I didn't mean that
22:41:30 Quit faemir (Remote closed the connection)
22:42:17 Join faemir [0] (
22:42:59Jaykaygevaerts: imo this requires a rewrite of this paragraph, im to tired to do it now :) ill look through the whole faq because i spotted some more outdated things.....
22:43:11***Saving seen data "./dancer.seen"
22:43:38kugelZagor: what did you all need to change? only the watermarks?
22:44:16 Join leachim6 [0] (i=614c84e8@gateway/web/ajax/
22:45:08leachim6...if I use the Apple OS, all my albums and songs display fine...but when I look at it in the Database view in rockbox, it says I have doubles of every single song
22:45:14leachim6is this a known bug ?
22:45:41kugelLlorean: btw: to show that I actually care about remotes, I think I did a good job when optmising the pitchscreen for the remotes/small screen (which was 1 goal when viewport'ifying the pitchscreen), didn't I?
22:46:01LloreanDunno, don't use that screen.
22:46:10kugelah ok
22:46:11Zagorkugel: yes, only the watermarks.
22:46:24kugelZagor: uhm..not good
22:46:41kugelas that probably doesn't help to get mp3 playing on a fuze/e200v2
22:46:57Zagorhow much ram does they have?
22:47:10leachim6mb or kb ?
22:47:17kugeleven though the free audio buffer doesn't quite match with
22:47:17Zagorand you get the same skipping as on svn clip?
22:47:41kugelI wouldn't call it skipping, rather reboot upon starting playback :S
22:48:18ZagorI'll put up an early patch. you can always try.
22:48:26leachim6so anyway
22:48:28kugelcool, will do
22:48:40leachim6is there a way I can delete my database all together
22:48:41leachim6and start from new
22:48:51Lloreanleachim6: "initialize now" should delete it
22:48:59LloreanIf you really want to do it the manual way, delete all the .tcd files in /.rockbox
22:49:02 Quit faemir (Read error: 104 (Connection reset by peer))
22:50:28leachim6database is being retarded...
22:50:35leachim6no...I'm sorry
22:50:36kugelLlorean: sorry if I was trying to rush it in, it's just that it's a major problem in the quickscreen to me. But let's forget about it, it'll get a better fix after the release
22:50:41leachim6Apple is being retarded
22:50:47leachim6by making the stupid db in the first place
22:51:11kugelwhat database are you talking about?
22:51:22leachim6the database feature of rockbox
22:51:32leachim6the ability of it to read the iPod db
22:51:34 Quit Jaykay ("ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]")
22:51:43gevaertsIt doesn't
22:51:43LloreanNo, the database feature of Rockbox doesn't read the apple one.
22:51:50LloreanIt has nothing to do with it. We create our own database and read it.
22:51:57leachim6my bad
22:52:09leachim6I meant the ability of it to decipher the apple file-naming scheme
22:52:18Bagderdo we have rbutil builds around ready for 3.1?
22:52:27pixelmait doesn't read the Itunes db and if you don't want to listen to your songs in the Apple firmware you are perfectly fine just copy-pasting your songs to the device
22:52:30kugelZagor: the database tool compiles now?
22:52:46leachim6pixelma yeah but I like to use the apple firmware from time to time
22:52:53kugelhow about putting the tagcache into a plugin (the database browser remains in the core)?
22:53:00gevaertsBagder: I think so. bluebrother and domonoky posted a few links a few days ago
22:53:09leachim6mainly because rockbox does not as of yet support lyrics
22:53:11leachim6at least emedded ones
22:53:30Zagorkugel: yes it does. I haven't verified how well it works though.
22:53:57Zagorthe songdb patch is an improvement of the database tool
22:54:00Bagderzips done...
22:54:23 Join gregzx [0] (
22:54:27leachim6any significant changes in 3.1 ?
22:54:29leachim6should I upgrade ?
22:54:35gevaertszips done here too :)
22:54:43kugelZagor: I mean, wouldn't it be worthwhile to put the database creation stuff outside of the core and only keep the reader in?
22:54:52gevaertsleachim6: have a look at
22:54:59kugelas far is I understood tagcache and tagtree are already pretty much seperated
22:55:07leachim6rockbox has come so far in such a short time
22:55:17leachim6I remember last january it wasn't even able to play sound
22:55:24Zagorkugel: I haven't really looked into the code that much
22:55:28leachim6is rockbox going to be participating in SoC ?
22:56:33gevaertsleachim6: maybe on your specific model, but rockbox itself is a lot older than that
22:56:44kugelleachim6: what target are you talking about when you say it couldn't play sound last january?
22:56:46leachim6on the iPod 5g I mean
22:56:49 Join faemir [0] (
22:56:53leachim6I should have specified
22:57:00kugeluhm, that can play music a lot longer
22:57:11Bagdergevaerts: did the fonts zip build for you?
22:57:27gevaertsBagder: no
22:57:46pixelmakugel: wouldn't that mean disk spin ups when browsing the database?
22:57:53Bagderhm right, that fix was only done in trunk...
22:58:21Lloreanpixelma: No, browsing would still be in core, building would be in a plugin.
22:58:36kugelpixelma: I mean only to move the creation tool into a plugin
22:58:50kugelreading the database remains in the core
22:59:02leachim6kugel: what would the advantage of that be ?
22:59:04pixelmamisunderstood then, but how about auto-update and the liek?
22:59:08 Quit Slack_ (Read error: 104 (Connection reset by peer))
22:59:14kugelleachim6: binsize
22:59:18leachim6...I like the integration the db has now
22:59:28Lloreanpixelma: We can do background plugins, I think it'd work as long as you didn't want to use another plugin while it was running.
22:59:37leachim6I wouldn't mind if some of the advanced audio settings were abstracted to plugins
22:59:53 Join Slack_ [0] (
22:59:54leachim6but database is a HUGE feature
22:59:55kugelpixelma: auto-update doesn't quite work as it should anyway, does it?
22:59:56Lloreanleachim6: I doubt it would be done if it couldn't be done transparently to the user.
23:00:02leachim6kugel: not at all
23:00:04kugeli mean it doesn't detect deleted songs afaik
23:00:11leachim6kugel: very true
23:00:17Lloreankugel: It detects deleted songs.
23:00:23leachim6I just had to delete the db and start again
23:00:27leachim6because I had doubles and triples
23:00:27Lloreankugel: It doesn't detect modified tags.
23:00:31pixelmaI think it does if dircache and database in RAM is enabled
23:00:38kugelthe auto update too? I thought it doesn't do that for performance reason
23:01:20 Quit Slack_ (Client Quit)
23:01:26LloreanEither way, couldn't a plugin auto-update just as easily as the core?
23:01:28leachim6when updating rockbox
23:01:33leachim6should I use the apple OS or rockbox ?
23:01:36leachim6does it make a difference
23:01:53kugelLlorean: I suppose yes, but you couldn't do a battery bench with it enabled
23:01:54Lloreanleachim6: Rockbox doesn't have USB on the iPod, so you're stuck using the Apple OS right now anyway
23:02:11kugeland a plugin that always runs isn't really worth it
23:02:15leachim6leachim6: is that planned ?
23:02:21Lloreankugel: You couldn't do a battery bench while the database is updating. This would be true without automation too
23:02:21leachim6...I just talked to myself...great
23:02:40leachim6Llorean: are there any plans to make a version of rockbox available for the iPod touch ?
23:02:44leachim6I wouldn't think so...
23:02:56Lloreanleachim6: They aren't planned. They happen if people do the work, nobody's doing it.
23:03:16leachim6...well...I don't see an advantage to you ?
23:03:26Lloreankugel: It wouldn't have to always run. It would stop when the update is done, and re-run when an update is needed. Auto-update would just be a background update called by the core instead of by the user.
23:03:27leachim6rockbox was created because the features of the players were insufficient
23:03:43leachim6the iPod touch OS is badassin enough as it is ?
23:04:00krazykitleachim6, it has nothing to do with "advantages", but everything to do with "does somebody want it enough to do the hard work of getting it to run"
23:04:09Lloreanleachim6: The iPod Touch music player lacks most of the features of rockbox, but discussing it is irrelevant.
23:04:22kugelLlorean: Are we talking about the same thing? There's "update now" and "auto update", I'm talking about the latter one
23:04:23leachim6Llorean: yeah, but there is third-party software to make up for it
23:04:39kugeland I thought "auto update" is always running, not just sometimes
23:04:40Lloreankugel: Think about it. "Auto update" is basically an "update now" that happens under certain conditions.
23:04:51leachim6...y'know...this conversation isn't even relevant at all...nevermind
23:04:53LloreanIt may always be running now, but there's no reason it needs to be always running.
23:05:04leachim6the little cd in the top-right corner means what ?
23:05:08leachim6the db is updating in the bg ?
23:05:16Lloreanleachim6: It means the disk is being accessed, that's all
23:05:31Lloreankugel: Since we are the OS, we know when files have been moved, etc, as well as knowing when USB has happened.
23:05:35kugelI gues it's worth a try at least
23:06:02leachim6is anyone working on reading embedded albumart in rockbox ?
23:07:30kugelI don't think so
23:08:10Lloreanjpeg album art needs to be done first.
23:08:18 Join Slack_ [0] (
23:10:42Bagderrbutil 1.0.8 binaries uploaded
23:12:09leachim6so....pictureflow is just a glorified slideshow ?
23:12:13leachim6you can't get to the music through it
23:12:16leachim6just look at the albums
23:12:33gevaertsShould RockboxUtility be updated to point to 1.0.8?
23:12:47*gevaerts means the wiki page of that name
23:12:58Bagderyes, the release is pending
23:14:06BagderI'm holding on the release announcement mail just to give the download mirrors more time to mirror the bins
23:14:35kugelZagor: you add 1.5 to an int?
23:14:42 Join onlysoaa [0] (
23:14:46Lloreanleachim6: That's why it's in the "Demos" category, yes.
23:15:06gevaertsBagder: do you have URLs for the rbutil binaries handy?
23:15:22leachim6Llorean: to be is nice to showuff with
23:15:28Bagdergevaerts: exactly like the 1.0.7 ones but with 1.0.8
23:15:42leachim6is anyone working on lyric support for RB ?
23:15:46gevaertsBagder: except for the one "b" I guess?
23:15:47Lloreanleachim6: Just like anything, unless someone shows up and does the work it won't get improved.
23:15:58leachim6no I here right now
23:16:01Bagderah yes, the 'b' is not there
23:16:32Lloreanleachim6: I was more talking about pictureflow.
23:16:33Bagderand we have no 64bit linux rbutil 1.0.8 yet
23:17:15 Quit itcheg (" ajax IRC Client")
23:17:15kugelZagor: do you think the calculation is needed at all for flash? I assume you can safely hardcode a very low watermark for those
23:17:22Lloreanleachim6: Most things that people have actually done any work on beyond a certain point get posted to the patch tracker. You can check there for things
23:17:38leachim6link ?
23:17:45Llorean"Patches" on every page of the site.
23:18:50gevaertsBagder: done (but obviously not tested very hard, as the mirrors aren't there yet)
23:18:51linuxstbWhat do others think about making rbutil for windows available as a .exe download, rather than a .zip? The .exe is already compressed, so the size different is negligible - it just adds an extra step for the user.
23:19:25scorche|shi would prefer that, yes
23:19:38kugelhm, no, it's better to caluclate one
23:19:40 Quit Horscht ("Snak 5.3.3 Unregistered copy. Evaluation period is over. Program will now quit. Thanks for using Snak.")
23:20:14Bagderlinuxstb: I agree, a zip with a single compressed exe inside seems pointless
23:21:49gevaertsHow do you build rbutil statically?
23:22:41bertrikhm, I'm not getting any sound at all playing an mp3 with the watermark patch on a 1GB clip
23:23:22Lloreanbertrik: Pause and resume
23:23:46bertrikdidn't have an effect
23:24:37Zagorkugel: even on flash, we want a little margin. and the margin differs a *lot* between wav and 32kbit mp3.
23:25:08kugelyea, I already came to that conclusion :)
23:25:09Zagorbertrik: odd
23:25:34kugelbertrik: sounds like what I experience on my fuze (it doesn't always reboot, sometimes just refuse to play)
23:26:02Zagorbertrik: does your "view buffering thread" show anything strange?
23:26:46bertrikI'll do a make clean and revert any other (probably unrelated) patches and try again
23:26:52 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
23:27:08*kugel waits for his clip to arrive
23:28:07LloreanZagor: "last file in the buffer" meaning "previous file played" or "the last file buffered"?
23:28:18Llorean"the file at the end of the buffer" rather.
23:28:34Zagoryes, "the file at the end of the buffer"
23:28:53LloreanI can see that as being fairly problematic with mixed playlists.
23:29:11LloreanWell, a 770kbps FLAC followed by a 128kbps MP3?
23:29:32kugelit's calulating the watermark using the bitrate of the very last file
23:29:44kugelno matter of which format that is
23:29:54Zagorit is the last file that needs the watermark, the others are already in the buffer
23:30:19kugelZagor: I just wonder how vbr proof that is
23:30:22gevaertsWon't that go wrong with a very short very low bitrate file at the end of the buffer, followed (in the playlist) by a 24/96 wav?
23:30:27 Join Hillshum [0] (
23:31:04LloreanI think I got my example backward.
23:31:12Zagorgevaerts: as soon as the 24/96 is loaded the watermark is recalculated
23:31:15LloreanBut I think either way it's okay.
23:31:50bertrikZagor, mp3 playback seems to work ok now, after pausing/resuming
23:31:58Zagorbertrik: nice
23:32:10gevaertsZagor: can't that be too late in some cases?
23:32:17LloreanBut if a small portion of the last file is in the buffer, you'd have a bad estimate of how many seconds of audio remained. If the prior file was higher bitrate you might have less seconds than you think, right?
23:32:52LloreanLike, if you have a 770kbps FLAC and then 0.2 seconds of MP3 and you need 2 seconds for spin up and seek, don't you end up with about 0.7 seconds of actual audio before you start rebuffering?
23:33:02ZagorLlorean: true. total and buffered length of the last file should be checked
23:33:12LloreanThat's what I was trying to say, only badly.
23:33:13bertrikI do notice some skipping/stuttering now
23:33:40Zagorbertrik: yes I got some too after a while. it came and went in a strange way.
23:36:45kugelanyone want me to build a 64bit linux rbutil build?
23:37:37gevaertskugel: you know how to build statically?
23:37:43Zagorflac works fine btw
23:38:27kugelgevaerts: it's mentioned here:
23:38:57 Join itcheg [0] (i=62db4767@gateway/web/ajax/
23:41:35linuxstbAnyone know why the rbutil version numbers are prefixed with "m" in some places, and "v" in others?
23:42:06amiconnBagder: Did you bump the flash image version in the plugins and the manual before starting the 3.1 builds? Or is the final build not yet running?
23:42:28gevaertskugel: I won't bother I think. I don't feel like building Qt
23:42:31BagderI didn't, and the builds are already done pretty much...
23:42:48gevaertsamiconn: I changed 3.0.1 to 3.1 in the manual
23:43:07*gevaerts isn't sure that this is what's being talked about
23:43:14amiconnI'm not referring to the rockbox version, but to the flash image version
23:43:24kugelgevaerts: yea, I just noticed that I haven't built qt at all, so this needs to wait
23:43:33amiconnThe ones with the 3.0.1 release are v2
23:44:45amiconnThe version is compared whenever you flash an .ucl, and a message is displayed if it's not the latest
23:45:04amiconnThis means that rockbox_flash.rock needs to know which is the latest
23:45:40gevaertsAt least we have more things right than for 3.0. At this rate 3.2 will be perfect :)
23:46:57amiconn*If* we want new flash images, they need to be rebuilt. The ones currently available for download contain rockbox 3.0.1, so if you changed the manual to state 3.1 ...
23:47:53amiconnAnd new flash images mean bumping the version, hence a rebuild of the release for archos
23:47:54gevaertsWe could just change the manual and rebuild.
23:48:01amiconnThis is in fact a hen-and-
23:49:10amiconnegg problem, as bumping the version in svn means that the next build will have a rockbox_flash that complains about an outdated flash image, while building the flash image requires the release .uclk
23:49:32 Quit perrikwp (" ajax IRC Client")
23:49:39amiconnThat's why I didn't do it in advance, but wanted to wait to the last minute, which is now over :\
23:51:17 Join Acky [0] (
23:51:47Bagdervoices and manuals are now being copied over
23:55:15 Quit tyfoo2 ("Carpe diem")
23:56:56gevaertsAt least one mirror has the new rbutil now
23:57:31BagderI'll mail in 2 minutes to get the release date right ;-)
23:58:51 Part onlysoaa

Previous day | Next day