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

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

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

#rockbox log for 2009-03-06

00:00:07*stripwax guesses ... sudo rbutil ?
00:00:33BigBambiOr at least someone told me it wasn't when I was checking whether to add it to the manual
00:00:33LambdaCalculus37(Assuming rbutil is in the /Applications folder in OS X) "sudo open -a"
00:00:48BigBambiLambdaCalculus37: Surely it is either needed or not?
00:01:02LambdaCalculus37BigBambi: I don't think it's needed, but I use it anyway.
00:01:07bluebrotherBigBambi: it seems to not be required for ipodpatcher but for some reason for rbutil
00:01:31BigBambiLambdaCalculus37: So if it isn't needed we shouldn't tell people to use it
00:01:35BigBambibluebrother: Ah, OK
00:01:58LambdaCalculus37bluebrother: It doesn't seem to be required for Sansapatcher, either.
00:02:02BigBambiI assume that that is the case for ipods and sansas?
00:02:09bluebrotherBigBambi: no idea why this is happening or whats causing this
00:02:40BigBambiI'll pop a note in the manual then if it is needed
00:03:28BigBambiLambdaCalculus37: I assume the line you gave above is the standard OS X one?
00:03:29LambdaCalculus37BigBambi: You have permission to pop a note in the manual. :)
00:04:00LambdaCalculus37BigBambi: That is assuming the user has copied the file into their /Applications folder.
00:04:35BigBambiLambdaCalculus37: Would that be normal, or would people download it to the desktop for instance?
00:04:38*BigBambi has no idea how OS X works
00:06:02stripwaxHow do I confirm if rockbox regular build server uses arm-elf-gcc 4.0.3 and arm-elf-ld 2.16 ? (wondering if my cygwin builds are using something that's recently become incompatible with something)
00:06:57JdGordon|doesnt it say in the server build log?
00:07:04domonokystripwax: if you hover over the numbers in the build server, you can see which gcc and ld server this particular build server used..
00:07:11BigBambiMind you, we don't tell people how to do root/admin for linux/windows either, just that they need to
00:07:14JdGordon|or that :)
00:07:24 Quit JdGordon| (" ajax IRC Client")
00:07:30LambdaCalculus37BigBambi: OS X doesn't care where apps are run from, so people can copy them anywhere.
00:07:34 Join JdGordon| [0] (i=836b0049@gateway/web/ajax/
00:07:36BigBambiSo maybe it would be enough to just add OS X to that
00:08:01stripwaxdomonoky - ah, right - thanks!
00:08:09LambdaCalculus37BigBambi: The /Applications folder is there for you to just keep your files in a safe place. But anyway, that's enough of that talk. :)
00:08:45 Quit J-23 (Read error: 145 (Connection timed out))
00:08:56BigBambiLambdaCalculus37: Sure, but if they were running it from the desktop say, would they run a different command - e.g. change to the corect directory then run sudo ./rbutil like linux?
00:09:12 Join salty-horse [0] (
00:09:22 Quit flydutch ("/* empty */")
00:09:25 Join J-23 [0] (
00:09:35LambdaCalculus37BigBambi: In those sorts of cases, you would just switch to the folder and type "open <foo-app>"
00:09:41 Quit ender` (" Anyone who thinks people lack originality should watch them folding roadmaps.")
00:09:45salty-horsehi. getting a warning about character clipping in some fonts when running "make zip" −− is this a known problem?
00:10:14BigBambiLambdaCalculus37: So open is the OS X version of sudo?
00:10:26LambdaCalculus37BigBambi: No, open is to launch an app.
00:10:28BigBambisalty-horse: You can ignore them
00:10:50*LambdaCalculus37 will be in the other channel with OS X talk
00:11:18 Quit domonoky (Read error: 104 (Connection reset by peer))
00:11:27 Quit peanus (Read error: 104 (Connection reset by peer))
00:11:35salty-horseok BigBambi
00:13:28BigBambiLambdaCalculus37: a) this is OS X regards to rbutil and b) we have been talking about sudo so far, so we suddenly switch to open?
00:14:14 Part salty-horse ("Leaving")
00:14:14BigBambiLambdaCalculus37: The whole point is that rbutil on OS X seems to need root rights
00:15:13BigBambiLambdaCalculus37: Perhaps you could re-read my questions and answer again :)
00:16:09bluebrotherwould be interesting to know why this issue happens for rbutil but not the patchers
00:16:37LambdaCalculus37BigBambi: In OS X, if you have to launch any app with root permissions (and not commandline apps like wget or anything), and assuming the application package resides in /Applications, you can type "sudo open -a <>" in the Terminal.
00:16:44 Quit dfkt (Connection timed out)
00:17:05 Quit dfkt_dt (Read error: 110 (Connection timed out))
00:17:06LambdaCalculus37If you just double click the app in the Finder, it'll run without root permissions.
00:17:16 Join dfkt [0] (
00:17:23 Join dfkt_dt [0] (i=dfkt@unaffiliated/dfkt)
00:17:26BigBambiLambdaCalculus37: Right, and if it isn't in /Applications, but instead they have downloaded it to e.g. the desktop?
00:18:34*linuxstb wonders if peanus's problem was simply because he "formatted as fat32", rather than "converted the ipod to fat32" - i.e. if the Apple firmware still worked?
00:18:48linuxstbLambdaCalculus37: Didn't you succeed using rbutil on OS X without root rights?
00:19:12LambdaCalculus37linuxstb: No. I had to sudo it.
00:19:26 Quit n1s ("Lämnar")
00:19:53bluebrotherLambdaCalculus37: have you checked the permissions on the executable? Maybe some x bits are set differently?
00:19:59BigBambiAnd can you run an application that is not in /Applications with sudo, and if so, how?
00:20:38LambdaCalculus37BigBambi: Then in that case, what you would do is cd to the folder where the app package resides, and then type "sudo open <>"
00:20:54 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:20:54 Quit pixelma (Nick collision from services.)
00:20:54 Quit amiconn (Nick collision from services.)
00:20:54 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:21:06LambdaCalculus37bluebrother: Haven't checked them. I should do so.
00:21:06 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:21:12 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:21:19BigBambiThank you
00:21:45LambdaCalculus37linuxstb: Older versions needed sudo. I haven't tried with 1.0.9; just with the SVN build bluebrother gave me.
00:24:27linuxstbLambdaCalculus37: Are you sure? Maybe you had the same problem that stopped that version working the first time?
00:25:23LambdaCalculus37linuxstb: It's possible. I want to try testing this again, though.
00:26:13bluebrotherLambdaCalculus37: have you tested the svn binary with sudo too?
00:26:40linuxstbbluebrother: Does rbutil do anything other than search for an ipod's raw disk device?
00:27:07LambdaCalculus37bluebrother: No, I didn't. And I also can't find the build you rolled for me; can you send me another one?
00:28:25bluebrotherlinuxstb: during the remount wait? No, it only checks if the disk device that was used for installing is mounted again (using the getmntinfo syscall)
00:28:44bluebrotherLambdaCalculus37: domonoky created that one, I don't have the binary :(
00:29:04linuxstbbluebrother: No, I mean before the bootloader is written - i.e. before the "no ipods found" message?
00:29:15BigBambiI think we need to sponser either domonoky or bluebrother a Mac :)
00:29:15LambdaCalculus37domonoky: I need a new SVN build of rbutil for OS X.
00:29:19 Join gromit`` [0] (
00:29:24*LambdaCalculus37 needs a new Mac ;)
00:29:34bluebrotherit only runs the ipodpatcher checks
00:29:56BigBambiLambdaCalculus37: I was thinking more for rbutil development :)
00:30:57bluebrotherBigBambi: LambdaCalculus37 just needs to start developing on rbutil ;-)
00:31:25BigBambibluebrother: Or that :)
00:31:29JdGordon|LambdaCalculus37: you actually need one or just access to one? I can hook you up woth vnc/ssh access to my mini back home...
00:31:38JdGordon|its on a fast connection so vnc shuold be useable
00:31:46JdGordon|and its already got xcode/gcc
00:31:54 Quit bertrik ("Leaving")
00:32:24LambdaCalculus37JdGordon|: Sure, email me the details and I'll set it up tonight.
00:32:35stripwaxListening to drive access on OF usb file transfer versus Rockbox usb file transfer on ipod video, it sounds like OF is performing quite a bit of caching - with Rockbox file transfer I can clearly hear seeking for each individual file written, but with OF file transfer there's plenty of silence (just spinning) from the drive while many files are 'copied' followed by a burst of seeking activity from the disk
00:33:01 Join kachna|lappy [0] (
00:33:11JdGordon|LambdaCalculus37: ah crap.. i just remembered vnc server in osx is useless so only the logged in person can use it... :<
00:33:27JdGordon|my family might not like you trying to dev while they are watching tv :p
00:34:45 Quit MethoS- (Remote closed the connection)
00:35:09LambdaCalculus37JdGordon|: Ahh well... it was worth a shot. :)
00:35:40JdGordon|ssh access is no problem though if you want it
00:36:04LambdaCalculus37Sure. Email me the address to log into and details.
00:36:16*LambdaCalculus37 heads home
00:36:19 Quit LambdaCalculus37 (" ajax IRC Client")
00:38:27***Saving seen data "./dancer.seen"
00:38:50stripwaxSo maybe OF is simply caching writes in memory and flushing cache periodically and on eject. Not a bad strategy especially when host OS has been told to 'optimize for quick removal'.
00:39:19stripwax("This setting disables write caching on the disk and in Windows", etc)
00:45:17 Quit ZincAlloy ("CGI:IRC (EOF)")
00:45:50 Quit bluebrother ("leaving")
00:45:54stripwaxSigh, ipod no longer responds to menu+select reboot - is there any known workaround or do I need to let the battery run down?
00:46:14stripwax(btw I think FS #8668 is the cause of my crash, turns out build wasn't quite clean)
00:46:17JdGordon|flick the hold switch and try again?>
00:48:44 Quit gromit` (Read error: 110 (Connection timed out))
00:50:03stripwaxJdGordon - any idea how long for? tried flicking hold 'on' for a minute, flicking back to hold 'off' for a minute, then holding menu+select for a minute..
00:50:33JdGordon|no idea... i thought it just needed a quick flick
00:50:46stripwaxhm, never mind. seems to work with a really quick flick. thanks
00:59:29 Quit JdGordon| (" ajax IRC Client")
01:11:47 Quit planetbeing_ ()
01:23:33 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
01:24:33 Quit bmbl ("Woah!")
01:27:28 Join homielowe [0] (
01:31:19 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
01:37:15 Join Nico_P [0] (
01:37:46 Quit J-23 (Read error: 104 (Connection reset by peer))
01:37:46 Join J-23_ [0] (
01:43:12 Join fdinel [0] (
01:46:08 Join CaptainKewl [0] (
01:46:38 Quit homielowe ()
01:47:29Unhelpfulhwcodec pictureflow patch... works on sim, and i don't see anything obviously wrong. any comments/suggestions?
01:47:46 Quit stripwax (Read error: 104 (Connection reset by peer))
01:48:16 Join homielowe [0] (n=homielow@unaffiliated/homielowe)
01:49:55kadobanUnhelpful: is there supposed to not be a _ on line 78 ("UNIQBUF SIZE")?
01:50:18 Join peanus [0] (n=peanus@
01:50:26peanusfuck yeah i got it installed
01:51:14peanusHey, thanks, whomever, for now dumbing down the manual-install instructions. it took 4 hours, but it forced me to learn a ton about how my computer and ipod works.
01:51:22kadobanoh, looks like it's just displaying wrong. scratch that
01:51:31peanus(that wasnt sarcasm)
01:51:34Unhelpfulhuh? it looks like it says "UNIQBUF_SIZE" to me, both in the line removed and its replacement.
01:51:45Unhelpfulweird font issue?
01:52:00kadobanyeah i guess...something about the highlighting seems to mess it up
01:53:08Unhelpfulit's really pretty straightforward, add one API export missing on hwcodec, and use the audio buffer if the plugin buffer is tiny.
01:54:08kadobanlooks okay to me, for what little that's worth :)
01:58:17 Part toffe82
01:59:45 Join Hawson_ [0] (
02:03:31 Quit jhMikeS (Nick collision from services.)
02:03:37 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
02:04:07 Join thomc [0] (
02:05:00 Join Hawson__ [0] (
02:06:59thomcHi my iPod (Video 30GB) is formatted just for use as an external hard drive. It doesn't have any software on it at all. When I try to install the bootloader with the ipodpatcher script it reports "no ipods found". The ipod is mounted as a drive and I can access it like that. Am I missing something?
02:07:16krazykitthomc, yes, you need a working original firmware
02:07:33scorchethomc: see the IpodManualRestore wiki page
02:08:23thomcRight, thanks very much.
02:10:01 Join Hawson___ [0] (
02:10:01***Alert Mode level 1
02:10:01DBUGEnqueued KICK Hawson
02:10:01DBUGEnqueued KICK Hawson_
02:10:01***Alert Mode level 2
02:10:01DBUGEnqueued KICK Hawson__
02:10:01DBUGEnqueued KICK Hawson___
02:10:01***Alert Mode level 3
02:10:26 Quit Hawson_ (Read error: 60 (Operation timed out))
02:13:14 Quit MTee (Remote closed the connection)
02:13:29 Join Darksair [0] (n=user@
02:13:35 Join rocko [0] (
02:14:21 Quit Hawson__ (Read error: 60 (Operation timed out))
02:15:56 Join gregorovius [0] (n=diego@
02:17:12scorcheIf no one has any more valid ideas for the SummerOfCode2009 wiki page, perhaps we can flesh out the existing ideas with a basic outline of what the implementation of X idea might be?
02:17:45scorcheas the wiki page stands, i am not so confident of our getting accepted to participate in GSoC this year...
02:17:57 Quit Hawson (Read error: 110 (Connection timed out))
02:19:19pixelmaUnhelpful: you tested in an Ondio sim, right? Just trying to find out whether to expect something weird wrt controls...
02:19:41pixelmapictureflow that is, of course
02:20:00LloreanI think both the "Touchscreen" and "Lua" features on there fall more under "wishlist" than anything else.
02:20:02 Join Hawson [0] (
02:20:03***Alert Mode OFF
02:20:23scorcheLlorean: agreed, but the last thing we need to be doing to the page is pruning it...
02:20:25 Quit Hawson___ (Read error: 60 (Operation timed out))
02:20:45scorcheof course we need to only be accepting projects that we would like to see, but we really really need more ideas - not less of them
02:20:49Lloreanscorche: Is it actually required that we have a large number of ideas?
02:21:04LloreanI thought the idea was that students could also propose things as the time comes.
02:21:31scorchesure they do, however the ideas list is heavily considered when accepting orgs
02:21:35peanusthome: woah, i just instlalled on a 30gb ipod an hour ag
02:22:06scorchethe most common reason why an org was not accepted last year was a poorly fleshout out/developed ideas list
02:23:09Lloreanscorche: That seems odd. I mean, can't we be accepted but then receive no student allocation because of dumb ideas?
02:24:07scorchestudent allocation isnt related to the ideas list at all
02:24:54scorchealso see
02:25:01 Join Hawson_ [0] (
02:25:34 Quit Nico_P (Remote closed the connection)
02:26:24 Join notlistening [0] (
02:26:41thomcwell I'm trying to restore using the instructions on the IpodManualRestore wiki page, but my iPod only has one partition. Do I need to create a second?
02:27:11notlisteningOk is the rockboxplayer a waste of time?
02:27:30scorchenotlistening: that depends on what you want to get out of it
02:27:51 Quit perrikwp (" ajax IRC Client")
02:28:04notlisteningin thhe end a player that runs rockbox that people want
02:28:41 Quit Hawson (Read error: 60 (Operation timed out))
02:29:19Lloreanscorche: I've added several more ideas to the page.
02:29:22scorchenotlistening: then you would have to talk to "people"...for my view at least, i dont see any features or abilities over any of my existing device to motivate me to get one....of course i speak only for myself when i say this...
02:29:33Lloreanthomc: If you follow the instructions, it will end up with multiple partitions.
02:30:44thomcLlorean: well I've rewritten the MBR and copied over the firmware and got to the point where I'm instructed to format the second partition... but it's not there.
02:30:53thomcfdisk lists only one partition.
02:31:19notlisteningok scorche I understand your point it has to be what people want and at the moment it is not
02:31:39Lloreanthomc: Did you force your OS to re-read the MBR?
02:32:13Lloreannotlistening: At the moment, it's almost certainly going to be too expensive. The featureset he's offering can mostly be purchased as a refurbished player for less than USD $20
02:32:40notlisteningso what would make it something you want?
02:32:45 Quit Hawson_ (Read error: 60 (Operation timed out))
02:33:55thomcLlorean: I repeated the process and it's working now. Must've made a mistake. Thanks.
02:34:10scorchei think that question has been answered a few times in the earlier pages of the thread
02:34:39notlisteningok I will look again it take a long time my eyesight is not so good
02:34:47Lloreannotlistening: The first step (in my mind) would be to try to offer _all_ the functionality of Rockbox. So that when people are looking for a player, they see one that is obviously "better" than all the others, rather than one that's just "as good" but costs more.
02:35:05 Join Hawson_ [0] (
02:35:41Unhelpfulpixelma: yes, ondio sim. next/prev scroll the covers, up toggles between cover / track list, down brings up the menu (that seemed weird), and there appears to be no scrolling the track list :/
02:36:07Lloreannotlistening: the unfortunate truth is that this player is almost definitely going to cost more than any other player you can run Rockbox on, just because it's going to be new and produced in smaller quantities even in the best case. So it needs to stand out as something worth paying a premium for (in my opinion).
02:36:46notlisteningI agree
02:37:06pixelmaUnhelpful: similar to c200 then (but worse). Why doesn't it surprise me though? :\
02:37:20notlisteninghumm well back to the drawing board i think
02:37:54notlisteningis there a feature set of rockbox and also a future planned feature set?
02:38:30***Saving seen data "./dancer.seen"
02:38:36LloreanThere's not really a future planned feature set.
02:38:58notlisteninglol well i thought i would ask
02:39:00LloreanThe best way to stay "future" friendly is to pick a CPU a bit more powerful than you need so that things like more advanced video codecs or new audio codecs are likely to perform well
02:39:53notlisteningright and how powerful are we looking at?
02:40:10 Join Hawson [0] (
02:40:36scorcheas well, there are few devices that have, say, a digital out...some people like to have certain aspects for recording...others like very high capacity, still others might like other features..
02:41:32Lloreannotlistening: Well, it depends in part on whether you want to offer video (on the player's screen, or with an actual output to TV). If video isn't too important, the one already chosen is probably good enough. It will also probably handle the existing video playback fairly well.
02:42:10 Quit gregorovius ()
02:42:14LloreanThe most powerful Rockbox target at the moment (I believe) uses the i.MX31 which is 532mzh ARM1136JF-S
02:42:36 Quit Hawson_ (Read error: 60 (Operation timed out))
02:42:42 Part thomc ("Leaving")
02:42:49LloreanA lot of us like the idea of a new HD target, though honestly that's probably a niche and may not align with what the larger audience wants.
02:42:59LloreanHD as in "hard disk" not "high definitino"
02:43:13notlisteningDo you think it is wise to go for everything everyone wants?
02:43:36notlisteningyeah i got hdd as disk
02:43:58scorcheyou cant do everything everyone wants, as some liek small form factor, others like a lot of features
02:44:00LloreanI don't know if it's wise to go for _everything_ necessarily. It depends on how expensive it will end up, of course.
02:44:31notlisteningWell we are looking a making this for the people that want rockbox which in the main is you guys
02:44:50LloreanIf I were designing it, I would list every possible feature, then organize them by "importance" (as in, how much I thought 'everyone' wanted that feature) then try to guess how much each would add to the final cost, pick a target cost, then add them in from the "most" important, working down the list until I reached that cost.
02:44:59Unhelpfulpixelma: suggestions on a fix? define my own contexts for PLA rather than using combining contexts?
02:45:06notlisteninghumm it is very ahrd to know what to do
02:45:28Lloreannotlistening: I think you should make it for the users, not the developers. Many of us have several players anyway, and get new players to work on rather than to use.
02:45:30 Quit Lss ()
02:47:31 Quit Hawson (Read error: 60 (Operation timed out))
02:47:49notlisteningLlorean when we start talking about costs i and the other guys involved do not really have an idea of this
02:48:04 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
02:50:34notlisteningdo you think that the developers here even without hardware initially will want to work on it if we have a sensible feature / design list
02:50:51pixelmaUnhelpful: if you ask me - get rid of PLA there </semi-serious>. Combining contexts is the worst you can do, so maybe as a first step you could try if it's possible to use one context in one screen and a second in the other. If you mean with defining your own contexts to invent a whole new one, I'd object as this is entirely against what PLA should be about
02:51:29notlisteningIf we could get the hardware designed and a way of producting the prototype will they follow
02:51:56notlisteningbuild it and they will come - field of dreams like
02:52:20pixelmaUnhelpful: then it's not different to have your own button (or action) definition in the plugin but hidden away. Or do you mean own action definitions in pictureflow?
02:52:53Lloreannotlistening: If the hardware is _really_ interesting you might get a few developers to work on it (and as casainho's demonstrated, it doesn't necessarily take many to get things rolling). But it's not really something I'd gamble on necessarily
02:53:31 Quit Aurix_Lexico (Remote closed the connection)
02:53:47Unhelpfula context per screen, if there's a context that has all of the controls each screen needs, would be nice
02:54:06notlisteningand are there hardware guys who can help with the design as are skills are limited
02:54:39notlisteningSo don't bother
02:55:35Lloreannotlistening: Honestly, casainho's general idea isn't bad innately. It's just got some strange design choices that cripple it for being something you can attract people with (in my opinion)
02:55:52LloreanAs I said, the amount of RAM on it doesn't make sense for it being a flash-based player.
02:56:16notlisteningmaybe what we are trying to do is outside the scope of our ability
02:56:18LloreanThe vast majority of RAM is used for buffering from a hard disk. If there's no hard disk, an awful lot of it can go without causing a significant problem.
02:56:54notlisteningwekk that was something on the dev board but maybe he had more optuins if the choice of that
02:57:23LloreanYeah. My concern was just that when or if someone's looking around to try to get this produced, the people looking at producing it will thing that 64MB was chosen for a reason.
02:58:03notlisteningok, well this is something that we need to highlight to them
02:58:11LloreanIf you reduce the RAM to save cost, and pick a screen more modern (something 320x240 is pretty standard), and if the sound quality is good enough (I don't have personal experience with the codec he's chosen AFAIK) it doesn't actually sound bad.
02:58:26notlisteninglike most of the features on thatdev board they are not used at all
02:59:25notlisteningokay my concern was we are totally barking up the wrong tree
03:00:14LloreanNo, I wouldn't say that.
03:00:55notlisteningagain our mutual friend has chosen it because of the ease of development rather than its merits
03:01:17notlisteningand i understand why this is due to the lack ofany of our experience in the type of work
03:01:19LloreanI think one problem is that he's depending on the fact that it's "open" as a selling point.
03:02:41notlisteningyeah i agree it is what he is stressing and that might float a bit with googles efforts but the hardware must be as good as if not better
03:03:29notlisteningyou can have the best software ever running on a can of beans it is still only a can of beans
03:04:53LloreanIt's not _bad_ hardware or anything. It's just, it doesn't have anything to make me pick it over players that will certainly be cheaper and smaller.
03:05:06LloreanWell, the screen's kinda bad.
03:05:11LloreanBut everything else isn't really. :)
03:05:42LloreanIt will only really have an advantage if people _really_ don't want to buy refurbished, and if the Sansa AMS port either fails, or falls behind and a new generation of Sansas come out.
03:06:07JdGordonLlorean: why do you think the playlist insert thing is only useful for people who want to eventually save the playlist?
03:06:18LloreanI didn't actually say that.
03:06:48LloreanI said I think that for people who will save playlists, "Insert" is more useful than "Queue" because you're creating permanent playlists.
03:07:01LloreanFor people who won't save playlists, "Insert" is the same as "Queue" mostly, because in neither case are you keeping the list.
03:07:26LloreanBut I also said I'm not really too aware of the usual use cases for queue, in general.
03:08:01LloreanMy understanding of it is that it's for intentionally temporary things, so to me at least I wouldn't see someone queuing a particularly large number of songs. If you're going to queue more than just a small number, you may as well just insert them and not save the list.
03:08:05JdGordonI dont think having in mind that you might want to save it is relevant...
03:08:28LloreanSo the button should be used for Queue and not Insert?
03:08:49JdGordonyou're right that f you dont save there is no difference... bt you might want to save later and you know that you wouldnt want this track added
03:09:17JdGordonNO... my opinion is that choosing one f those options and hardcodng it is a bad idea
03:09:17LloreanMy point is about which single function is most appropriate. Do you think Queue is more valuable as a single-button function than Insert?
03:09:30peanusHALP (just a little) I was sent over to <>, by a link saying it had info on how to compile a lang file (i want Bork), but I can't find any info on how to compile one. is there another page with this info?
03:09:51LloreanJdGordon: Yes, but if that's the case, many of our hardcoded buttons are a bad idea. You're just trying to focus on one to get your foot in the door.
03:10:27JdGordonyou know i'm not...
03:10:53JdGordonespecially when we make a big dealabout this butto/action being the PLAYLIST ACTION
03:10:54LloreanJdGordon: "Some people might want other things" is exactly the same argument you could give for, say, making the Pitch shortcut keys configurable so that they can be used for other things.
03:11:21JdGordonyes it is.. but thats not the topic
03:11:30LloreanOr making the Volume buttons on targets with spare ones configurable, because you don't use them in the list when music isn't playing.
03:11:54LloreanNearly _every_ button we have, except ones explicitly labeled, can be argued as us "simply picking the function we thought would be most valuable to have."
03:12:08LloreanSo instead of arguing "this thing we've always done, we need to stop doing it" why not offer a single function that would be better on the button?
03:12:32 Quit peanus (Remote closed the connection)
03:13:46JdGordonI'm not suggesting all out customisable buttons... none is and only very few people want it... having a very limited subset of available actions for a button which is not in the "standard set" would be bloody useful and i think its a bad decsision that we dont do it
03:14:04JdGordonthe rec button in the list is one such button
03:15:06JdGordonand btw, i think the extra support argument is nonesense if its done correctly
03:15:35LloreanTell me, how do you do it in a way that doesn't require us to either ask additional questions, or explain how to verify what the setting is, before telling people a series of buttons to press to perform an action?
03:15:51LloreanWhat is this "done correctly"?
03:16:45JdGordonwell, for one thing if the maual doesnt usethat button as the standrd way of doing it... so it explain you need to get to the context menu to find the insert options instead of sayig "press rec to insert"
03:17:36JdGordonthen you can add "you can use the rec button to speed this up by choosing whch insert method to use, we dont recommend changing te value if you have problems "
03:17:47JdGordoni dont know.. something like that... its not hard
03:17:58LloreanHow is that not extra work for those providing support then?
03:18:16JdGordonhow is it extra work?
03:18:16LloreanIt's a bunch of extra stuff we wouldn't have to say if it's a hard-coded button.
03:18:33JdGordonits in the manual
03:18:39JdGordonyou say use defaults and rtfm
03:19:47LloreanWhat "limited subset" do you think is necessary then. Insert, Insert Shuffled, Queue, and Queue Shuffled?
03:20:00LloreanOr even simply "Insert" and "Queue"
03:20:24JdGordonand repeat last action...
03:20:54JdGordoni dont see why you cant give the full set..
03:20:59LloreanYou don't need the full set
03:21:15LloreanThere is no reason, ever, to need to repeat the other options more than once unless you don't pay attention and choose the wrong song.
03:21:31JdGordonits less work (code and support) to just give all the avilable option
03:22:16LloreanIf you're talking about adding in a new action anyway, I don't see why shortening the list is "hard"
03:22:28 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
03:22:57JdGordonwhy say you have the choice of these 2 options when you can just say you hae the choice of ay of the options
03:23:41LloreanIgnoring shuffle, those are the only two necessary to get the full explicit functionality sped up.
03:24:14JdGordonfine, whatever... shuffle shouldnt be ignored though
03:24:25LloreanI still don't think the button needs to be configurable.
03:26:01JdGordonI still see it as a completly pointless chang if its not
03:26:22LloreanSo it doesn't add ANYTHING that will help ANYBODY AT ALL if we don't make it configurable?
03:26:55JdGordonit add something to some unknown amount of users, instead of adding plenty to all users who use playlisting
03:28:23LloreanThis adds something to all users who use playlisting except those who want to save a playlist, haven't saved it yet, and want to add more songs into it that won't be saved.
03:28:34LloreanThat's a pretty select group.
03:29:42LloreanEven then, they would have to never, ever, ever create a playlist without the intent of saving it, and never ever create a playlist with the intent of saving it without also wanting to queue songs in it to not benefit from this.
03:29:46LloreanWhich narrows down the group even further.
03:29:48JdGordonalso those that want to add a track and dont care when its played
03:30:55LloreanAnd never create playlists the other ways.'
03:32:09JdGordonif this is supposed to be a shortcut, why cant i choose what I tihnk would be the best shortcut for me?
03:32:37JdGordonor more correctly, why should you choose for me which you think I would use most...
03:32:38LloreanYou could say the same thing about the "resume playback" button then.
03:33:11LloreanWhy can't it always go to the last plugin, or the FM radio?
03:33:27JdGordonhow? doesnt it already return you to the plast music screen? i.e wps or fm?
03:33:29LloreanI don't even know where you're getting calling it a "shortcut" from
03:33:33 Quit dfkt (Read error: 110 (Connection timed out))
03:33:35LloreanI'd simply call it "the playlist insert button"
03:33:41 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
03:34:03LloreanJdGordon: I said "always" and "plugin OR FM"...
03:34:03JdGordonit purpose is to speed up playlisting... how ca that not be called a shotcut?
03:34:25LloreanAre you claiming it won't speed up playlisting for a significant portion of users?
03:34:39 Quit tvelocity (Remote closed the connection)
03:35:00JdGordonyes, yes i am... it might speed up for some, but w have no idea how many users we have.. let alone how many use playlisting
03:35:03 Join serendipity_ [0] (n=serendip@
03:35:03Unhelpfulhrm, there's really no "generic directions + select/cancel" context, eh?
03:35:07LloreanYou're simply trying to use misguided semantics to prove a point. Calling it a shortcut doesn't make it any more or less reasonable that it be configurable.
03:35:07 Join kugel [0] (n=kugel@rockbox/developer/kugel)
03:35:51LloreanI could call the Menu button a "shortcut" to the menu. Should that be configurable? You can't deny it's faster than pressing left in the filetree until you get to root, then pressing left again.
03:35:58LloreanIt's definitely a shortcut.
03:36:21serendipity_hey all. i'm new to Rockbox. I want to install a plugin, but i cant find where to download them. For instance, has info on the TextEditor plugin, but no download link.
03:36:29kugelJdGordon: We're doing on DAPs, any action which wouldn't have something to do with music would be inappropriate. And As we cover all audio stuff already, playlisting is the only thing left anyway
03:36:29JdGordonshortcut can be used because its designed to be a quicker alternative to the usual way...
03:36:32Lloreanserendipity_: Rockbox comes with all its plugins.
03:36:50LloreanJdGordon: It's designed to be quicker alternative to the usual way of inserting songs. Your own definition.
03:37:02serendipity_all plugins listed in are already installed?
03:37:05 Quit _lifeless (Remote closed the connection)
03:37:10JdGordonLlorean: yes, thats what i said
03:37:14 Join _lifeless [0] (n=lifeless@
03:37:19Lloreanserendipity_: If they're appropriate to, and working on, your player, yes.
03:37:44LloreanJdGordon: So the menu is a shortcut, but shouldn't be configurable. The insert button is a shortcut, but should. Because it's a shortcut. Even though Menu shouldn't?\
03:37:46kugelJdGordon: You call it shortcut, I call it "making use of a unused button, which actually makes a *core
03:37:47Lloreanthat makes no sense.
03:37:53kugel* feature usable without pain
03:38:04JdGordonthe menu isnt a shortcut... how are you supposed to get to it otherwise?
03:38:16LloreanJdGordon: From the filetree, press left until root, then left once more.
03:38:29JdGordonkugel: have i ever said im against the button usage?
03:38:33LloreanThe menu button is entirely unnecessary.
03:38:36LloreanIt's just a shortcut.
03:39:06 Join plutonian [0] (
03:39:13Lloreanserendipity_: For many/most players it lists on the index whether your player has it or not
03:39:20plutoniandoes the AAFreak theme work with rockbox v3.1?
03:39:31kugelJdGordon: No, but you don't agree about making a core feature usable. You rather discuss about complicating things, which wouldn't be there for a good deal of our targets anyway
03:39:42JdGordonLlorean: is the only way for me to get you to actualy argue the pointis to se hebrew so you cant argue the meaning of the word?
03:39:44serendipity_this is great
03:39:54Unhelpfulplutonian: we don't make or maintain the themes, did you try it?
03:40:05plutonianit doesn't say anything about a patched build but it does mention that "Without Viewports, this theme wouldn't have been possible."
03:40:05LloreanJdGordon: You're the one who was saying "it's a shortcut" is the reason it should be configurable. I was merely pointing out it's not the only shortcut.
03:40:14kugeleven if it was a shortcut, we actually cannot choose a non-redundant action, as the button isn't free on many targets
03:40:17plutonianare viewports in the official builds or are they a patch?
03:40:19JdGordonkugel: no.. im arguing to make it useable for everyone... you lot are sugesting only a unknmown subset should benefit
03:40:29Unhelpfulviewports are an official feature for some time now
03:40:34LloreanJdGordon: If you don't think "being a shortcut" is a reason, feel free to retract that. Otherwise, feel free to explain how the menu button isn't a shortcut, when it does exactly what you defined a shortcut as doing.
03:40:42kugelAnd I still think we don't need configurability at all. Playlists is a core feature, so it just fits it's purpse
03:40:57Lloreankugel: He wants it configurable for all Insert/Queue options, rather than just one.
03:41:07 Join kevin__ [0] (n=chatzill@
03:41:24kugelthis doesn't make any sense
03:41:35kugelvirtually any of those is covered by insert
03:41:41LloreanQueue isn't really
03:41:46LloreanNor are the two "Shuffled" variants
03:42:06kugelyes, shuffled not. But for queue, as you said, you just don't save the playlist, and are done
03:42:10LloreanBut "Queue" only matters if you have an unsaved playlist with songs inserted, and don't wish to save it before adding songs that you don't wish to have saved in it.
03:42:24kugelqueue is something I never understood the use for anyway
03:42:31LloreanFor shuffled, you should be able to turn shuffle off and back on again to reshuffle your playlist anyway.
03:43:04LloreanIt's not _perfect_ I'll admit, but it's about equally fast to going into the menu, reconfiguring what the one-button-press does, coming back, and then insert-shuffling everything
03:43:15LloreanAt least, if you have "Shuffle" on the quickscreen
03:43:20kugelWith insert, you can do almost anything and still need vastly less button presses for
03:43:21Unhelpfulkugel: "don't save the playlist" doesn't cover the case where you want to have some songs that will be saved with the playlist, and some not... though i can't really say why you'd want to do that, that seems to be what queue is for.
03:43:38kugelUnhelpful: I never used queue, actually
03:43:44LloreanI've used queue.
03:44:06LloreanBut usually if I'm listening to a short, repeating playlist and just want to hear a single song on this cycle, and not in future repeats.
03:44:33kugelwell, once or twice, to see what it does. But after I saw what it does, I also saw that I won't ever need it again
03:44:45plutoniansweet, it does work
03:44:59LloreanI would bet money that, on average, among our userbase, "Insert" would be the most used action if all the playlist actions had an individual button each and people knew the full use of each of them. Since it's impossible to test though, that statement means nothing.
03:45:03plutonianthe wiki article on viewports seemed to imply that it wasn't in the main build
03:45:20kugelinsert is the most flexible one
03:45:30LloreanBy far.
03:45:35JdGordoninsert is broken...
03:45:43JdGordoninserting multiple anyway
03:45:46LloreanJdGordon: Broken how?
03:46:05LloreanAnd the feature should be considered off of a non-broken insert anyway.
03:46:17JdGordonyou can get playlists like 1ABC23 (assuming 123 and ABC re in 2 seprate dirs)
03:46:48LloreanSo it doesn't put the pointer at the end of the list after doing an "insert" on a folder?
03:46:55kugelof course you can. play 1, insert ABC at once, then insert 2 and 3 seperately
03:47:06plutonianwhat does "Data abort at 000098BC (0)" mean?
03:47:17Lloreankugel: No, he means "If I insert the folder ABC then 123, I get A123BC instead of ABC123
03:47:24LloreanWhich is broken if he's not using "Insert Next"
03:47:29JdGordonLlorean: yeah, there is a long tracker task for that wth an argument if its actually broken or not
03:47:42kugelThat's a bug then, which I've never experienced
03:47:46LloreanIt's very broken if that's what it does right now.
03:47:55Llorean"Insert Next" is what you'd use to get the situation you've got.
03:48:42Llorean"Insert" should always insert *after* the last song of whatever you last inserted. That's not only the most flexible way, but the fact that it doesn't do it right now makes it impossible to get certain playlists without inserting things in reverse order.
03:49:04JdGordonalthough.. fine.. i do see the argumet that only insert queue and shuffled (Q and I) is needed and not the whole lot
03:49:45kugeland I still think that shuffle is mostly covered too, by taking a quick tour into the quickscreen
03:50:03JdGordonthat ASSumes you still have shuffle on the qs
03:50:11kugelor just hit shuffle playlist, once after you're done adding 100s of songs
03:50:34Lloreankugel: There is the case where you don't want the songs you've already heard to be included in the shuffle, at least. So there is one specific case where it's true that "Shuffled" does something you can't accomplish otherwise.
03:50:55kugelyea, I said mostly. Sure there's edge cases
03:51:00LloreanJdGordon: Even going through the menus to find shuffle is going to take approximately as long as going through the menus to change the "Insert" setting to "Shuffled"
03:51:28LloreanSo even if you don't assume it's in the QS, it's still a time saver vs the old method of having to go to the context menu for every item.
03:51:41JdGordon? thats why im saying have the button there.. if you know you are usually going to want to insert shuffled have that set to the button
03:52:20JdGordonyour making the assumption that most people will only use regular insert.. im saying make it possibl to make everyonehappy
03:53:08LloreanJdGordon: Do you think we should have the "resume playback" button configurable to mpegplayer, midiplayer, FM radio, and WPS?
03:53:23JdGordonno, and thats irrelevat anyway
03:53:30kugelYou make everyone happy, I make 98% (the other 2% are the ones who don't want the songs they heard shuffled too) happy, but without adding unnecessary complexity
03:53:35 Quit timc (Connection timed out)
03:53:43LloreanJdGordon: It's more or less the same situation, so it's not irrelevant. The arguments you used here apply exactly to that one as well.
03:53:50LloreanJdGordon: Why is this one case so "special" to you whereas you don't care about that one at all?
03:53:51JdGordonkugel: thats a stupid number... i can show you the same statement in reverse
03:54:01kugelI don't think so
03:54:13 Join timc [0] (n=aoeu@
03:54:13BUMBACL0Tany status on 6th gen
03:54:13JdGordonits not the same at all... why cant you tke each on seperatly?
03:54:14 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
03:54:25LloreanBUMBACL0T: No. Please don't ask. The forums will have updates if it ever changes.
03:54:31krazykitBUMBACL0T, no, and there won't be unless someone actually takes it upon themselves to do the work
03:54:32JdGordonkugel: you've got stats to support your 98% claim?
03:54:33LloreanJdGordon: the same arguments apply.
03:54:52JdGordonLlorean: yes, but we arnt talking about changing the play button...
03:54:56LloreanJdGordon: Your argument is based _entirely_ on "we could make more people happy if we did this." There are a million things we could change for that reason.
03:55:15LloreanWhy is this one one that we _should_ change, while those are ones we _shouldn't_ exactly?
03:55:34saratogajhMikeS: seen dreamlayers DMA audio playback patch for PP ?
03:55:40JdGordonI'm not saying we shouldt.... I am saying that we art talking about it... im saying it doesnt have to be all or none
03:56:06LloreanJdGordon: So your real position is "we should make many buttons configurable, but I'm only arguing one at a time"?
03:56:34LloreanMy point is that this is a bigger issue than "just one button" and trying to pass it off as such is silly, because it's unrealistic to expect it won't then come up for several other buttons.
03:56:36 Join tarbo [0] (n=me@unaffiliated/tarbo)
03:58:00CtcpIgnored 4 channel CTCP requests in 0 seconds at the last flood
03:58:00*JdGordon gives up and just continues to outright reject the patch as it is
03:58:46LloreanYou reject it because we won't make a change to it that we won't make elsewhere in the source?
03:59:34JdGordonno, i reject it because its a stupid addition and its my right to object to patches i feel are useless... weather that counts for anything is another question
03:59:59LloreanWhat is a better addition that does not require a configurable button, then?
04:00:19LloreanIf your rejection has nothing to do with configuration, suggest something that is actually better but non-configurable.
04:00:48JdGordonim not sayin there is a better suggestion that doesnt give configurability
04:01:00LloreanSo it's the best suggestion that's non-configurable.
04:01:28JdGordonno, im saying the patch as it stands is stupid and pointless
04:01:45JdGordonand with minor tweaks wold be very usefull to everyone
04:01:47LloreanStupid and pointless, in your opinion, but there's nothing better without configurability?
04:02:11LloreanI mean seriously, when you use words like "stupid and pointless" on a patch you're even willing to admit will be pretty useful, you just sound bitter.
04:02:30 Quit saratoga ("CGI:IRC (EOF)")
04:02:39JdGordoni cant iagine why i ound bitter... maybe going round in circles on irrelevancies did that to me
04:02:47LloreanBut if you can't suggest something better without requiring a configurable button, then even if it's "bad", if it's the best option within those bounds and it's still what we should do, since by definition the other options are worse.
04:03:14*kugel agrees
04:03:19JdGordonwe already have plenty of configurability in rockbox
04:03:34JdGordonshould we remove the volume setting because surely -6is good for everyone?
04:03:34LloreanJdGordon: You're the one who's going around in circles. You keep trying to "prove" your point with arguments that apply to dozens of other things, but you say they don't need changed why this does. And yet you've yet to explain what it is about this one thing that makes the arguments apply to it, but not anything else.
04:03:51JdGordoni have never said they dont need changes
04:04:03JdGordonall i have said is WE ARNT TALKING ABOUT THEM
04:04:32LloreanJdGordon: And I'm asking you why they aren't allowed to be talked about
04:04:43LloreanIf your arguments are true, you should be arguing for a change in the configurable button policy in general
04:05:00LloreanIf that's the case, it has nothing to do with his patch, but rather ALL buttons that could be configurable based on your reasons why they should be.
04:05:00JdGordonim not alone in my opinion of that policy
04:05:27Lloreanyes, but you're trying to get that policy changed by avoiding arguing the policy, and calling a good patch "stupid" over and over in the hopes that you can get your toe in the door
04:05:46LloreanIf you want to get the policy changed, argue the policy itself. Don't try to sneak in changes.
04:06:27notlisteningTx Llorean good night and take not prisoners ;)
04:06:38notlistening*no :P
04:06:51 Join blkhawk- [0] (
04:07:03kugelnotlistening: hey, I just wanted to add that I agree that 64 are way more than needed.
04:07:18kugelI think 8MB should plenty already.
04:07:46kugelespecially if you can get a non-crappy lcd for that
04:07:46JdGordonLlorean: besides I'm not sure why you get to call it a polisy when there is over 50 different settings
04:07:57 Quit tarbo_ (Connection timed out)
04:08:01notlisteningthanks kugel
04:08:14JdGordonincluding one to change the star screen which actually can cause support headaches
04:08:19JdGordonand the quickscren
04:08:21notlisteningu mean for the cost difference
04:09:02LloreanJdGordon: So now your statement is "a configurable button is exactly the same as a configurable anything else"? Like the volume example you gave earlier?
04:09:06kugelJdGordon: You're basically objecting to the button policy, but yet you keep telling "we aren't talking about the other buttons", that doesn't make any other sense than trying to sneak in your toes
04:09:30kugelnotlistening: yes
04:09:42LloreanJdGordon: All I'm saying is, if you want to get the button policy, argue the button policy. For this patch, when discussing it, we don't do configurable buttons so discuss this patch on the assumption that configurable buttons aren't an option.
04:09:51LloreanOnce they are an option, it can always be re-visited.
04:10:01kugelwell, I doubt the RAM will yield much savings nowadays, but still. 64MB is basically overkill.
04:10:11kugelfor flash based players, that is
04:10:33notlisteningI think maybe we need a this is what we are using and this is what we are aiming for page to clear up any myths about ethernet support etc etc
04:11:32notlistening64 gigs of ram etc :D
04:11:57notlisteningnight all
04:12:30kugelnotlistening: If you want to sell the thing, you need to pay attention on the selling points. As it has been said, you won't get rich ever just by saying it's an open design. And you won't attract much more people with an overdose of RAM either. But a decent LCD, that will get you some attention for sure
04:13:48serendipity_oh man, rockbox is so much cooler than ipodlunix
04:14:12serendipity_you all rock
04:14:32*serendipity_ just got it installed, am playing with it
04:14:41notlisteningbattery lcd and flexibility it storage options are my key features and that it lasts
04:14:53 Join jordoex [0] (
04:16:41 Quit notlistening ("Leaving")
04:17:05 Quit intrados_ (Connection reset by peer)
04:18:13 Join intrados_ [0] (
04:22:54 Quit Llorean (Read error: 104 (Connection reset by peer))
04:23:08 Join Llorean [0] (
04:23:20 Quit blkhawk (Read error: 113 (No route to host))
04:23:49 Nick blkhawk- is now known as blkhawk (
04:24:05 Join Barahir_ [0] (
04:24:11 Join planetbeing_ [0] (n=planetbe@
04:24:23 Join AndyI [0] (i=AndyI@
04:30:13 Join yhuang [0] (n=yhuang@unaffiliated/yhuang)
04:31:32 Quit serendipity_ (Read error: 54 (Connection reset by peer))
04:32:14 Join Strife89 [0] (n=nds@
04:34:17kevin__hey, I'm having some problems unbricking my sansa c250
04:34:29kevin__I've followed the instructions here:
04:35:03kevin__but every time the c250 resets, I end up with the same problem (only blue led is on)
04:35:33 Quit AndyIL (Read error: 110 (Connection timed out))
04:35:47Strife89I thought there was a c200 page now.
04:36:25 Quit dfkt (Read error: 60 (Operation timed out))
04:36:38kevin__hm I didn't find one
04:36:45kevin__I'll try looking again
04:36:49Strife89c200tools, at the least.
04:38:22kevin__Sorry, but I can't seem to find anything ):
04:38:32***Saving seen data "./dancer.seen"
04:38:42Strife89Well, nuts.
04:39:01 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner]")
04:39:12Strife89I have no browser on me, so I can't look....
04:40:30 Quit Barahir (Read error: 110 (Connection timed out))
04:41:41kevin__hm do you know what might be the problem though?
04:41:49kevin__everything seems to work fine
04:42:00kevin__I've been able to get into recovery mode using e200tool
04:42:05kevin__and I can see the 16mb drive
04:42:33kevin__but after copying mi4 and disconnecting, nothing seems to change
04:42:39Strife89There are many different versions of the OF.
04:43:04Strife89You might be using the wrong version .mi4.
04:43:11kevin__oh ok
04:43:37Strife89Double-check the page you got it from.
04:44:09Strife89Also, aren't you supposed to copy a certian text file as well?
04:44:20Strife89I can't remember.
04:44:25kevin__there's only 1 c200 file
04:44:52Strife89That might not be the page you need.
04:45:05Strife89There was another...
04:45:38Strife89I can't look unless I get off IRC and reboot my DS.
04:45:47kevin__should I try the c200erase instructions?
04:46:01Strife89Not just yet.
04:46:26Strife89Will you wait here while I go look?'
04:46:37kevin__btw I really appreciate the help
04:47:22Strife89Not a problem. :)
04:48:16 Quit yhuang ("Leaving")
04:48:31 Join yhuang [0] (n=yhuang@unaffiliated/yhuang)
04:48:33 Quit dfkt_dt (Read error: 110 (Connection timed out))
04:49:48Strife89kevin__: I'll go once I've settled a conversation with scorche.
04:50:07Strife89Oops, I meant soap.
04:52:31 Quit miepchen^schlaf (Read error: 101 (Network is unreachable))
04:54:36 Join serendipity_ [0] (n=serendip@
04:54:56 Quit Strife89 ("Gone for research.")
04:57:25 Quit rocko ("Leaving")
04:59:11 Quit serendipity_ (Client Quit)
05:01:25 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
05:02:18 Join Strife89 [0] (n=nds@
05:02:44Strife89kevin__: Found it.
05:04:27Strife89Scroll down to Recovery Mode and carefully follow what you can.
05:05:40 Part martian67 ("Leaving")
05:07:25kevin__well I can't get into recovery mode normally :/
05:07:54Strife89But you can with the e200 tool?
05:08:42Strife89Hmmm... you might have to mix'n'match a little...
05:09:26Strife89Also, I think this is a page:
05:09:40 Join rocko [0] (
05:10:04kevin__doesn't work ):
05:10:19 Join Dauron [0] (
05:10:24Strife89Hmmm, darn.
05:10:30 Quit _Auron_ (Read error: 113 (No route to host))
05:10:32 Nick Dauron is now known as _Auron_ (
05:10:46Strife89I can't remember its name, then.
05:12:56Strife89Can you look at the list of recently changed Wiki pages?
05:13:25Strife89Do you know what the link for that is?
05:13:57kevin__i don't think so
05:13:59 Quit plutonian ("leaving")
05:14:49kevin__oh, thanks
05:15:07Strife89cool_walking_: Saves me some tapping, thanks. :)
05:15:47Strife89Look for an entry with "tools" in the name....
05:16:09kevin__there's UsefulTools
05:16:14Strife89I'm quite certain it was updated...
05:16:40Strife89UsefulTools might help, but I'm not certian.
05:17:01Strife89Damn, I wish I had my laptop on me....
05:18:12kevin__only other ones with tools are DesktopTools and DevelopmentTools
05:18:21kevin__but those were updated in 2008
05:18:32Strife89Those aren't it.
05:19:19Strife89There was another page... I can't remember the title....
05:19:28cool_walking_What are you looking for?
05:19:58Strife89The page of c200 recovery tools.
05:20:15Strife89Such as the one gevaerts coded.
05:21:06kevin__well I found an irc log with a similar problem to mine
05:21:31kevin__and there were links to a c250repair tool
05:21:37kevin__but the links are dead now :/
05:21:43Strife89I can't search by hand unless I reboot the DS, load the browser, find it, and reboot again to reconnect to IRC.
05:21:56cool_walking_ links to e200tool
05:22:26Strife89He used that, I believe,
05:22:37Strife89with minimal success.
05:23:11cool_walking_oh, I should read more
05:24:20Strife89I badly wish I had the DS Opera browser kit.
05:24:24 Quit Seed ("cu, Andre")
05:24:24 Quit Horscht ("Verlassend")
05:25:44Strife89Or that my Palm had a WiFi radio.
05:27:15Strife89... I'm sorry, kevin__, but I need some sleep.
05:27:37kevin__do you think i should try the erase tool :/
05:27:47Strife89I'll check in sometime tomorrow.
05:28:16Strife89kevin__: If you can't find anything else, it's worth a shot.
05:28:31Strife89Just have some firmware handy.
05:29:11Strife89But anyway, good luck, and good night. :)
05:29:21Strife89No problem. :)
05:29:52Strife89See you later.
05:30:01 Quit Strife89 ("ClIRC - IRC client for Nintendo DS")
05:31:47 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
05:31:49 Quit kadoban (Remote closed the connection)
05:32:13 Join Shlee [0] (
05:35:04 Join midijunkie [0] (
05:35:17 Quit dfkt (Read error: 110 (Connection timed out))
05:36:42 Join dfkt [0] (n=dfkt@unaffiliated/dfkt)
05:37:51 Join kadoban [0] (
05:39:15 Part Shlee
05:41:42 Quit midijunkie (Read error: 104 (Connection reset by peer))
05:44:24 Join dfkt_dt [0] (i=dfkt@unaffiliated/dfkt)
05:45:27 Quit rocko ("Leaving")
05:55:00 Quit FlynDice (Remote closed the connection)
05:58:28 Join FlynDice [0] (
06:04:24 Quit dfkt (Read error: 110 (Connection timed out))
06:08:05Unhelpfulhrm, how hard would it be to have a custom-drawn list like the pictureflow track list "work like" the core file or DB browser? ie, use the same buttons for scrolling, select/exit, etc?
06:09:55LloreanWhat's the problem with doing that?
06:11:01Unhelpfulonly that i have no idea how to do it :)
06:12:14Unhelpfulit seems that the PF track list would be easiest to use if it had the same button map as the core browsers, with the back/cancel button returning you to the album view.
06:12:16LloreanCan you use core contexts in plugins?
06:12:42Unhelpfulthe album view, if it's going to keep using PLA, should probably only be using generic_left_right_fire
06:13:57Unhelpfulugh, that leaves it without a menu key. i can see some dislike PLA so much...
06:14:12LloreanPLA only works well in _really_ simple situations.
06:15:19Unhelpfulit really seems like *every* context ought to include a menu key.
06:16:53 Quit FlynDice (Remote closed the connection)
06:17:24 Quit dfkt_dt (Read error: 110 (Connection timed out))
06:17:31LloreanI kinda think that there should be one standard context for "up, down, left, right, select, menu" with the option to either use that and customize the other keys, or go all-out customization.
06:17:40LloreanBasically, make the job "easier" in a general sense, but don't get all complex.
06:17:56Unhelpfulperhaps i ought to avoid PLA entirely, and just use the rb->get_custom_action that it calls :)
06:18:22 Join plutonian [0] (
06:18:42plutonianis it possible to get rockbox to work on an ext3 filesystem?
06:19:02plutonianor anything other than fat32 for that matter?
06:19:05 Join FlynDice [0] (
06:19:28 Quit kevin__ (Remote closed the connection)
06:20:20 Quit FlynDice (Remote closed the connection)
06:21:56Unhelpfuli believe FAT16 support can be enabled at compile time, although it's not in standard builds. why would you want something else on your DAP, anyway? FAT is widely supported, simple to implement, and not a large drain on CPU.
06:22:50plutonianfat32 doesn't support filesystem labels with no capital letters
06:23:11plutoniannot a real issue by any stretch of the imagination, but it strikes me as silly
06:24:05 Join midgey [0] (n=tjross@
06:24:32 Join FlynDice [0] (
06:25:36plutonianit's also sort of annoying to not be able to have :'s and ?'s in filenames
06:26:10plutonianI don't like to have to look at Batboy_ The Musical
06:26:11 Join dfkt_dt [0] (i=dfkt@unaffiliated/dfkt)
06:26:56plutonianor Crisis_ What Crisis_
06:27:23 Quit FlynDice (Remote closed the connection)
06:27:28 Quit jfc (Read error: 104 (Connection reset by peer))
06:27:42 Join jfc [0] (
06:28:10Unhelpfuli have tracks with / in their metadata, but i'm not too keen on implementing some new FS that will allow absolutely anything but NULL. you could use the database...
06:29:39 Join FlynDice [0] (
06:34:50 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
06:34:53 Join nataku92 [0] (n=chatzill@
06:35:09 Quit nataku92 (Client Quit)
06:35:19 Join kevin__ [0] (n=chatzill@
06:37:32UnhelpfulLlorean: it looks like i'd just call rb->get_action to use any core context i want. i don't think there's one that will do for the album list, is there?
06:38:33***Saving seen data "./dancer.seen"
06:40:23LloreanUnhelpful: Don't you just want to duplicate the context the normal list uses?
06:43:10Unhelpfulfor the track list, yes. for the album list, we want horizontal scrolling, select, quit, and menu actions.
06:44:45LloreanSorry, got mixed up.
06:44:56LloreanYeah, I can't imagine one for the album list in core, no.
06:46:10 Quit planetbeing_ ()
06:46:13Unhelpfuli can see the WPS context making sense, possibly, with the seek/skip buttons mapped to scrolling... but that wouldn't be right at all on the e200, where clearly one ought to always scroll with the wheel.
06:48:50 Quit plutonian ("Lost terminal")
06:51:05 Join rocko [0] (
06:53:04 Quit kachna|lappy (No route to host)
06:55:02 Quit yhuang ("Leaving")
06:59:23Unhelpfulit seems like PLA *might* be ablet to handle the album list, if there's some way i can rely upon keys defined in the left/right/fire mapping taking precedence over the generic actions one...
07:06:25 Join planetbeing_ [0] (n=planetbe@
07:06:52 Quit dfkt (Read error: 110 (Connection timed out))
07:07:11 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
07:13:17 Quit daurnimator ("Cyas later...")
07:16:49 Join daurn [0] (n=daurnima@unaffiliated/daurnimator)
07:16:51 Join daurnimator [0] (
07:32:10 Quit agaffney (Read error: 104 (Connection reset by peer))
07:33:54 Join agaffney [0] (n=agaffney@gentoo/developer/agaffney)
07:41:30 Quit CaptainKewl (Read error: 110 (Connection timed out))
07:48:46 Join ender` [0] (
07:49:42 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
07:53:09 Quit planetbeing_ ()
07:55:43 Join kachna [0] (
08:02:33*amiconn is building for recorder with pf-hwcodec patch
08:05:55 Quit Darksair ("Do you hear that? This is the sound of inevitability. This is the sound of your death, Mr. Anderson.")
08:13:10 Join planetbeing_ [0] (n=planetbe@
08:13:21 Quit bmbl (Read error: 104 (Connection reset by peer))
08:14:57 Quit kevin__ (Read error: 104 (Connection reset by peer))
08:15:22 Join yhuang [0] (n=yhuang@unaffiliated/yhuang)
08:22:40 Join tvelocity [0] (
08:27:36 Quit kkurbjun ("Leaving.")
08:29:08 Quit dfkt_dt (Read error: 110 (Connection timed out))
08:29:27 Join dfkt_dt [0] (i=dfkt@unaffiliated/dfkt)
08:30:40 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
08:33:19Unhelpfulamiconn: actually, it worked on the sim, it's plugin-only, and i don't think anything further can be done to test it except on targets, anyway, so it's also r20213 :)
08:34:32amiconnUnhelpful: It's working :) Some remarks: (1) The startup screen looks funny, as it is black-on white with the logo looking inverted. Also the progress bar is huge. (2) Creating the album art cache took ~15 minutes. Not too bad when taking into account that it also takes a few minutes e.g. on the irivers.
08:35:10Unhelpfulrecorder have the same funky control issues as ondio sp sim?
08:35:17amiconn(3) 4 fps while scrolling around. 3 fps with steady display
08:35:33Unhelpfulahahahah... that's pretty awful :)
08:35:33rasherHow does that look?
08:36:09 Quit perrikwp (" ajax IRC Client")
08:36:17amiconnNo, the controls are almost like I expect them to be. Only leaving the tracklist isn't
08:36:30amiconnThe tracklist achieves ~6 fps
08:37:34amiconnActually the tracklist fps depends on the track names
08:37:54 Join kevin__ [0] (n=chatzill@
08:38:13*Unhelpful is proud of having done so much for something so useless :)
08:38:36***Saving seen data "./dancer.seen"
08:40:08amiconnThe covers are quite recognisable (my recorder has the white backlight mod).
08:41:15*amiconn needs to think a bit about full screen animation on this thing
08:41:43Unhelpfulin what way?
08:42:11amiconnThe greylib itself achieves 36fps, so that's not the limiting factor
08:42:45Unhelpfulwell, there are a few fmul and fdiv calls per column rendered
08:43:41amiconnI know. Perhaps some could be reduced to 16*16->32. That is two instrcutions instead of a subroutine
08:44:42Unhelpfulgiven what you've said about shifts on sh1, you might also consider changing PFREAL_SHIFT to 8 (it's 10 now)... but that means redoing the sine table, as well.
08:44:49 Join petur [50] (n=petur@rockbox/developer/petur)
08:45:26amiconnBut even at 4fps, pf is already usable. It would probably be better to reduce the slowdown-when-achieving-centre if the fps are low
08:45:37amiconnAre these signed or unsigned shifts?
08:46:37Unhelpfulwell, PFreal is defined as a signed type, but the values in question are going to be unsigned in many of the right-shift cases
08:47:36Unhelpfulwould a ((some unsigned type)val) >> shift suffice to make the assembler use logical shifts without adding any other nastiness?
08:48:01amiconnI would think so
08:48:57amiconn>>10 for signed is 10 instructions on SH1 (and probably using a subroutine with those instructions), for signed it's two instructions (shlr8 + shlr2)
08:49:27amiconnEh, for *unsigned* it's two instructions
08:50:03Unhelpfulso making sure the shift is unsigned is probably worth more than changing the shift amount
08:50:56 Quit cool_walking_ (" ajax IRC Client")
08:51:59Unhelpfulyou might also want to look at fdiv. i experimented with a function that would shift left as far as possible and return the amount shifted, instead of using clz and shifting by that amount. that might be a win vs clz calculation followed by a variable shift.
08:52:18 Quit BHSPitMonkey ("Ex-Chat")
08:54:27 Join Rob2222 [0] (
08:55:06Unhelpful... yes, i'm pretty sure all of the right shifts can be forced to unsigned, in the loop that fetches the actual pixels.
08:58:49 Join jordoex_ [0] (
09:02:32amiconnAre you referring to the big loop in render_slide()?
09:07:12Unhelpfulthe while loops betwen lines 1488 and 1541
09:07:17 Quit jordoex (Nick collision from services.)
09:07:56amiconnI tried it, and it does indeed get faster. Now it almost always shows 4fps (once the cache is loaded), and sometimes 5fps
09:08:07Unhelpfulthe divide by PFREAL_ONE at 1473 should also be of a positive value
09:08:46 Nick jordoex_ is now known as jordoex (
09:09:45amiconnIt's quite noticeable when scrolling around. Line 1473 will have much less an effect (only once per column instead of every pixel)
09:11:52amiconnUnhelpful: Do the alpha cases occur often (the reflection stuff)?
09:12:22amiconnI mean the last of the 3 while loops in the if() and the else block
09:12:31 Quit faemir (Read error: 110 (Connection timed out))
09:12:50 Quit Rob2223 (Read error: 110 (Connection timed out))
09:12:51 Nick Barahir_ is now known as Barahir (
09:13:18amiconnThese are some good candidates for multiplication optimisation (especially fade_color())
09:13:47Unhelpfulthose alpha values serve different purposes, actually. the one in the if is passed to render_slide, and is used when fading whole slides in and out, as for the outermost slides, and the fade when transitioning to tracklist
09:14:35 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
09:14:50 Quit midgey ()
09:14:59Unhelpfulthe one in the last while loop is done in etiher case, and handles reflection fading. the reflection height is about a third of the screen height, less for columns rendered further away.
09:15:01amiconnWell, the "colours" passed to fade_color are actually 8 bit values, as is the alpha (well, alpha can be 256), correct?
09:15:34*amiconn adds casts to (unsigned short) in fade_color()
09:17:58 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
09:18:09Unhelpfulwell, the alphas in the reflection table are always <256 in practice, and the passed alpha is always <256 when it's applied, since 256 is special-cased to avoid the multiplication entirely... so i think you could get away with treating the value as 0-255, if there are any relevant optimizations, such as an unsigned 8x8->16 multiply?
09:18:50amiconnSuch does not exist on SH1, but 16*16->32 does (both signed and unsigned)
09:19:08amiconnDoesn't seem to help atm - will do further tests tonight
09:19:12Unhelpful a 16*16->16 would suffice if it exists and is cheaper
09:19:37amiconn16*16->32 is the only native multiplication instruction on SH1
09:19:51 Quit kachna (Read error: 148 (No route to host))
09:20:31amiconnIt also has mac (won't help in pf though). 16*16->32+42->42
09:20:59amiconnThe 42 is not a typo
09:25:35 Quit BHSPitMonkey ("Ex-Chat")
09:29:01 Join Thundercloud [0] (n=thunderc@
09:34:36 Quit agaffney (Read error: 110 (Connection timed out))
09:37:17 Join Lss [0] (
09:43:41 Quit z35 ("Leaving")
09:45:09 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
09:45:21 Quit _lifeless (Remote closed the connection)
09:45:32 Join _lifeless [0] (n=lifeless@
09:57:06 Join kevin____ [0] (
09:57:18 Join agaffney [0] (n=agaffney@gentoo/developer/agaffney)
09:59:38 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
10:03:38 Quit Thundercloud (Remote closed the connection)
10:09:25 Quit dfkt (Read error: 110 (Connection timed out))
10:09:46 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
10:11:46 Join sarixe [0] (
10:13:37 Quit kevin__ (Read error: 110 (Connection timed out))
10:18:09 Quit rasher (Read error: 104 (Connection reset by peer))
10:19:47 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
10:21:12 Join rasher [0] (
10:22:08 Quit BUMBACL0T (Excess Flood)
10:22:17 Quit agaffney (Read error: 104 (Connection reset by peer))
10:22:26 Join BUMBACL0T [0] (n=ORF@unaffiliated/bumbacl0t)
10:23:54 Join agaffney [0] (n=agaffney@gentoo/developer/agaffney)
10:32:11 Quit planetbeing_ ()
10:36:13 Quit dfkt (Read error: 110 (Connection timed out))
10:38:39***Saving seen data "./dancer.seen"
10:39:04 Join dfkt [0] (n=dfkt@unaffiliated/dfkt)
10:39:15 Quit Lss ()
10:50:25 Quit dfkt_dt (Read error: 110 (Connection timed out))
10:54:51 Join dfkt_dt [0] (i=dfkt@unaffiliated/dfkt)
10:58:04 Join pyro_maniac [0] (
10:58:48 Quit nuonguy ("This computer has gone to sleep")
11:00:32 Join planetbeing_ [0] (n=planetbe@
11:04:27 Quit dfkt (Read error: 110 (Connection timed out))
11:07:11 Join einhirn [0] (
11:09:58pixelmarasher: for responses of cabbiev2 (port) makers you could also try posting in the "ongoing default theme contest" thread, some of them are still quite active in the forums and notice
11:10:23 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
11:11:15rasherAnd are we sure we *do* want to relicense to cc-by-sa-3?
11:13:21pixelmano, I'm not. Though I don't see what could be wrong with it when themes you want to upload to the theme site would require it. I don't know a lot about licensing so I'm probably not the right one to asl
11:13:22 Quit dfkt_dt (Read error: 110 (Connection timed out))
11:13:52 Quit planetbeing_ ()
11:14:17 Quit kevin____ (Read error: 110 (Connection timed out))
11:30:17 Nick J-23_ is now known as J-23 (
11:32:18 Quit dfkt (Read error: 110 (Connection timed out))
11:44:13linuxstbrasher: Regarding the themes site, what functionality have you implemented (or are planning to implement) ?
11:44:38 Join bazzawilleee [0] (
11:45:12 Join planetbeing_ [0] (n=planetbe@
11:45:20bazzawilleeehey can anybody tell me quickly how to determine the firmware version on a sansa e250 so I can tell a ebay seller?
11:45:39rasherSo far displaying of themes, uploading of themes, verification of the .zip structure (no .wps syntax check yet - but that's the plan) and verification of email, the beginnings of an admin panel where themes can be hidden/deleted.
11:46:10linuxstbSo every theme uploader will need to provide a valid email address?
11:46:24rasherNot only valid, but one they control
11:46:28linuxstbOK, sounds sensible.
11:46:46BigBambibazzawilleee: It is in system - info or something similar in the Sansa firmware
11:46:53linuxstbAre themes automatically published, or does an administrator need to press a button?
11:47:11rasherI think the idea is to publish automaticly
11:47:25BigBambiI think one of these silly US laws needs that
11:47:39bazzawilleeethanks bigbambi I will pass it on
11:47:41rasherWell, it puts us in a better situation iiuc
11:47:53BigBambiDMCA or whatever, to qualify for safe harbour
11:48:09*scorche nods in agreement
11:48:15Unhelpfulamiconn: any multiply larger than 16*16->32 goes to a libgcc function on sh1, right? i may be able to speed up the downscaler a good bit, some of the per-output-column-per-input-row multiplies can be limited to the hardware multiply size without restricting image size unreasonably
11:48:31rasherIt's not really an issue to do one or the other by default
11:48:40rasherSo whichever we decide on
11:49:01linuxstbI think it makes sense to publish automatically anyway - it's easier for us.
11:49:08BigBambiI think given that it is hosted in the US (or will be), it pretty much has to be automatic
11:49:14 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
11:49:23BigBambiand it is easier :)
11:50:58scorcheare there themes still created that are not compatible with svn rockbox?...for a reason other than "age", i guess
11:51:01 Join dfkt_dt [0] (i=dfkt@unaffiliated/dfkt)
11:51:49linuxstbrasher: When you check a theme zip, do you also look for a remote wps?
11:53:04linuxstbscorche: I expect so (multi-font for one), but I thought it was agreed we would only host svn-compatible themes? So the validation should hopefully reject them - but IIRC the wps parser doesn't reject unknown tags, it just ignores them. It only gives errors when a tag is misused.
11:53:43rasherlinuxstb: I detect if zero or one rwps exists, yes. But the fact is not currently stored (that's based on the target the uploader selects). It should probably be a bit smarter there
11:54:00 Part bazzawilleee
11:54:49scorchelinuxstb: multifont was the only one i could think of, yeah....and i dont see a problem with hosting only svn-compatible themes now...especially as it will be linked in to rbutil...
11:58:53 Join Seed [0] (
11:59:05linuxstbrasher: That's always the kind of thing that can be added later (i.e. scanning the uploaded files and identifying those with a .rwps). I don't want to distract you from the basics, so ignore me...
11:59:16 Quit homielowe ()
11:59:50linuxstbrasher: But what list of targets are being used? I wasn't sure about that when I was working on the site. Are you actually using target names, or just lcd sizes/depths ?
12:00:10rasherA bit of a mix, actually
12:00:30 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
12:00:51rasherUploaders give the site a target name, but it stores the LCD dimensions
12:01:10rasherVisitors pick a target, but the site picks themes that match that target's LCD dimensions
12:01:47pixelmawhat about themes with more than one rwps (e.g. cabbiev2) - is the site clever enough to package "on demand"?
12:02:04rasherA theme can't have more than one rwps.
12:02:45 Join kugel [0] (n=kugel@rockbox/developer/kugel)
12:02:48rasherThe theme site deals in individual themes, not packages like the source
12:03:17pixelmaa theme can as quite a few in SVN do, of course there will only be one for each target's zip... ok, that's what I wanted to know
12:04:21rasherI don't think it'd be worth it to add that complexity. I think it's quite reasonable to have uploaders upload different theme versions individually
12:06:42pixelmaI think it won't happen very often but if someone wants to upload a theme for more than one target it could be a bit annoying
12:06:55scorchealso, just as another reminder, if anyone has any ideas for GSoC, please put them down here...if we wish to be accepted, we really need to have this page fleshed out a bit more If we have no more ideas, perhaps we can give each idea in this a bit more detail about how it should be accomplished...
12:07:33 Join kushalone [0] (n=kushal@
12:08:21rasherpixelma: Frankly, I think having to package your different themes into the "package" format would be more work
12:09:45 Quit dfkt_dt (Read error: 110 (Connection timed out))
12:11:47 Join dfkt_dt [0] (i=dfkt@unaffiliated/dfkt)
12:12:03linuxstbscorche: What's the deadline for GSoC?
12:12:39scorchelinuxstb: organization apps are accepted next week (9th to the 13th)
12:14:51rasherMan, timestretch is only a 3+3kb bin/ramsize increase?
12:15:12rasherThat's bloody nice going
12:18:51pondliferasher: Perhaps you could try the latest version (090306b)? ;)
12:19:20pondlifeAny reports welcome, I'm hanging around for them...
12:20:33 Join tyfoo [0] (
12:24:20pixelmapondlife: I too think it doesn't belong into the thread you posted too, but I don't understand how your button "table" works with 6 buttons
12:24:52pondlifeDid I count wrong?
12:25:18pondlifeLEFT/RIGHT/UP/DOWN/MENU/SELECT/PLAY/STOP - oops, that's 8
12:25:32pondlifeI knew I was making some silliness
12:26:27pondlifeFor some reason I was looking at UP/DOWN and LEFT/RIGHT as single buttons :/
12:26:41linuxstb"PLAY/STOP" is one button on at least ipods though, so that could make 7...
12:26:54pixelmaand yes, the Ondio is the most restricted wrt buttons or better "actions", it has 6 buttons (the Ipod has 5 + scroll fwd/back). Still it has a very intuitive mapping, better than other targets
12:27:34pondlifeForget the existing mapping, I was trying to remove that from the equation... but I got it horribly wrong.
12:27:52pondlifeIs the Ondio mapping in the wiki still correct?
12:30:20pixelmathe only limitation there is, is it doesn't have a quickscreen due to that. Which wiki page (never looked at it so can't answer the question)?
12:31:20 Join lasser [0] (
12:31:29pixelmaAlso - if you mean button combos with the "shift" key, there are some targets where button combos are not possible
12:31:45pixelmapondlife: I'd suggest looking at an Ondio manual
12:31:47pondlifeYes, I'm aware that multi-buttons is of limited use
12:31:51pondlifeOK, will do
12:32:12pixelmaor a sim
12:33:15pondlifeHmm, so the Ondio lacks a PLAY button (i.e. resume from browser or menus)?
12:34:23 Quit dfkt_dt (Read error: 110 (Connection timed out))
12:34:27lasserhi! I just compiled RB 20213 and got this error:
12:34:29lasserWarning: character 8491 goes 1 pixel(s) beyond the font's ascent, it will be clipped
12:34:31lasserWarning: DWIDTH spec > max FONTBOUNDINGBOX
12:34:33lasserWarning: 200 characters out of 812 were clipped (162 at ascent, 38 at descent)
12:34:34lasser max overflows: ascent: 4, descent: 2
12:34:36lasserWhat is going wrong?
12:34:46pixelmapondlife: ok, it has no "Play" button but resume is a short press of "Mode" from the browsers or menus
12:34:51 Join dfkt_dt [0] (i=dfkt@unaffiliated/dfkt)
12:35:21pondlifeAh, so Mode = PLAY, long Modw = MENU ?
12:35:24 Quit dfkt (Read error: 110 (Connection timed out))
12:35:36pixelmalasser: nothing, it just tells you that some fonts are a bit broken, it's just a warning
12:35:36 Join Darksair [0] (n=user@
12:35:49pondlifeNote my capitals refer not to physical buttons, just to my imaginary ones ;)
12:35:50pixelmapondlife: yes
12:35:59 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
12:37:36pondlifepixelma: Is there a way to stop playback from the browser?
12:37:44pondlifeI can't see it listed.
12:38:19pixelmapondlife: it also has no real one button access to the menu from the WPS (either go through the WPS context menu or through the file browser), I don't miss one and find the other two more important to access
12:38:43***Saving seen data "./dancer.seen"
12:38:52rasherlasser: the font converter got more strict. Nothing to worry about
12:38:54pondlifeBasically, it lacks dedicated MENU and STOP buttons in my terminology
12:39:25pixelmapondlife: seems to stop playback in the browsers when I press the "On/Off" button
12:39:40lasserpixelma, rasher : ok, thanks!
12:39:48pondlifeAh, yes
12:43:51pondlifepixelma: One more question if you don't mind: In the WPS, long On/Off is used to stop playback. Does this mean there's an even longer hold required for power off?
12:44:00pondlifei.e. short/long/longer
12:45:40pixelmayes, and I can't remember shutting down accidentally though
12:46:07pondlifeOK, that's fine
12:46:23pondlifeIt explains my misunderstanding, will think some more
12:48:06pixelmaBasically, it lacks dedicated MENU and STOP buttons in my terminology <- to me the main difference is that there is no dedicated "Select" button
12:48:32pondlifeAh - right is used instead...
12:49:30pondlifeCan you pause in the WPS?
12:49:41pondlifeAh, short On/Off
12:51:42 Quit lasser ("ChatZilla 0.9.84 [Iceweasel 3.0.6/2009020409]")
12:52:00pixelmathe right/select thing is something people claim to be almost redundent on other targets. The only difference is that there is no horizontal scroll due to it (well there is with a combo) but that's not important at all to me
12:52:48pondlifeIt's largely redundant. I mainly use select for the context menu.
12:53:24pondlifeLong MENU makes just as much sense for that anyway.
12:53:29 Quit planetbeing_ ()
12:55:57 Quit dfkt_dt (Read error: 110 (Connection timed out))
12:57:36 Quit dfkt (Read error: 110 (Connection timed out))
13:12:41 Quit BUMBACL0T ()
13:13:00pixelmapondlife: someone reminded me of the fact that teh Player also only has 6 buttons ;)
13:13:05 Join BUMBACL0T [0] (n=ORF@unaffiliated/bumbacl0t)
13:13:30pondlifeWe should only need 6 buttons :)
13:15:29domonokyindeed, my archos player has only 6 buttons. it is the absolute minimum we need for rockbox. more are better :-)
13:21:36pixelmaas it seems though, the number of buttons alone don't tell the whole story... :\
13:42:00 Quit rocko ("Leaving")
13:47:16 Quit FlynDice ("Konversation terminated!")
13:59:23 Join midijunkie [0] (
13:59:26 Quit midijunkie (Remote closed the connection)
14:14:04 Join nibbler [0] (
14:31:36rasherWhy does switching language stop playback?
14:33:27 Join moos [0] (i=Mustapha@rockbox/staff/moos)
14:38:46***Saving seen data "./dancer.seen"
14:40:55LinusNrasher: maybe to load the voice file?
14:44:25rasherBut there is no voicefile. And why not restart it afterwards anyway?
14:45:15 Join nuonguy [0] (
14:47:24LinusNbeats me
14:47:53 Join CaptainKewl [0] (
14:51:07 Join n1s [0] (n=n1s@rockbox/developer/n1s)
14:53:31 Join ldp [0] (n=ldp@
14:55:18 Join faemir [0] (
15:02:43 Quit nuonguy (Read error: 113 (No route to host))
15:06:38 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner]")
15:11:57 Join Ubuntuxer [0] (
15:12:39 Part Ubuntuxer
15:13:07 Join kugel [0] (n=kugel@rockbox/developer/kugel)
15:16:56 Join Ubuntuxer [0] (
15:17:31 Part Ubuntuxer
15:18:45 Join Sedgewick [0] (n=Sedgewic@
15:19:55 Join pyro_maniac1 [0] (
15:24:30 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
15:33:25 Quit pyro_maniac (Read error: 110 (Connection timed out))
15:35:52 Quit jordoex (Read error: 145 (Connection timed out))
15:36:50 Join evilnick [0] (i=0c140464@gateway/web/ajax/
15:37:03 Nick fxb__ is now known as fxb (
15:40:01 Quit bmbl (Read error: 110 (Connection timed out))
15:50:28kugelpondlife: are you planning to commit before 3.2 or after?
15:50:41pondlifeBefore, I hope
15:51:30pondlifeWhy do you ask?
15:51:53kugeloh, I'm just curious. freeze is in 3 days
15:52:02pondlifeYep :)
15:52:03*kugel would like to have the disable wps update or rec button for playlist before 3.2 too, but tends to mess things up before releases too
15:52:41ldpAre you guys the developers?
15:53:11evilnickIf a new feature is added before the freeze then you have all the time in the freeze period to fix it up!
15:53:28pondlifeevilnick: ^
15:53:36ldpGood work
15:54:00pondlifeBut I'd rather not commit something that needs fixing up...
15:54:10The-Compilerldp: some of the guys here are ;)
15:54:53ldpI see
15:55:34pondlifeI think of the freeze as a beta test period. Patches on the tracker don't get enough testing until the get into SVN, often.
15:56:37LloreanIf you want something to get wide testing, you should commit it right after 3.2. Not immediately before.
15:56:46evilnickpondlife: It does make it a lot easier to get feedback as all the users who can't/don't compile get to report their findings.
15:57:12LloreanIf it doesn't work out right before the freeze, you're just putting more things in we might have to cut and potentially end up with a 3.2 less stable than it could be. The whole point of release versions is to hope to have them stable.
15:57:19LloreanIs there some rush why the feature can't be in 3.3 instead?
15:57:20pondlifeLlorean: Does is make much difference? I am happy to wait if that's what people would prefer.
15:57:50Lloreanpondlife: In my mind, the freeze is the "formal" freeze. Shortly before that, people should be cautious about what they commit, and commit things that are _very_ unlikely to have problems.
15:58:04LloreanWe want 3.2 as stable as possible, rather than aiming for a lot of new features.
15:58:10LloreanOthers may disagree, though
15:58:23pondlifeAh, ok. I thought of it as a "feature complete" point, mainly.
15:58:36pondlifeNo rush on my behalf, certainly.
15:59:05LloreanI don't think two weeks is long enough to reliably go from "feature complete but possibly buggy" to "we're confident there's a minimal set of bugs" personally.
15:59:47LloreanIn my mind the best time to get new features in, is when they have the longest period to be tested.
16:00:19pondlifeMarch 24th then :)
16:00:43LloreanA week before that, because you can just commit it to the non-release branch
16:02:38LloreanThough if it's expected to work very well, I don't really see a problem with going for it now.
16:02:42 Quit CaptainKewl (Read error: 110 (Connection timed out))
16:02:56Lloreankugel: The record button for playlist still needs to be adapted for all targets, and have a function picked for in the menus, right?
16:03:38pondlifekugel: Were you still allowing "no action" for that button?
16:04:14 Part ldp
16:04:20 Part LinusN
16:04:29n1spondlife: did the speedup patch get tested with audio that needed to be upsampled?
16:04:50pondlifen1s: Not as far as I know, but please do
16:05:16pondlife(Or, if you can provide me such audio, I'll do it.)
16:05:27Lloreanpondlife: We don't typically give users the option to turn off and on buttons...
16:05:36*n1s has no such audio
16:05:55LloreanInstead of thinking "configurable buttons" think "which button function adds the most functionality to Rockbox's featureset."
16:06:43pondlifeGuess I can always toggle hold.
16:07:21LloreanThe issue of configurable buttons is an issue that can be discussed, but it really should be outside what this patch does (since if it's accepted, then new functions can be added to this or possibly other buttons anyway)
16:07:46LloreanBut for this patch's sake, it shouldn't be held up on what may or may not be an indefinite debate on whether configurable buttons (either in a limited sense, or unlimited) are a good idea.
16:08:16*n1s is starting to like the idea of configurable buttons actually, probably limited to the "non essential" ones though
16:09:14*pondlife doesn't like the idea of core buttons being configurable, nor of "special buttons", and believes that ultimately Rockbox will have .keymap files or somesuch...
16:15:26Lloreann1s: The problem is, we're still picking and choosing what we think is essential.
16:15:42LloreanWe've already had people try to argue that things like seeking are nonessential and the buttons should be freed up for other things.
16:16:09n1sLlorean: i mean essential as "enough to fix messed up buttonmaps" basically
16:16:25Lloreann1s: I don't understand.
16:16:36n1swhat is more important is the hw limitations though, not all buttons can be used in combos etc
16:17:41n1sLlorean: that depends on how the maps are done of course, it it is like a theme file then i guess i think any button that can be mapped should be possible to map
16:27:12 Quit ender` (" Women and Cats will do as they please. Men and dogs had better get used to it. -- Robert Heinlein")
16:31:16 Join midgey [0] (n=tjross@
16:31:58 Quit nibbler ("Ex-Chat")
16:36:18 Join robin0800 [0] (
16:38:49***Saving seen data "./dancer.seen"
16:39:12kugelLlorean: a function picked for the menus? So we did not agree on Insert?
16:42:00 Quit robin0800 (Remote closed the connection)
16:42:11rasherkugel: It'd probably be nice to include something other than the FS number in the commit message.
16:43:22kugelrasher: well, why? You can klick on the FS and get all information
16:43:35 Join serendipity_ [0] (n=serendip@
16:43:37rasherThat only works on the front page...
16:43:50 Quit SirFunk (Read error: 60 (Operation timed out))
16:44:03 Join CaptainKwel [0] (i=2669ecc2@gateway/web/ajax/
16:44:06rasherNot in the commit mails, not in viewvc, not in svn log
16:44:18 Part CaptainKwel
16:45:15kugelpondlife: how would I allow no action? There's no configurability in this patch
16:45:30 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
16:45:44 Join SirFunk [0] (
16:45:50pondlifeI just wondered it you'd consider adding an option for that, that's all. Not a biggie though
16:46:16pondlifeI'm just very used to using REC to turn on the backlight
16:46:25pondlifeWill toggle hold instead
16:46:27serendipity_are there development tools for mac?
16:46:45kugelpondlife: I'm not..I exclusively use hold for that
16:46:58 Join toffe82 [0] (
16:47:11pondlifeI'll just re-educate my fingers...
16:47:13kugelvery rarely other buttons too, since I have "first button press enables backlight only" enabled
16:47:33n1sserendipity_: the standard linux tools should work fine on osx
16:47:50serendipity_oh, nifty
16:47:52n1sthe wiki has all the juicy info on setting up a dev environment
16:49:46 Join Scoth [0] ( the WTFPL considered a compatible license for releasing a WPS under (and not having it end up in the graveyard)? Thats what I like to release any of my work involving photoshop under
16:54:46rasherserendipity_: It'd be nice if you dual-licensed it, just so there's less confusion
16:55:13serendipity_you can do that?
16:55:30serendipity_release something under two?
16:55:34serendipity_humm. ok
16:56:34 Join planetbeing_ [0] (n=planetbe@
16:57:20rasherAs long as it's all your work, you can do whatever you like with it. We encourage people to pick CC-BY-SA 3.0, so there's less confusion about licenses
16:57:21serendipity_thanks rasher
16:58:12linuxstbrasher: Will your themes site have a choice of license, or are you intending to force the CC-BY-SA 3.0 on all uploads? (I would be in favour of the latter, for simplicity).
16:59:13rasherThe latter
17:01:04rasherSomeone like serendipity_ could of course say in the comment text that it's also available under a more liberal license
17:01:28 Join midijunkie [0] (
17:01:49serendipity_you cant get much more liberal than WTFPL =P
17:02:39*rasher knew it already
17:03:36serendipity_but okay, i'll dual under wtfpl and a cc ok-corporate no-give-credit derrivativs-under-any-liscence whatvever
17:05:13Lloreankugel: How do you "Insert" while in the menu? (Not Filetree, MENU)
17:06:19Lloreanserendipity_: The idea is a specific CC, not "whichever CC license you like"
17:06:48 Join Lss [0] (
17:07:08Lloreankugel: Have to go. Will check logs later. But think. The button still needs a "job" within the menu tree where 'insert' doesn't do any good (you can't insert "Sound Settings" into a PL) and possibly in the playlist viewer itself.
17:08:09 Quit orsonj (Remote closed the connection)
17:08:12rasherIt's simply to make people have to think less when re-using stuff from other themes. If all themes use the same license, things get far simpler. (and yeah, WTFPL on its own would do it in theory, but to save people the bother of having to look it up)
17:08:14kugelLlorean: well, there's nothing to insert, so it has no action yet
17:08:46serendipity_do they even have a no-attrib yes-commercial yes-derrivative not-share-alike?
17:09:13kugelplaylist_viewer could be done. But I fear that user don't get the difference between menus and database/filetree, which would then lead to unexpected behavior
17:09:23rasherserendipity_: There's the very recently released CC0, but the "standard set" doesn't have no-attrib I believe
17:09:35serendipity_philosophically....yes-commercial, because as a pirate, i benefit from corporate works all the time || and no-credit because i hate writing documentation, and want to spare others from it (and foss purists usually give credit anyways) || and derrivatives-under-any because ... meh
17:09:52serendipity_where do i get cc0?
17:10:39rasherserendipity_: It doesn't really matter though, as long as you have it available as WTFPL also, surely?
17:10:45 Join orsonj [0] (
17:11:08serendipity_whoop, found it
17:11:19rasherThis is getting off-topic though
17:11:52serendipity_back to work for me
17:12:01 Quit Zagor ("Don't panic")
17:12:12 Nick serendipity_ is now known as serendipity_|not (n=serendip@
17:13:09 Quit petur ("gimme a beer!")
17:24:03 Join serendipity_ [0] (n=serendip@
17:25:26 Join rocko [0] (
17:27:46 Join serendipity__ [0] (n=serendip@
17:27:47 Quit serendipity_ (Read error: 104 (Connection reset by peer))
17:30:30 Join CaptainKwel [0] (i=2669ecc2@gateway/web/ajax/
17:31:39 Quit Sedgewick ("off")
17:34:32 Quit serendipity_|not (Read error: 110 (Connection timed out))
17:35:10 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
17:37:13 Join gartral [0] (
17:37:58 Join jordoex [0] (
17:38:04 Part gartral
17:38:49 Quit jhMikeS (Nick collision from services.)
17:38:54 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
17:39:48 Quit midijunkie ("?(???~•~)?")
17:41:12 Join MethoS- [0] (
17:43:18pixelmakugel: if I'm reading the FS task right (just skimming though), the database format has changed now and would need a reinitialise?
17:43:53 Quit orsonj (
17:43:53 Quit n1s (
17:43:53 Quit AndyI (
17:43:53 Quit BlakeJohnson86 (
17:43:53 Quit Ridayah (
17:43:53 Quit tchan (
17:43:53 Quit maraz (
17:43:53 Quit shadearg (
17:43:53 Quit dionoea (
17:43:53 Quit shodanX (
17:43:53 Quit rphillips (
17:43:53 Quit lostlogic (
17:43:53 Quit Tuplanolla (
17:44:49NJoinorsonj [0] (
17:45:37NJoinshodanX [0] (
17:46:08NJoinn1s [0] (n=n1s@rockbox/developer/n1s)
17:46:08NJoinAndyI [0] (i=AndyI@
17:46:08NJoinBlakeJohnson86 [0] (
17:46:08NJoinRidayah [0] (
17:46:08NJointchan [0] (
17:46:08NJoinshadearg [0] (
17:46:08NJoinmaraz [0] (
17:46:08NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
17:46:08NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
17:46:08NJoinrphillips [0] (n=rphillip@unaffiliated/rphillips)
17:46:08NJoindionoea [0] (n=dionoea@videolan/developer/dionoea)
17:48:36kugelpixelma: yes
17:49:01kugelit's reinitialise automatically, due to the version number
17:49:28 Nick fxb is now known as fxb__ (
17:50:29 Quit Darksair ("People who are zhuangbility want to show their niubility but only reflect their shability.")
17:50:46pixelmathat should have been mentioned in the commit message then. Won't people lose their runtime data, rating etc. if they don't export them before and then import (I just think I read it somewher, don't use it so am not sure)
17:51:26 Join dfkt_dt [0] (i=dfkt@unaffiliated/dfkt)
17:52:55kugelhm, yea, I think so. I haven't thought of that, I don't gather runtime data
17:52:56 Join flydutch [0] (
17:53:11 Join jgarvey [0] (
17:53:38pixelmaI also think that commit messages should be quite explicit in general
17:54:33linuxstbpixelma: Absolutely... They should be self-contained and say _why_ something is done, not just what (the code itself normally says what).
17:54:37 Quit serendipity__ ()
17:55:56linuxstbkugel: It's not helpful to just put the FS number in the log message - it should really describe what the patch actually does, and then give the FS number for further information.
17:56:41kugelok, I'll be more verbose again next time, sorry :(
17:58:00rasherFrankly, I don't like "oops" and "fix red" commit messages either
17:59:11 Join _Wonko [0] (
17:59:34 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
17:59:56 Quit kushalone ("Leaving. I cannot promise to be back but most likely will.")
17:59:56CtcpPing from gevaerts!n=fg@rockbox/developer/gevaerts
18:00:00_Wonkohi, i was wondering, how can you get rockbox to power off, and not tax the battery when i'm not listening
18:00:17CtcpPing from gevaerts!n=fg@rockbox/developer/gevaerts
18:00:28_Wonko*power down is the correct expression i guess
18:00:51scorchethat depends on the device you are using, however this and many more such questions are answered in the fine manual...
18:01:29rasher_Wonko: Look for "Idle Poweroff"
18:01:34rasherWhich does exactly what you'd expect.
18:02:06 Join gartral [0] (n=gareth@
18:02:09 Part gartral
18:03:13 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner]")
18:03:44_Wonkothank you! yes, that seems to be exactly the thing i'm looking for
18:04:18_Wonkodoes rockbox auto power down on other devices?
18:04:51gevaertsCan we start GSoCApplicationTemplate2009 as a copy of GSoCApplicationTemplate2008, or do we expect too big differences?
18:05:04scorchegevaerts: that should be fine
18:05:10rasher_Wonko: I don't understand that question
18:06:42pixelma_Wonko: "Idle poweroff" exists for all targets in Rockbox
18:07:21_Wonkoassuming the intended behaviour is shutting down the player when you pause the playback for an extended period of time... does rockbox do that on other players than my first gen ipod?
18:07:47evilnick_Wonko: Yes
18:08:24evilnickPause or music stopped and no button presses, I believe
18:09:20_Wonkoyes, stopped music i mean
18:09:34 Quit dfkt (Read error: 110 (Connection timed out))
18:10:49*amiconn dislikes the whole idea of configurable buttons
18:10:55_Wonkoi will try the "workaround" i found when searching for idle poweroff.. hope it works, cause at the moment, my battery is gone within perhaps a day at the most
18:15:31 Nick dfkt_dt is now known as dfkt (i=dfkt@unaffiliated/dfkt)
18:26:35 Join jordoex_ [0] (
18:27:17linuxstb_Wonko: What workaround have you found? AFAIK, Rockbox doesn't have any form of poweroff on the 1st/2nd gen ipods (assuming you're talking about the original 1st gen)
18:27:50 Part pyro_maniac1 ("Leaving.")
18:30:36gevaertsscorche: do you think the USB bits on SummerOfCode2009 need more details?
18:31:28scorchegevaerts: sure...anything on the page could use some expanding, likely...i was planning on expanding the USB bits if it came to it, but if you want to... ;)
18:32:46gevaertsscorche: I can work on the USB bits, but I'm not really sure what to add
18:33:37 Quit jordoex (Read error: 145 (Connection timed out))
18:35:02scorchei was thinking about perhaps separating out the individual modes, say what they do/why it would be handy to have...perhaps even point in a direction of an existing implementation that could be looked at
18:35:23gevaertsah ok.
18:35:33_Wonkolinuxstb, i guess i'm talking about that one... the link is this:
18:35:44_Wonkothe comment by Jens Arnold
18:35:55 Quit SirFunk (Read error: 110 (Connection timed out))
18:36:05_Wonkoi have yet to try it out, as i'm currently loading the batteries ;)
18:36:32_Wonkooh fuck, i need to be going
18:36:40_Wonkowell, cya!
18:36:49 Join SirFunk [0] (
18:37:25 Quit _Wonko (" HydraIRC -> <-")
18:38:52***Saving seen data "./dancer.seen"
18:39:53pixelmaahh... 1st or 2nd gen Ipod. I forgot :\
18:40:48 Quit moos ("Rockbox rules the DAP world")
18:41:52linuxstbpixelma: You were technically correct though, the "idle poweroff" option does exist for those ipods, it just doesn't work...
18:42:32pixelmawell, it works... but will get you into a reboot cycle
18:43:29 Join miepchen^schlaf [0] (
18:43:30rasherMaybe it should be disabled?
18:43:55rasherPeople might rely on it, not knowing that it's going to reboot
18:45:49 Part pondlife
18:46:32pixelmathat won't help them completely though, they still need to be aware having to reboot into the OF
18:48:23rasherYes, but that's better than the current situation in my opinion. Having an Idle Poweroff that doesn't actually work is bad.
18:48:30 Quit miepchen^schlaf ()
18:51:41 Quit sarixe ("Leaving")
18:54:53 Nick jordoex_ is now known as jordoex (
18:56:00midgeygevaerts: do you have an mtp h10?
18:56:02Slasherifortunately the db shouldn't reinitialize automatically when the version changes so users can revert to a previous version and do export using that version
18:57:53 Join midijunkie [0] (
18:59:03rasherSlasheri: Any hope for FS #7287 - Sort tags?
19:03:04Slasherirasher: technically that patch probably should be ok, but it's hard for me to say how important feature that would be (i wouldn't need that). And probably it would significantly increase the db size
19:04:17 Join kushalone [0] (n=kushal@
19:04:45rasherWell, two things: I don't think we *ever* want to display the sort tags. And how much larger are we talking? And is it significant, in the grand scale?
19:07:58Slasherirasher: looking from the patch, it would at least add 16 bytes for each song and then the actual data for the sort tags (not sure if that data may get to the ram for faster searches)
19:08:07pixelmaI still don't understand why it shouldn't display sort tags, I mean in a record store it will be "Presley, Elvis" too
19:08:14pixelmabut maybe I'm special
19:08:41 Quit Nico_P (Remote closed the connection)
19:09:33rasherWell, the whole point of sort tags is to sort differently than the correct name
19:10:29pixelmayes, I could still imagine reading "Presley, Elvis" in the browser and "Elvis Presley" in the WPS if I had a song by him
19:11:37pixelmain my eyes it would read weird to read something starting with "E" among other words starting with "P"
19:11:42 Quit kushalone ("Leaving. I cannot promise to be back but most likely will.")
19:11:53 Join kushalone [0] (n=kushal@
19:13:31Lloreanrasher: As long as the size increase is "per song" and possibly "only for songs where the tags are present" the amount shouldn't matter since it's user choice whether to increase their own DB size.
19:13:54rasherLlorean: I think it'd be for all songs.
19:14:18LloreanWell, as I said "possibly" for that one.
19:14:52rasherHuh, did I not just *move* tools/release/
19:15:00*Llorean thinks that if the database is set to filter by "artist" it should use sort_artist when present and artist when not, but that maybe there should be an option to explicitly sort by sort_artist that simply shows the contents in that tag and sorts by it, for people like pixelma
19:15:12LloreanSince sorting by a tag without any defaulting doesn't require any particularly fancy new logic anyway
19:16:08 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
19:16:37saratogawas there any discussion about possibly using some of our money to send funman a Clip?
19:17:18rasherNot much discussion iirc. I think I said it's quite likely it'd be possible
19:17:57saratogaI sent him my fuze, but its 1) not a clip, and 2) probably going to float around the US postal service for quite a while
19:21:04LloreanWell, the fuze probably allows him to make more progress than a Clip does anyway. Fix the >2gb thing, at least.
19:21:22 Join qurvel [0] (
19:21:50gevaertsmidgey: I had, but I decided that the mtpness of it was just annoying, so now I have an msc h10
19:22:31midgeyah, well i quickly whipped up a patch for FS #9966
19:23:42 Join bertrik [0] (
19:24:13 Quit jordoex (Read error: 104 (Connection reset by peer))
19:25:09XavierGrrasher: regarding themes licensing, I got in contact with Ioannis Koutoulakis (iCatcher & UniCatcher) and he told me that he doesn't mind any relicensing that will happen on his themes. Specifically he told me "What fits better for Rockbox."
19:25:42XavierGrrasher: also he said there was once a similar discussion about it here between him (genre9mp3) and linuxstb.
19:26:56saratogaalso, we have only two interested mentors listed on teh GSOC page
19:27:09saratogai'm sure we have more interested people then that
19:27:44saratogaperhaps a front page news item about it would be a good idea, as we did on this day, last year
19:31:24 Join bs66_1 [0] (
19:34:01*domonoky adds himself to the page..
19:34:53amiconnUnhelpful: The downscaler is only used in PF during cache creation, correct?
19:35:52 Part bs66_1
19:37:09 Join miepchen^schlaf [0] (
19:46:29 Quit bs66_ (Read error: 110 (Connection timed out))
19:49:08 Quit lostlogic ("Reconnecting")
19:49:16 Quit rocko ("Leaving")
19:49:19 Join lostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
19:51:40gevaertsscorche: I've fleshed out the USB device side things a bit. It's not really good yet, but there's a lot more information now. I haven't done the host side yet. Feel free to improve on sight :)
19:56:14 Join FlynDice [0] (
19:56:47 Quit miepchen^schlaf ()
19:58:37 Join MethoS-- [0] (
20:00:22 Join jaykay [0] (
20:01:42 Quit MethoS-- (Remote closed the connection)
20:03:51 Quit jhMikeS (Nick collision from services.)
20:03:57 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
20:07:41 Join jaykay_ [0] (
20:09:47FlynDiceBertrik:I got some strange results with your ams sansa rtc patch
20:10:25bertrikyeah, the day of week is incorrect
20:10:57FlynDicesetting the date in rockbox sets it 2 days earlier also
20:11:31bertrikyou mean, the OF shows 2 days earlier than RB?
20:12:19FlynDiceno, when I set March 6 in the setting page and save, it saves March 4
20:12:19 Quit MethoS- (Read error: 113 (No route to host))
20:12:47FlynDicemarch 8 goes to March 6 etc
20:12:57bertrikhmm, weird, I'm thinking of another, simpler, way of fixing it
20:13:15bertrikthanks for letting me know
20:13:38FlynDicesure I left a couple of comments on FS
20:14:58 Quit Lss ()
20:15:29 Quit FlynDice (Remote closed the connection)
20:16:25 Quit jhMikeS (Nick collision from services.)
20:16:31 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
20:21:26 Quit jhMikeS (Nick collision from services.)
20:21:32 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
20:23:24 Quit jaykay (Read error: 110 (Connection timed out))
20:30:28 Part qurvel
20:38:54***Saving seen data "./dancer.seen"
20:39:07 Join mirak [0] (
20:43:08 Join arohtar [0] (
20:46:14 Quit faemir (Read error: 104 (Connection reset by peer))
20:50:16 Quit CaptainKwel (" ajax IRC Client")
20:52:44 Quit yhuang ("Leaving")
20:53:28 Join JdGordon| [0] (i=836b0049@gateway/web/ajax/
20:57:24 Quit saratoga ("CGI:IRC (Ping timeout)")
21:00:53 Join yhuang [0] (n=yhuang@unaffiliated/yhuang)
21:07:35 Join Lss [0] (n=Lss@
21:15:03 Quit tyfoo (Read error: 104 (Connection reset by peer))
21:15:38 Quit gevaerts (Nick collision from services.)
21:15:47 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
21:17:16 Join pyro_maniac [0] (
21:20:01bertrikoh nice, something was changed from short to int32_t and binsize went down ...
21:21:31 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
21:27:42n1sbertrik: not really surprising, dealing with regular ints is usually the simplest
21:29:08n1sah, the lack of frequent yielding when benching the midi plugin seems to have been the cause for it to crash
21:29:29n1sa little strange that it just turned the sansa off though
21:29:32 Join Aurix_Lexico [0] (
21:30:16n1san overeager watchdog timer thingy somewhere perhaps?
21:33:03 Quit yhuang (Read error: 110 (Connection timed out))
21:33:18pixelmaI'm not sure if it's still the case but I remember that playing midi did not reset the idle power off timer
21:35:21 Join shotofadds [0] (n=rob@rockbox/developer/shotofadds)
21:36:15pixelmaseems it's fixed though
21:36:21n1spixelma: this was not that at least, it happened much too quickly and suddenly (no splash or anything) and also happened while charging from usb
21:37:13n1sanyway, this is no problem for the svn version just when i disabled the actual pcm_play_data call
21:40:26shotofaddsgevaerts: ping
21:40:41*gevaerts wakes up
21:40:45pixelmaugh, seems this backlight fade out looks most flickery (taking longer) when the CPU is busy, very noticable while playing midi on my c200 (I noticed earlier that it looks different now and then but wasn't sure under which circumstances)
21:41:03gevaertsshotofadds: pong
21:41:16shotofaddsthere's a patch on flyspray to enable USB serial for the d2 (fs#9917) which I don't entriely understand. not least because USB serial works without the majority of that patch.. I was wondering if you had any thoughts about it?
21:41:33n1spixelma: possibly that is at least partially caused by the too infrequent yielding, /me will fix
21:43:11 Join ender` [0] (
21:43:16shotofaddsgevaerts: all I did was add USB_NUM_ENDPOINTS, define usb_attach(), and define USB_SERIAL
21:43:16shotofadds in usb_core.h. I don't really understand what the other changes are trying to achieve.
21:44:12*shotofadds thinks this is possibly better posted to the fs task...
21:46:37gevaertsshotofadds: I don't really know either.
21:46:40 Join albinoblackrabbi [0] (n=519b4220@gateway/web/cgi-irc/
21:46:50shotofaddsgevaerts: ah, I see it. the logf text often seems to have a repeated or missing character at the end of each line. those read/write changes are presumably aimed at fixing that. i'll test it out.
21:47:28*gevaerts hasn't really looked at the datasheet for this controller, so it's all a bit magical
21:47:29 Quit albinoblackrabbi (Client Quit)
21:53:44 Quit jaykay_ (Read error: 110 (Connection timed out))
21:54:50gevaertsshotofadds: some bits of it seem to make no sense at all to me
21:56:06gevaertsLike e.g. why does it replace that for loop in usb_drv_write_packet() by a while loop that does exactly the same except it's more verbose?
21:56:30gevaertsor am I missing something?
21:57:29gevaertsSimilar things happen in handle_ep_in()
21:57:35shotofaddsgevaerts: I have no idea. it also makes the player crash on USB insert :/
21:58:52gevaertsIt looks to me like he changed things more or less at random until he got it working, and then posted a patch
22:01:09 Join efyx [0] (
22:02:27gevaertsI think you just need USB_NUM_ENDPOINTS and usb_attach(). The rest of that patch isn't very useful IMHO
22:03:28gevaertsexcept that the #define for USB_NUM_ENDPOINTS should go to tcc*.h, not config-cowond2.h
22:03:29 Join |amadeus| [0] (
22:03:57shotofaddsgevaerts: I concur. I think the patch is trying to address buggy handling of buffer lengths that are indivisble by 2, but it seems to be doing it in a slightly strange way. I'll look into it some more tomorrow.
22:04:28shotofaddsand yes, I thought it was strange adding that to config-*.h - but several other targets seem to do this
22:05:17gevaertsAll of those are mcuelenaere's :)
22:05:57shotofaddshehe, us New Ports amateurs ;-)
22:06:49gevaertsAnyway the driver has more bits that need fixing
22:07:23shotofaddsI'm sure it has, I haven't really looked into USB on this at all
22:07:28shotofaddsAnything in particular?
22:07:44gevaertsmostly the nice TODO issues
22:09:00 Quit SirFunk (Read error: 110 (Connection timed out))
22:09:14gevaertsthe driver spends too much time busy-waiting. It needs to use interrupts instead
22:09:37 Quit jgarvey ("Leaving")
22:10:00 Join SirFunk [0] (
22:10:06shotofaddssure, but I'd hope those improvements could wait until the functionaility actually works..
22:11:21*bertrik starts working on charging for e200v2
22:12:05gevaertsof course. It's just that I have this tendency to think that we have a driver for tcc usb, while in reality we have a partially working good enough for debugging one
22:12:07*shotofadds wonders if vitja has any intention of working on this in future
22:13:07shotofaddsgevaerts: indeed. although I'd say it's "just about passable" rather than good enough..
22:13:13shotofaddsUMS is rather unreliable
22:14:06gevaertsMaybe I should try to fix it before tackling things like USB host...
22:15:48shotofaddsgevaerts: I think that depends how useful it would be. Eg. there's not much need for fully working USB on the D2 right now, altough it is very helpful for debugging. If this controller is used on other targets, maybe it becomes more important
22:16:08 Part |amadeus|
22:16:33*shotofadds is away for a while
22:18:31gevaertsshotofadds: I think it's useful for various reasons. We want this driver for the meizus as well, I don't like leaving it in this state, and I've never really worked a FIFO-based driver before. Getting some experience with it will help with other drivers later on (like the AMS)
22:18:45gevaertsbut yes, that's all for targets in progress
22:27:40 Quit perrikwp (" ajax IRC Client")
22:29:10 Join matsl [0] (
22:29:44 Quit midgey ()
22:35:28 Join homielowe [0] (n=homielow@unaffiliated/homielowe)
22:38:57***Saving seen data "./dancer.seen"
22:51:12 Join robin0800 [0] (
22:52:33 Join robin0800_ [0] (
22:52:37 Quit robin0800 (Client Quit)
22:53:13 Quit SirFunk (Read error: 110 (Connection timed out))
22:53:35 Quit robin0800_ (Remote closed the connection)
22:54:22 Join SirFunk [0] (
22:57:18 Join robin0800 [0] (
23:00:05 Quit robin0800 (Client Quit)
23:00:37 Join robin0800 [0] (
23:07:06 Join Thundercloud [0] (
23:11:59 Join yhuang [0] (n=yhuang@unaffiliated/yhuang)
23:15:54 Quit JdGordon| (" ajax IRC Client")
23:16:35 Join miepchen^schlaf [0] (
23:19:13 Join JdGordon| [0] (i=836b0049@gateway/web/ajax/
23:25:32 Join dfkt_dt [0] (n=dfkt@unaffiliated/dfkt)
23:25:55 Join domonoky1 [0] (
23:26:13 Quit dfkt (Nick collision from services.)
23:26:16 Nick dfkt_dt is now known as dfkt (n=dfkt@unaffiliated/dfkt)
23:26:19 Quit domonoky1 (Client Quit)
23:26:33 Join domonoky1 [0] (
23:27:54 Quit domonoky (Read error: 104 (Connection reset by peer))
23:28:08 Nick domonoky1 is now known as domonoky (
23:29:58 Quit robin0800 ("No Ping reply in 30 seconds.")
23:30:19 Join dfkt_dt [0] (n=dfkt@unaffiliated/dfkt)
23:30:26 Join robin0800 [0] (
23:35:24 Quit evilnick (" byeeeeeeeee")
23:35:29 Quit shotofadds ("Leaving")
23:35:39domonokybluebrother: rbutil 1.1 ready for release ?
23:36:50 Join dfkt_ffs [0] (
23:37:05 Quit dfkt (Nick collision from services.)
23:37:22 Quit dfkt_dt (Nick collision from services.)
23:37:29 Nick dfkt_ffs is now known as dfkt_dt (
23:37:33 Nick dfkt_dt is now known as dfkt (
23:37:51bluebrotherdomonoky: yes, I already started building binaries
23:38:04bluebrotherat least I don't know of anything that could stop the show :)
23:38:11*domonoky begins checkout on jdgordons mac :-)
23:40:47bluebrotherhmm. I should update my build machine to Qt 4.5 for at least the windows binary. Might include accessibility fixes.
23:42:58 Quit rasher (
23:43:21 Quit AndyI ()
23:45:22 Quit pyro_maniac ("Leaving.")
23:45:41 Quit tvelocity (Remote closed the connection)
23:46:14NJoinrasher [0] (
23:48:13 Quit yhuang ("Leaving")
23:48:51 Join AndyI [0] (i=AndyI@
23:50:25 Quit miepchen^schlaf (Read error: 60 (Operation timed out))
23:50:59 Join midgey [0] (n=tjross@
23:53:40 Quit robin0800 (Remote closed the connection)
23:55:31bluebrotherBagder: rbutil 1.1 binaries for w32 and linux available here: . Please copy to download server.
23:56:36amiconnrasher: Is there a reason why you didn't remove
23:57:57rasheramiconn: I wasn't sure I hadn't - I thought svn move would... move, and wanted to figure out what happened, then forgot
23:58:27amiconnsvn move moves, but you need to explicitly commit the deleted (source) file
23:58:43amiconnA folder commit doesn't pick up deleted files

Previous day | Next day