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 2009-02-27

00:02:14 Quit xSlack (Read error: 60 (Operation timed out))
00:03:55 Join maxscelerate [0] (
00:04:04 Nick maxscelerate is now known as paulposition (
00:06:10 Quit buk_ ("Quitte")
00:06:42paulpositiongiven that there are scrollwheel targets, scratchpad(?) targets and maybe someday touchscreen targets, how come none ever figured out the h10's simple up-down scrollpad? Hardware problem?
00:07:11BigBambiI don't think it is a case of figuring it out
00:07:16BigBambiIt is just used as buttons
00:07:32 Join xSlack [0] (
00:07:36*linuxstb spots barrywardell
00:07:48 Quit evilnick_7 (" ajax IRC Client")
00:08:46paulpositionI do know that.. Was wondering why.. Two years ago, there was trouble reading the dac and it gave jumpy values so I reckon the problem might still be there..
00:09:00kugelBigBambi: afaik it is a matter of figuring it out
00:09:09 Quit bluebrother (Nick collision from services.)
00:09:14 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
00:09:24BigBambihmmm, I thought it was known, it is just it is a stupid control method
00:09:29kugelmaybe you should try to write some red pixel to the display, that magically helps on the fuze/e200 too
00:09:56BigBambiAs in I thought getting positions from it wasn't a problem
00:10:30kugelpaulposition: nothing changed in this regard, so expect it to be still the same issues
00:10:30n17ikhhow does one write a specific bootloader to a sansa?
00:10:34BigBambiah well
00:10:39gevaertsrasher: can you add your findings to FS #9957? I could try as well, but you're the one who saw the problems first hand...
00:10:41kugelNobody works on it, so..
00:11:05rashergevaerts: Will do
00:11:19 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
00:11:21rasherThe playlist size didn't seem to matter after all
00:11:25 Quit xSlack (Read error: 60 (Operation timed out))
00:15:57 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:15:57 Quit pixelma (Nick collision from services.)
00:15:59 Quit amiconn (Nick collision from services.)
00:16:00 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:16:09 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:16:18 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:16:28 Quit bluebrother ("leaving")
00:16:40n17ikhdo I use the sansa firmware updater?
00:16:43 Join midgey_ [0] (
00:18:20paulpositionIs there some button to press or whatnot to plug usb in *without* getting into disk mode (ie, keep the player running on usb charge)?
00:18:29n17ikhhold select, I think
00:18:36paulpositionI'll try
00:19:16paulpositionbingo. ty
00:21:51linuxstbn17ikh: To upgrade the Rockbox bootloader, do "sansapatcher -a filename.mi4"
00:22:00n17ikhoh, ok
00:22:05 Quit thegeek (Read error: 104 (Connection reset by peer))
00:22:16 Join thegeek [0] (
00:22:18n17ikhwriting the file pp5022.mi4 to the / directory and then booting into the OF works too, right?
00:22:40linuxstbWell, it it will, but you will overwrite the OF.
00:22:40n17ikhseems to have worked for me...
00:22:52n17ikhOh. did I break it? >_>
00:22:54linuxstbs/it it will/it will work/
00:23:05linuxstbYes ;)
00:23:16n17ikhhow do I fix it? just run sansapatcher on it?
00:23:16linuxstbAlthough now you have Rockbox USB mode to rescue you
00:23:56n17ikhactually, yeah, I don't care that much
00:24:23linuxstbI think you will need to use recovery mode to restore the OF, then run sansapatcher -a
00:24:34linuxstbOr just forget about the OF
00:24:48n17ikhyeah, I might do that
00:25:34n17ikhso how did that work? I overwrote the OF *and* the bootloader with just the new bootloader?
00:25:49kugelonly the OF
00:26:01n17ikhbecause it boots into rockbox on usb plug now, like it should
00:26:02kugelthe OF bootloader is still present
00:26:06n17ikhso the old bootloader is gone
00:26:13n17ikhthe old rockbox bootloader, that is
00:27:12kugelhm. I guess the OF bootloader loads the old rockbox bootloader, which in turn tries to load the OF (which you replaced by the new rockbox bootloader)
00:27:14 Join FOAD_ [0] (
00:27:30n17ikhthat's... convoluted
00:27:40kugeleasy fix: replace the OF bootloader too ;)
00:28:04 Quit paulposition ()
00:28:24n17ikhI think I'll get it in recovery mode and fix it
00:28:47kugelwell, do what you want
00:29:01kugelI know a way to reduce the boottime down to <2 seconds
00:29:08kugel~2s rather
00:29:55n17ikhhow is that?
00:30:09kugelsansapatcher -bl
00:30:35kugelbut be careful, it can mess all up, so that you'd sansapatcher, if you pass the wrong file
00:30:50n17ikhwhat do I replace the bootloader with?
00:31:14 Quit flydutch ("/* empty */")
00:31:26kugelrockbox bootloader
00:32:00n17ikhoh, ok
00:32:16 Quit mcuelenaere ()
00:32:20kugelbut not the PP5024.mi4 file!
00:32:53 Quit midgey (Read error: 113 (No route to host))
00:32:54 Quit LambdaCalculus37 ("foot")
00:33:04n17ikhso pp5022.mi4 is the rockbox bootloader... but it isn't?
00:33:39kugelit's the Rockbox bootloader, yes
00:33:58kugelbut it has some extra stuff with it which would break it
00:34:01 Join gregzx_ [0] (
00:34:06 Quit gregzx_ (Client Quit)
00:34:41 Join robin0800 [0] (
00:34:54***Saving seen data "./dancer.seen"
00:35:07kugelfor sansapatcher -bl you need the plain bootloader (bootloader.bin)
00:36:41 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
00:36:48n17ikhok, I see
00:37:53 Quit gregzx (Read error: 60 (Operation timed out))
00:37:56 Quit FOAD (Read error: 110 (Connection timed out))
00:37:57 Nick FOAD_ is now known as FOAD (
00:38:11n17ikhwhere can I find bootloader.bin?
00:38:59kugelIt's not distributed, you need to compile it yourself
00:40:05n17ikhso that means setting up a cross-compile environment.
00:40:20 Quit maddler ("Lost terminal")
00:40:23n17ikhwell, forget it
00:43:39 Quit barrywardell ()
00:44:12 Join cbecht [0] (
00:45:00 Quit perrikwp (" ajax IRC Client")
00:45:15*linuxstb wonders why whoever added the -bl option used the .bin version, not .mi4 like everything else
00:45:38n17ikhnow sansapatcher won't run anymore because I don't have the sansa OF
00:45:45n17ikhit thinks it's the old-style rockbox install
00:45:54linuxstbWell, it is.
00:46:06n17ikhneeds more force option
00:46:22linuxstbWhat are you trying to do?
00:46:31n17ikhwrite a new rockbox bootloader
00:46:57linuxstbI thought you just installed the new bootloader?
00:47:02n17ikhyeah, over the OF
00:47:14n17ikhI want it installed where the bootloader should be
00:47:25n17ikhbecause I planned on replacing the OF
00:47:25cbechtI have a sansa e260v2. I checked out rockbox from svn and built firmware for it. But I don't know how to dot he actual install. I can't find install instructions for the v2 since it isn't officially supported yet. Any tips?
00:47:33linuxstbn17ikh: So you have the bootloader.bin file now?
00:47:41n17ikhno, that's not what I'm trying to do.
00:48:06linuxstbThen I don't understand. The OF doesn't exist any more - hence the error from sansapatcher.
00:48:08n17ikhI'm replacing an old rockbox bootloader with a new rockbox bootloader, I'm leaving the OF bootloader alone.
00:48:29linuxstbUnless you've restored it.
00:48:40n17ikhwell, I tried
00:48:51n17ikhsansapatcher said "the OF doesn't exist!" and refuses to do it.
00:49:06n17ikhusing the -of option
00:49:14linuxstbHelp me catch up. You firstly installed the new bootloader by copying the .mi4 file to drive and updating in the OF?
00:49:30linuxstbThen what did you do?
00:49:44cbechtsansapatcher gives me this:
00:49:47cbecht[INFO] Scanning disk devices...
00:49:47cbecht[ERR] No E200s or C200s found, aborting
00:49:47cbecht[ERR] Please connect your sansa and ensure it is in UMS mode
00:50:18linuxstbn17ikh: I already said that that method of installing the bootloader overwrites the OF. That's why sansapatcher is telling you it doesn't exist.
00:50:18cbechtThe device is connected and in UMS mode. I've tried with it mounted to /sansa and unmounted.
00:50:27n17ikhyeah, I know
00:50:29linuxstbcbecht: Are you root/Administrator?
00:50:38cbechtlinuxstb: Yep.
00:50:45n17ikhI realize that the OF no longer exists. I've downloaded it again and am trying to replace it.
00:51:00linuxstbYou need to use recovery mode for that.
00:51:12cbechtDoes it require libusb or soemthing?
00:51:12n17ikhsansapatcher -of doesn't work?
00:52:29 Join xSlack [0] (
00:53:04kugeln17ikh: you can also drop the PP5024.mi4 into recovery mode
00:53:18kugelI mean the rb bootloader
00:54:21n17ikhwow, the OF takes a long time to boot
00:54:26n17ikhusing the OF bootloader
00:54:35cbechtI'm using the sansapatcher linked from the e200v1 install instructions. Is there a later one that would understand the v2 hardware? Or should I be building that from svn?
00:54:35 Quit shotofadds ("Leaving")
00:56:10 Quit ender` (" Where there's a will, there's an inheritance tax.")
00:57:51linuxstbcbecht: sansapatcher doesn't work (and will never work) on the v2s.
00:59:18cbechtlinuxstb: Ok. Well, my problem is that I am building svn firmware for my v2 and I can't find install instructions. So I was using the instructions from the v1 hoping they were sufficiently similar. Is there a better source of instructions? Or could you tell me how to get the rockbox bootloader installed? (I already have the .rockbox directory on the player)
01:01:15linuxstbYou need to use the mkamsboot tool in utils/mkamsboot/ - this combines the Rockbox bootloader (bootloader-xxxx.sansa) with an original firmware file to create a new firmware that you install on the device.
01:05:59cbechtAh Ha!
01:06:08cbechtThat was the missing bit of information I needed. Thanks
01:07:04 Quit Lss (Read error: 104 (Connection reset by peer))
01:07:15 Join Lss [0] (
01:21:23Unhelpfulamiconn: and shift back to center when scrolling stops? the only problem i see there is if the number of slides that can be rendered on screen exceeds the number that can be cached, since it could turn into "the center slide and two to the right" while scrolling, and then shift to the center and the two adjacent to it when scrolling stops... and the cache will basically *always* be too small on most color targets that i've looked at, if
01:21:24Unhelpful it's restricted to the plugin buffer.
01:21:36 Quit T0paz ("Leaving")
01:22:50 Join webguestington [0] (n=424c70e3@gateway/web/cgi-irc/
01:23:00webguestingtonHello there.
01:23:52webguestingtonQuestion: What might be causing my ipod to appear in ubuntu as a read-only file system?
01:24:05linuxstbFilesystem corruption.
01:24:10 Quit Llorean (Read error: 113 (No route to host))
01:24:26webguestingtonwhat should i do? ? ? ?
01:24:37linuxstbRun fsck.vfat
01:24:47webguestingtonin terminal right?
01:25:16linuxstbYes. Google can probably help you with the details (I don't know them off-hand)
01:25:27webguestingtonokay. thanks man.
01:25:42linuxstbYou want to run it on the second partition - i.e. /dev/sdX2
01:30:22webguestingtonso the format of the command would be like..
01:30:44webguestingtonfsck.vfat -/dev/sdX2 ?
01:32:46n1swebguestington: run 'mount | grep -i ipod' to fid out which device it is
01:33:05n1s(I *think* that shoudl get it)
01:33:30webguestingtonthanks dude! that helped
01:34:05webguestingtondamn permission denied. should i just do sudo?
01:34:28 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
01:34:30webguestingtonschweet. progress.
01:35:12webguestingtonokay it says this: FATs differ but appear to be intact. Use which FAT?
01:35:19webguestington1) Use first FAT
01:35:24webguestington2) Use second FAT
01:35:43 Quit MethoS- (Remote closed the connection)
01:35:45linuxstbToss a coin and pray...
01:35:47n1swebguestington: this is not really rockbox related, ask google
01:35:55webguestingtonhahah.. okay thanks guys.
01:36:10 Quit Zagor ("Clint excited")
01:38:03 Quit n1s ("Lämnar")
01:39:11 Quit Thundercloud (Remote closed the connection)
01:43:04 Quit cool_walking_ (" ajax IRC Client")
01:44:11webguestingtonso uh.. i'm pretty excited about watchmen, are you guys excited too?
01:44:47krazykitwebguestington, this is an on-topic channel. please read the guidelines linked in the /topic
01:44:59webguestingtonyes sir.
01:45:02 Quit webguestington ("CGI:IRC")
01:47:59 Join fdinel [0] (
01:50:19 Quit MTee (Read error: 110 (Connection timed out))
01:54:34 Join Llorean [0] (
02:01:13 Join robin0800_ [0] (
02:04:53 Quit robin0800 (" - Chat comfortably. Anywhere.")
02:05:58 Quit dfkt (Read error: 54 (Connection reset by peer))
02:06:29 Join robin0800 [0] (
02:07:33 Quit robin0800 (Remote closed the connection)
02:07:59 Quit robin0800_ (" - Chat comfortably. Anywhere.")
02:08:02 Part toffe82
02:08:33 Join robin0800 [0] (
02:12:50 Quit perrikwp (" ajax IRC Client")
02:15:41 Quit moos ("ciao ciao")
02:19:58 Join jordoex [0] (
02:22:41 Quit robin0800 (Remote closed the connection)
02:27:07 Join evilnick [0] (i=620ec27e@gateway/web/ajax/
02:33:09kugelFlynDice: ping
02:34:57***Saving seen data "./dancer.seen"
02:36:58 Quit rakslice_work_ (Read error: 113 (No route to host))
02:49:27JdGordondid someone change the default value for charge on usb connect?
02:49:47LloreanJdGordon: Huh?
02:50:14LloreanDreamlayers reported charging isn't actually set up properly for iPod USB charging, I think.
02:51:13JdGordonim talking about h300 though... was the default always to not chage on usb connect?
02:51:30LloreanYeah, don't know why.
02:51:41LloreanIs the Gigabeat S expected to charge over USB?
02:52:33JdGordonok wierd.. i thought it was always enabled... that explains wy it didnt chagre last night
02:52:36linuxstbLlorean: I think it should in Rockbox itself, but not the bootloader.
02:52:55JdGordonits only a PP thing that it could be problematic right?
02:53:54Lloreanlinuxstb: It didn't *seem* to be charging in my car, but I'm not sure. It showed an icon next to the battery in the status bar, but the WPS didn't show charging, and when I reached my destination I don't think the battery level was any higher (but the Beast seems to have unreliable enough readings I'm not certain)
02:54:22linuxstbLlorean: I'm just going by what (IIRC) jhMikeS has said.
02:54:59JdGordonLlorean: check the seting... thats what my h300 was doing last night and i couldnt figure out why it was still flat this morning
02:55:08LloreanJdGordon: I think the h300 is the only one *with* a setting
02:55:28linuxstbLooking at the source, jhMikeS may have enabled it for other targets.
02:55:30 Join robin0800 [0] (
02:55:39LloreanWhere's it located?
02:56:04JdGordonah yes, you're right.. so why is it disabledby default?
02:56:24LloreanJdGordon: I think it has something to do with how the OF works?
02:56:26linuxstbAh no, it seems only the H300 has HAVE_USB_CHARGING_ENABLE defined.
02:56:28 Quit midgey_ (Read error: 110 (Connection timed out))
02:56:35LloreanThere's a setting in the OF for it, so for some reason we have a setting too.
02:56:49JdGordonthats a bit... odd
02:57:38LloreanI seem to recall some mention of it being a switch between 100 and 500, and if it's at 500 some PCs won't recognize it?
02:58:00LloreanI asked about it once out of curiosity, but it was a long time ago, and once again I could be confusing something else.
02:58:39linuxstbIt all comes from here -
02:59:33linuxstbIt seems the H300 can draw too much power from an unpowered hub (to quote LinusN)
02:59:45 Join midgey [0] (
03:00:45 Join spidermang [0] (n=424c70e3@gateway/web/cgi-irc/
03:00:58spidermanghey uhh my ipod doesn't appear to be charging through usb
03:01:57spidermanghow come you think?
03:04:48kadobanJdGordon: Have you gave any more thought to FS #9922 (statusbar not drawing in plugin menus) by any chance? i've checked into it a bit more and I don't see a much better way to fix the problem. It seems to be either force drawing it, or setting it changed (they both work).
03:05:24JdGordoni havnt looked into it yet, coding on thi laptop isnt much fun...
03:05:42linuxstbspidermang: I think that's a known issue. Which ipod?
03:06:00kadobanokay, no worries
03:08:12JdGordonkadoban: is it every plugin (which uses one of the 2 standard menus?)
03:09:05kadobanJdGordon: i believe so. It was every one where it was easy to get the menu to show up and disappear at least...let me look.
03:09:33spidermang4g greyscale
03:10:00spidermang40 gigs.
03:10:30JdGordonhmm.. start seems to be fubar
03:11:43*JdGordon slaps head... checking or this while charign is not a good idea
03:12:52kadobanJdGordon: yeah, it's all of them it appears
03:13:30kadobanstar has a different graffical bug, but i think that's separate (it draws the statusbar over the menu title for some reason)
03:18:12 Quit spidermang ("CGI:IRC")
03:19:53 Quit evilnick (" ajax IRC Client")
03:26:08 Join bs66_ [0] (
03:27:48 Quit robin0800 (Read error: 110 (Connection timed out))
03:32:18FlynDicekugel: Saw the patch I'll try it later tonight
03:32:44 Join midijunkie [0] (
03:34:18midgeybah, someone broke building the sim on os x again
03:46:04 Quit bs66_1 (Read error: 110 (Connection timed out))
03:53:06 Join CaptainKewl [0] (
03:54:19 Join Platypus-Man [0] (
03:55:36Platypus-Manhello people, is the Sony NWZS738FB supported, or are there any future plans to support it?
03:56:52 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
03:56:52advcomp2019no and there is no plans too
03:57:07Platypus-Manthanks for clearing it up
03:57:10 Part Platypus-Man
03:57:11saratogai guess you could check the wiki to see if anyone has ever expressed any interest in the device
03:57:48kugelsaratoga: hi
03:58:05kugelany updates on the compiler/mp3?
03:58:29saratogakugel: i'll probably take a look at it this weekend
03:58:34saratogaright now i'm really busy with research
03:59:11saratogadid you ever benchmark on newer gcc versions?
04:00:23 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
04:03:30kugelsaratoga: no, but given that my experiences were similar I wouldn't expect btter performace than 4.4.0
04:04:14kugelI kept the compilers though (the binaries), so I should be able to compile new builds
04:04:15 Quit rocko ("Leaving")
04:06:23saratogakugel: someone reported some codecs being faster in 4.2 ages ago
04:06:39saratogaso maybe theres some chance
04:07:01saratogabut i guess it would be most interesting to look at coldfire and the gigabeast since gcc has more improvements for those processors
04:09:43kugelI have the m68k compiler built too
04:10:04saratogai need to get one of those players to test with
04:14:41 Join blkhawk- [0] (
04:15:23 Quit Llorean (Read error: 104 (Connection reset by peer))
04:15:41 Join Llorean [0] (
04:15:48 Part Llorean
04:29:56 Quit kugel (Read error: 110 (Connection timed out))
04:30:40 Quit blkhawk (Read error: 110 (Connection timed out))
04:31:39 Nick blkhawk- is now known as blkhawk (
04:32:40 Quit midijunkie (Read error: 110 (Connection timed out))
04:35:00***Saving seen data "./dancer.seen"
04:47:45midgeybuilding the sim on os x has been broken for over a month :\
04:48:02midgeyr19775 was the last one working
04:52:46midgeyUnhelpful: ping
04:54:35 Join low_light [0] (i=ad58bb86@gateway/web/ajax/
04:54:45midgeythe loader initialized plugin api changes seem to disagree with mac
04:55:48Unhelpfulmidgey: any idea why?
04:56:21midgeyi'm trying to check it out right now, but it's a bit of a huge commit
04:56:27Unhelpfuli sadly don't have a mac to test on, but my primary dev platform is linux, and sims work just fine, so it's some bsd-ish or osx-ish distinction...
04:56:30midgeylet my pastebin the error
04:56:48midgeyosx seems to break things all the time...
04:58:10 Join webguestington [0] (n=424c70e3@gateway/web/cgi-irc/
04:58:26webguestingtonhey you guys.. my ipod's not charging on usb
05:00:02webguestingtonit just started doing this.. and before that, i wasn't able to do the trick where you hold Menu and charge it without mounting it.
05:00:23webguestingtonya know?
05:00:57webguestingtonmidgey, does that mean there's no answer?
05:01:23Unhelpfulwebguestington: i don't have your answer, and maybe he doesn't, either. be patient.
05:01:36midgeyyou haven't really given people much time to respond
05:02:00midgeywhat type of iPod are you using?
05:02:04webguestingtonoh okay i gets it. i'm sorry guys.
05:02:11webguestington4th gen greyscale
05:02:41midgeyand which version of rockbox (svn revision)?
05:03:04webguestingtonlatest release.
05:03:30midgeya revision number is preferred. latest release is ambiguous
05:03:57midgeysome people mean 3.1, some mean a version they think is latest but is slightly out of date, etc
05:04:00Unhelpfulmidgey: what compiler are you using?
05:04:25midgeygcc version 4.0.1 (Apple Inc. build 5490)
05:04:27low_lightJdGordon: is there an easy way to not display the statusbar on a remote, but still have it on the main screen? I kind of have the m:robe 100 remote working, but because of it's LCD size (79x16) the status bar doesn't fit.
05:06:10midgeyUnhelpful: removing the attribute for the plugin_api in plugin.h seemed to help
05:06:14Unhelpfulmidgey: it *looks* like whatever gcc puts in the assembly to indicate that the variable has an assigned section doesn't agree with your assembler
05:07:18 Quit miepchen^schlaf (Read error: 101 (Network is unreachable))
05:07:32webguestingtoni'm fairly certain it's 3.1.
05:08:20webguestingtonmy ipod is dead, i was going to check rockbox info on there.. not sure how else i'd confirm.
05:11:49webguestingtonit's slightly aggravating because it just now started doing this..
05:12:22Unhelpfulthe attribute is not strictly needed, i *think* for sim? it solves an init problem on target when iram is used.
05:13:28midgeyi think ATTR_ICODE etc are blank for sims
05:14:44midgeywebguestington: my 4G seems to be charging fine through usb, but I'm on a more recent verision
05:15:37Unhelpfulmidgey: by "seems to help", do you mean that it works if you remove the attribute?
05:15:57midgeyr19777 was able to build successfully
05:16:07midgeyi'm checking with latest svn
05:16:48webguestingtonshould i get the latest daily build and try that?
05:17:17midgeyit would probably be useful
05:17:30midgeyyou said you've never gotten it to charge through usb?
05:20:30webguestingtonnah it worked fine up until very recently, suddenly stopped working.
05:21:13 Quit saratoga ("CGI:IRC (EOF)")
05:21:31midgeyand you hold menu (quickscreen should load) and insert the cable, then release menu?
05:21:43 Join sarixe [0] (
05:21:47midgeyand no charger icon appears in the status bar?
05:22:20webguestingtonman.. i may have just been doin it wrong then.
05:22:28webguestingtonnow it's just dead though lol
05:22:32Unhelpfulmidgey: you need to make sure that such a change is #ifdef'd for sim. it *will* break things on targets.
05:23:00midgeyright, osx builds for targets without complaint so i wasn't going to touch that
05:23:41midgeyis the attribute technically correct for the sim? i can probably come up with some ifdef magic to only change for osx
05:29:17 Quit ChanServ (
05:29:17 Quit cool_walking_ (
05:29:17 Quit tmzt (
05:29:17 Quit bluefoxx (
05:29:17 Quit z35 (
05:29:17 Quit GodEater_ (
05:29:17 Quit tarbo (
05:29:17 Quit advcomp2019 (
05:29:17 Quit lightbulbjim (
05:29:17 Quit keby (
05:29:17 Quit soap (
05:29:17 Quit yosafbridge (
05:29:18 Quit sarixe (
05:29:18 Quit FOAD (
05:29:18 Quit gromit`` (
05:29:18 Quit Nico_P (
05:29:18 Quit blithe (
05:29:18 Quit Galois (
05:29:18 Quit tchan (
05:29:18 Quit n17ikh (
05:29:18 Quit Neovanglist (
05:29:18 Quit jhulst (
05:29:18 Quit gibbon_ (
05:29:18 Quit scorche|sh (
05:29:22webguestingtonat which point should i release menu? is it too late by the time i see the usb cable graphic?
05:29:41NJoinsarixe [0] (
05:29:41NJoinFOAD [0] (
05:29:41NJoingromit`` [0] (
05:29:41NJoinNico_P [50] (n=nicolas@rockbox/developer/NicoP)
05:29:41NJoinblithe [0] (
05:29:41NJointchan [0] (n=tchan@lunar-linux/developer/tchan)
05:29:41NJoinn17ikh [0] (n=n17ikh@
05:29:41NJoinNeovanglist [0] (i=Neovangl@
05:29:41NJoingibbon_ [0] (n=joel@
05:29:41NJoinGalois [0] (
05:29:41NJoinscorche|sh [50] (n=scorche@rockbox/administrator/scorche)
05:29:41NJoinjhulst [0] (n=jhulst@unaffiliated/jhulst)
05:30:16 Quit Zoxc ()
05:30:45NJoinChanServ [0] (ChanServ@services.)
05:30:45NJoincool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
05:30:45NJointmzt [0] (
05:30:45NJoinbluefoxx [0] (
05:30:45NJoinz35 [0] (
05:30:45NJoinGodEater_ [0] (i=c2cbc962@rockbox/staff/GodEater)
05:30:45NJointarbo [0] (n=me@unaffiliated/tarbo)
05:30:45NJoinadvcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
05:30:45NJoinlightbulbjim [0] (
05:30:45NJoinkeby [0] (
05:30:45NJoinsoap [50] (n=soap@rockbox/staff/soap)
05:30:45NJoinyosafbridge [0] (
05:30:45Mode"#rockbox +o ChanServ " by
05:30:59 Quit tarbo (SendQ exceeded)
05:31:29Unhelpfulmidgey: it's only telling the compiler and linker that the variable should be placed in a particular section of the object file. it *should* be valid on sim on osx. :/
05:31:43 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
05:32:13midgeythe build environ tends to do crazy things on osx
05:32:37midgeythere are other hacks in rockbox to get it to build
05:32:48 Join tarbo [0] (n=me@unaffiliated/tarbo)
05:32:52 Quit Horscht ("Verlassend")
05:34:52midgeyfor example, look around in the bitmaps folders and you'll see plenty of osx.dummy.bmp files
05:44:06 Quit webguestington ("CGI:IRC (EOF)")
05:45:36 Quit Nico_P (Remote closed the connection)
05:48:35 Quit Aurix_Lexico (Remote closed the connection)
05:49:22 Quit Kopfgeldjaeger (Remote closed the connection)
05:51:15midgeyJdGordon: around?
05:52:01midgeyUnhelpful: thoughts on this patch? ->
05:52:32midgeywith those changes, it's possible to build the sim on mac again
05:53:24 Quit _lifeless (Read error: 110 (Connection timed out))
05:53:45Unhelpfulthat seems generally sensible, but i'm not sure i follow how it makes sure that definition doesn't happen on sim, on osx?
05:54:00Unhelpfuli'm assuming that's "outside" the patched portion?
05:54:48midgeythe changes to config.h set DATA_ATTR to blank just like the other _ATTR macros
05:55:31Unhelpfuli was about to say it should be defined in plugin on all targets, but i don't *think* it matters on sh1
05:56:11Unhelpfuli'd like an opinion from somebody who knows better about our hardware, and about iram init, but i'm *pretty* sure this is ok :)
05:56:37midgeyi figured it made sense to keep all the ATTR defines in one place so i went with config
05:57:53midgeyif anyone's opposed to me committing those changes, speak up over the next few hours
05:58:32midgeyan alternative would be to use #ifdef __APPLE__ but I didn't want to add even more ifdefs
05:59:24 Quit low_light (" ajax IRC Client")
06:02:05 Join FOAD_ [0] (
06:07:16JdGordonmidgey: yeah, hey
06:07:17midgeybah, ld on mac doesn't understand −−version
06:07:19 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
06:07:23*midgey smacks apple
06:08:08JdGordonmidgey: are you having the same problem with the osx asim as lostlogic? he rekons its gcc bugs
06:08:27midgeyerr, what problem was he having?
06:09:15JdGordoni cant rember is evil on mac
06:10:05midgeysh can't be build without modifying it to use make -r
06:11:49midgeythere's some status bar redraw problems in the debug menu
06:12:10midgeyprobably not critical to fix or anything, but might lead to hints on other problems
06:13:40midgeyi was messing around in the "View Battery" screen and upon exit the status bar is redrawn pretty late
06:13:51 Join nowbehave [0] (n=4a4b4933@gateway/web/cgi-irc/
06:14:00midgeyit's more noticable if you go to the power status screen
06:14:41midgeythe status bar is replaced with "Power status:" :P
06:15:25JdGordonyep, i see it...
06:15:40midgeystays like that until stepping back into the system menu or until the lcd goes to sleep and wakes back up
06:16:13Unhelpfulanybody handy who knows tagcache much? i'm looking at the function in pictureflow that builds the cache of album names, and trying to get rid of that cache, since it's a bunch of string data, could save some space, etc. create_album_index is storing the string, the "index" of the album (not provided by tagcache, but by counting the number of albums retrieved so far), and the result_seek member of the tagcache_search structure
06:16:41Unhelpfuli know that tagcache knows nothing of the index, and that can't be used to fetch the album name... can the seek value be used?
06:17:55JdGordonUnhelpful: Slasheri is the db person.. dont think anyon other than him knows its internals...
06:18:07midgeyyou could be the first!
06:19:12 Quit FOAD (Read error: 110 (Connection timed out))
06:19:12 Nick FOAD_ is now known as FOAD (
06:19:50Unhelpfulmidgey: careful, there, i ended up rewriting pretty hefty chunks of pictureflow
06:19:59nowbehavethis question is concerning a "usa spec" pa15-toy, its a interface for the toyota, connects to the radio and makes the ipod seem as if its an external cd changer, the person who posted the question on March 16, 2007 asked if rockbox would work with the usa spec device, the answer was "Nope, the Apple Accessory Protocol is not supported, thus you can not control a Rockboxed iPod from your head unit. If you head
06:20:18midgeytagcache could probably use some love ;)
06:20:28midgeythen go after playback.c
06:21:07JdGordonnowbehave: you gt cut off... but assuming your question is will it work, it might now...
06:21:22JdGordonmidgey: haha you sadistic bastard :D
06:22:36nowbehavejdgordon, so would a smart way of going about it would be to just buy the usa spec device, install it, and see if it works?
06:23:57JdGordonwell.. thats the only way to test it... nwe wont guarentee it will work though (unless somone has already reoprted it working)
06:24:18JdGordonoh.. it actuall might not
06:24:41Unhelpfulit all depends on which parts of the protocol it needs
06:24:42JdGordonit deends if it takes control of the ipod or just lets you skip/seek ike a remote
06:26:07 Quit BHSPitMonkey (
06:26:07 Quit cbecht (
06:26:07 Quit killan (
06:26:07 Quit scorche (
06:26:15nowbehavei think the interface just lets you skip around like a remote
06:26:19NJoinBHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
06:26:19NJoincbecht [0] (
06:26:19NJoinkillan [0] (
06:26:19NJoinscorche [50] (n=scorche@rockbox/administrator/scorche)
06:26:31nowbehavethere is text too, but i dont need that, just need power and sound
06:28:32 Join rocko [0] (
06:28:36JdGordonhow much is it?
06:28:55 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
06:34:56nowbehaveits 139, i am reading a forum now, it talks about this kinda it says teh status is closed and it is assigned to no one, thats a not a good thing right?
06:35:03***Saving seen data "./dancer.seen"
06:35:10nowbehavejdgordon : $139
06:37:06kadobannowbehave: look at the reason for closing ("Accepted"). that means that it was accepted to svn and is now in rockbox
06:38:47Unhelpfulit would appear that i can use the value saved for the purpose of finding the albums tracks to fetch its title from the tagcache, as well. i'm not sure if the performance will be reasonable... i'll get my local stuff in order and give it a try :)
06:41:07nowbehavethanks jdgordon, kadoban, and unhelpful,
06:42:58midgeynow why would ld -v print to stderr instead of stdout...
06:43:53 Quit nowbehave ("CGI:IRC (EOF)")
06:48:53Unhelpfulthe result field of a tagcache_search structure is char *, rather than an array in the structure - i assume i need to strcpy if i want to keep the contents, as it's allocated from a buffer somewhere in core?
06:57:38 Quit rocko ("Leaving")
07:03:11*midgey adds another osx hack
07:15:52Unhelpfulwe appear to be working without an album names cache, just a reserved string the size of the longest album name, and an array of albums and their seek values :D
07:16:34 Join fml [0] (n=4fd3ca24@gateway/web/cgi-irc/
07:18:22fmlHow about this for the "natural sorting" setting: Setting "Sorting of file names with (or: containing) numbers" with the options "Smart: 1_One, 2_Two, 10_Ten" and "As Text (or: Plain, or:Dumb): 1_One, 10_Ten, 2_Two"?
07:19:40 Quit fml (Client Quit)
07:20:22kadobanthat looks overly verbose to the point of being confusing
07:22:10*Unhelpful still prefers "numeric" sort, as there is absolutely nothing else especially "natural" about the implemented method :/
07:22:51kadobanit's natural in the sense of the natural numbers, i'm pretty sure. it's known as that in USA somewhat commonly
07:23:22kadobanbut yeah, i thought that the example list things primary purpose was to avoid having to call it anything in particular...
07:23:47 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:27:02 Quit CaptainKewl (Read error: 60 (Operation timed out))
07:27:23kadobanI'm working on adding Playback Control to all plugins possible, and i'm down to the ones that just don't have a menu at all. Here's what I'm thinking: instead of just quitting for many of them, instead go to a menu with Playback Control and Quit. Does that seem horrible?
07:28:38JdGordonhave you got a lit of the plugins wich dont use a stanard menu?
07:29:38kadobanthe ones that use a non-standard menu, or the ones that don't have any menu at all? (either way, it's in FS #9953). the only ones included are the ones i'm considering or have already added playback control to though
07:30:00kadobanthe second bunch are "doesn't have a menu", and the third bunch are "has a non-standard menu"
07:30:43 Quit perrikwp (" ajax IRC Client")
07:31:59Unhelpfulkadoban: pictureflow isn't on your list?
07:32:24kadobanUnhelpful: oh, i don't know if i even considered pictureflow...let me look at it
07:33:08Unhelpfulyou don't list it under any of your categories. it lacks playback control, and i'm in the process of making it actually work during playback (it doesn't work, now, if buffer is full)
07:33:45 Join einhirn [0] (
07:33:53kadobanyeah, i left some out for various reasons. i think pictureflow was purely laziness...i don't use album art, so it wouldn't run and i didn't get around to trying it
07:34:31kadobanif it works during playback though, i'd definitely work something up for it
07:35:02 Quit xSlack (Remote closed the connection)
07:35:04kadoban(once i does i mean)
07:35:54 Join xSlack [0] (
07:38:35 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
07:39:03kadobananyway, for the menu-instead-of-quiiting-directly thing, it can be made fairly unobtrusive by the standard move-out-of-menu buttons being mapped to quit unless Playback Control has been used since the menu opened. if anyone thinks that the idea is particularly stupid, let me know in the next hour or so...otherwise i'll code up a patch.
07:40:15Unhelpfulmaybe just have a menu with "playback control" and "quit" in it, for plugins that lack any menu right now?
07:40:50kadobanyeah, that's the idea. the other ones i've either just added a new menu item or converted them to standard menus and added an item
07:41:46 Join zimba42 [0] (
07:44:30 Part zimba42
07:47:50 Join Bagderr [241] (n=daniel@rockbox/developer/bagder)
07:50:07 Nick Bagderr is now known as B4gder (n=daniel@rockbox/developer/bagder)
07:57:44 Quit perrikwp (" ajax IRC Client")
07:57:59 Join perrikwp [0] (n=18ac0c41@gateway/web/ajax/
07:58:57 Quit BHSPitMonkey ("Ex-Chat")
08:08:19midgeyany thoughts on the DATA_ATTR patch I posted earlier?
08:09:49amiconnUnhelpful: Really? I can't imagine that a target won't be able to cache at least 5 slides.
08:10:49amiconnBut of course the cache center shifting while scrolling would be an optimisation, and trying to cache at least all visible slides would have priority
08:13:53amiconnAnother important point (when the cache is large enough) is to make sure the disk won't be spinning up every time you scroll
08:18:56 Join tvelocity [0] (
08:22:30 Quit Seed ("cu, Andre")
08:30:16 Join Rob2223 [0] (
08:31:45 Join ender` [0] (
08:31:51 Join petur [50] (n=petur@rockbox/developer/petur)
08:33:02 Join nuonguy [0] (
08:33:40 Quit sarixe ("Leaving")
08:35:07***Saving seen data "./dancer.seen"
08:35:19 Join MTee [0] (n=MTee@
08:39:09 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
08:39:17JdGordonkadoban: midgey: well, i have a fix which should fix every bar problem hen its displayed with lists... but its very bandaid and not very nice in its current form.. but it works
08:41:13midgeywell if you don't come up with a better fix, it will probably be at least worth committing to the 3.2 branch just before release
08:43:59 Join gartral [0] (
08:45:37Unhelpfulamiconn: i think more than the plugin buffer is needed, if you want to scroll more than a few slides without going to disk. based on the average slide size, and reported buffer remaining after init on my beast, there is room for ~9 slides in the cache
08:46:08JdGordongrr, well this fix works, but now the statusbar is displayed over the logo on boot... is that bad?
08:46:11 Quit cool_walking_ (" ajax IRC Client")
08:46:12amiconnBut 9 slides are already more than the visible ones
08:46:22amiconnJdGordon: yes
08:46:36JdGordonthts what i thought...
08:48:27 Quit Rob2222 (Read error: 110 (Connection timed out))
08:49:07JdGordonfixed :)
08:49:26gartralwould it be possible to have rockbox drop the USB connection on a long key press or something similar? like when you hold center to keep rockbox usable and not initialise USB, just charge?
08:49:44JdGordonit does that already
08:50:28gartralJdGordon: i mean after USB is initialised, not before
08:50:39midgeyin theory it is possible
08:50:48midgeyi believe it was suggested the other day
08:50:57Unhelpfulyes, but only by very few, so that scrolling one or two slides goes to disk... unless there's extra buffer from core. also, you can get, although it looks a bit cramped, 9 slides on screen on ipodvideo
08:51:01 Quit nibbler (Read error: 60 (Operation timed out))
08:51:19 Join _lifeless [0] (n=lifeless@
08:52:11amiconnDropping the usb connection "by force" would be very unclean
08:52:25gartralno worse than just unplugging the device
08:52:53amiconnWhat could (and probably should) be done is that rockbox exits the usb screen when you disconnect safely
08:53:16Unhelpfulright now it doesn't ever attempt to cache more than the "number of slides" setting * 2 + 1 slides, although if the cache buffer is large enough, more slides may be left cached in the direction you've just scrolled from, since they're cached LRU
08:54:01kadobanthe only reason i ever see that being useful is if rockbox gets confused and still thinks the USB is active when it really isn't. does that ever happen? (it doesn't to me)
08:54:22JdGordonOH FFS!!! part of the bar problem not redrawing is because of the screens which block in their button loops...
08:54:39gartralits an afterthough in response too
08:57:27JdGordonhmm... maybe not...
08:57:29*JdGordon confused
08:58:02*midgey commits some GSoC-related work
08:58:14midgeyvery minimal, don't get too excited
08:59:42 Quit _lifeless (Read error: 60 (Operation timed out))
09:00:02kadobanJdGordon: isn't the problem fairly simple? it seems that there is simply no force drawing of the statusbar when lists open, so unless there is new information the statusbar doesn't actually get drawn. i guess there is code that's supposed to be doing the right thing somewhere that i didn't see/understand?
09:01:09JdGordonkadoban: well, yes and no... its *supposed* to force a redraw when the bar gets toggled which should do it, but it looks like somebad ordering somewhere is stopping it
09:01:27kadobanoh okay, that makes sense
09:01:29JdGordonso yes your fix wil work, but mine is doing the same thing, 1 level further "down"
09:01:59kadobanyeah, i couldn't figure out that bit...if it can be done further down that makes a lot of sense
09:09:00 Join _lifeless [0] (n=lifeless@
09:11:27gartralhmm... i noticed after using rockbox solely without the OF on e200, the battery drops 2-3 percent almost mediately, where as charging till the OF says full, the level stays at 100 percent for a while
09:12:10gartralthis is before i start playing anything, or use any buttons, btw
09:16:18Unhelpfulamiconn: i wasn't sure, but i think midgey's patch here should be ok?
09:17:03Unhelpfulit seems that if iram isn't in use, the problem solved by moving the API pointer to the data section never arises anyway, right?
09:18:16 Join SirFunk [0] (
09:19:05 Quit offset (Read error: 110 (Connection timed out))
09:19:28JdGordonoh bloody hell... im a moron.. i tihnk i see the problem
09:20:38 Join pj188899 [0] (
09:20:49pj188899yo i just have a quick ipod 5th gem question!
09:20:57amiconnUnhelpful: Atm this is correct, although it doesn't hurt to have it always there
09:21:26pj188899anyon here use rock box on a 5thgen ipod?
09:21:26amiconnIt increases the plugin file size by 4 bytes. Memory usage doesn't change at all
09:21:41pj188899anyon here use rock box on a 5thgen ipod?
09:22:03midgeyplenty of people do, just ask your question
09:22:05kadobanpj188899: probably several people :) it's more efficient if you just ask your question and anyone who knows can respond
09:22:19pj188899well how do i add sort by artist?
09:22:39midgeyUnhelpful: see r20124 ;)
09:22:47Lsswhat do you mean by add?
09:22:52Lssisnt it already in?
09:23:08Unhelpfulmidgey: ah, i didn't know you'd already committed it. don't i look silly. :)
09:23:38JdGordonkadoban: i said it would be an easy fix... and it is... :'(
09:23:45JdGordonshould have seen this weeks ago
09:24:13kadobanJdGordon: ahaha, thanks a lot for rooting it out.
09:24:25pj188899its not in mine
09:24:32pj188899its only alphabeticaly
09:24:59 Join n1s [0] (n=n1s@rockbox/developer/n1s)
09:25:10Lsswhich version are you on?
09:25:22Lssare you accessing it from file system or database?
09:25:33Lsshow are your files ordered?
09:25:39pj188899like normaly?
09:25:43pj188899artist i think
09:25:49Lssthe folder structure?
09:25:56Lssthe file names?
09:26:07pj188899it just reorders them file banes
09:26:13kadobanfiles in the File Browser are simply sorted by filename. if you want something more exotic, you need to use the database
09:26:19Lssgive an example of the filename
09:26:46pj188899is at the top
09:26:47Lssthats why there is no way to sort by artist
09:26:51Lsslike kadoban said
09:26:52pj188899in sted of in like N
09:27:08Lssbrowsing by file is normal in this case
09:27:15pj188899oh ok
09:27:21pj188899its just hard as fuck with 600 songs
09:27:26gartralin reversi it says "Gave over. <TEAM> have won" it should be "has won"....
09:27:34pj188899and you dont know who the artist is
09:27:35Lssfile view doesnt access any tags the file might have
09:27:43Lsswhich include the artist name
09:27:52pj188899well how do i access the tags?
09:27:57Lssdude i have around 4000 on mine
09:28:01Lssuse database instead
09:28:30gartralim going to submit a patch to correct that typo, if anyone is intrested
09:28:47Lssgrammer nazi :P
09:28:54Lssjusnt kidding
09:29:04JdGordonmidgey: did you miss a file in your commit? you seem to have removed code only?
09:29:08Lssor just kill the have/has
09:29:15pj188899hold on why is it going over 600?
09:29:17Lssand just leave it as <TEAM> won
09:29:18 Quit SirFunk_ (Read error: 110 (Connection timed out))
09:29:24pj188899 only got 514 songs?
09:29:32Lssmaybe you have more than you think or something
09:29:40pj188899i cant i just reformated...
09:29:43Lssthat probably means some files are not tagged
09:29:50pj188899not tagged in my folder?
09:29:54pj188899i crtl a
09:29:57Lssno tagged in the file
09:30:02pj188899in the folder i was adding
09:30:04kadobani think it counts all files or something like that
09:30:07Lsstags are part of the file itself
09:30:09Lssnot the folder
09:30:16midgeyJdGordon: that's all there is to that commit. it was a small piece that shouldn't have any effect on svn
09:30:25pj188899ok so its done now what?
09:30:34Lssi suggest going through all your songs
09:30:41JdGordonmidgey: ok just checking.... looks odd
09:30:45midgeyi'll try saving the genlang changes and lang structures changes until later
09:30:52kadobanpj188899: have you taken a look at the manual by any chance? rockbox's manual is quite good
09:31:01pj188899yes i have
09:31:06Lssits too wordy for most ppl to be truthful
09:31:28pj188899this is like a major function for me
09:31:34pj188899i have a 30gb ipod
09:31:39pj188899and i can fill it in like 10 seconds
09:31:44Lsslol mine is 80gig
09:31:50Lssi have almost 3 times yoru problems
09:31:52pj188899 i dont want ot fi i can view files by artist..
09:32:25pj188899it dont matter..
09:32:30pj188899just asking why it was doing that
09:32:48Lssif you only have mp3 and dont mind itunes you could sync your ipod under the official firmware
09:32:53Lssso its properly tagged
09:32:58Lssand then fire up database
09:33:16Lssas a added bonus your songs would work both under rockbox and official firmware
09:33:17pj188899idk how
09:33:31pj188899i just added rock box
09:33:31Lssjust use the ipod the usual way with itunes
09:33:41pj188899nah im good i hat itunes..
09:33:45Lssafter you are done you boot to rockbox and initialise the database
09:33:51Lssor update it if its already there
09:34:00Lssthen its not a problem i gues
09:34:09pj188899idk how to boot something ona ipod..
09:34:23kadobanJdGordon: thanks again, that fix seems to work great so far :)
09:34:25Lss-.- its in the manual
09:34:31Lssit explains it faster than i do
09:34:43pj188899did you READ the MANUAL FULLY?
09:35:13Lssto be truthful probably around 40%
09:35:20pj188899ok then
09:35:31Lssi skipped alot of stuff which is either intuitive or not relavent to me
09:35:59pj188899comming data base 1/9?
09:36:59pj188899oh i got it thanks
09:37:20pj188899this is better
09:37:24pj188899thanks alot
09:37:35n1si'd recomend looking at the quick start chapter and the "browsing and playing" one
09:38:14pj188899i am now
09:38:23Lssbut there is merit when ppl say that the controls in rockbox isnt standardised
09:38:23 Quit z35 ("Leaving")
09:38:39Lssthe method of reversing to the previous page doenst exactly stay consistant
09:38:53Lssor exiting a game/application/demo
09:39:17UnhelpfulLss: it's pretty consistent in core, i'd say. plugins all handle keys themselves, so inconsistency there is not remoting surprising.
09:39:33Lssok what about the playlist menu
09:39:35 Nick JdGordon is now known as JdGordon|zzz (
09:40:32gartralok, patch at fs#9960
09:40:41n1syes, rockbox could be improved in this area, there are a few qurkiy button maps
09:40:51pj188899so how do i turn this off now?
09:41:01pj188899is there a restart ?
09:41:09pj188899the ipod itself i mean sorry
09:41:25Lsshit and hold play
09:41:33n1spj188899: that is also in the manual...
09:41:45pj188899yeah your to late..
09:41:48Lssdont bother turning it off
09:41:55Lssturning it on is a bitch
09:41:58Lsspretty long wait
09:42:01pj188899i already new i thought it would be diffrent
09:42:04pj188899no really
09:42:05gartralreversi also has a quirky long press read, it accepts the initial press then the long press afterward, im going to look into changing this
09:42:10pj188899not really its not that ong
09:44:13pj188899nvm now i see what you mean ..
09:44:17pj188899it only found nine?
09:44:41Unhelpfulprobably some of your files lack tags?
09:45:43pj188899well i guess ill be tagging some songs..
09:45:52Lssgood luck
09:45:55Lssi started in 2006
09:46:01Lssit never ended since
09:46:58pj188899i dont have that many songs..
09:47:05pj188899why didny you just get an autoer?
09:47:08scorchelets keep this ontopic please...
09:47:11gartrali have just shy of 1000 tracks
09:47:31Lssi realised that everyone has their own tagging style
09:47:36Lsswhich messes up my own style
09:47:44Lssso auto taggers are mostly bad
09:47:46*scorche coughs
09:47:54Lssok oops
09:47:55pj188899not if its custom lol
09:48:27 Quit pj188899 ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
10:01:29 Join Thundercloud [0] (
10:08:24*pixelma wonders whether rockblox works correctly on Ipod Mini or 1st-3rd gen Ipods now
10:12:05B4gder <= spam or just off-topic?
10:12:18B4gderand wrong forum section anyhow
10:12:59scorchei say is their first post and why do we care about converting to xen format?
10:13:03scorchezen too
10:13:04gartraljust off topic, but someone surly did a lot of work to make it
10:13:55scorcheno...certainly spam...and if not, it has nothing to do with rockbox anyway
10:14:01B4gderwell, the phrasing made it sound like a sales pitch more than a friendly advice
10:14:39B4gder"Designed and manufactured by Creative Technology, the new player of Creative family Creative Zen sounds fantastic and features a brilliant color screen capable of displaying photos and video"
10:14:57B4gderbla bla bla
10:15:10*scorche points to the trashbin
10:15:13kadobanthat's spam...i kept expecting it to start trying to sell me viagra
10:15:25linuxstbB4gder: Definitely spam...
10:15:38B4gderI moved it
10:17:45kadobanCould anyone take a glance at this and see if they have any big objections? I mentioned it earlier but it's a way to add the Playback Control menu to plugins without menus. Instead of immediately exiting, it brings up a small menu with Playback Control and Quit.
10:18:22kadoban(just want to check before i add it to a whole bunch of plugins only to find out it's awful)
10:20:17gartrali'm trying to figure out why reversi registers both the *press* of the center key, and *then* brings up the context menu... as far as i can tell, i should just be able to invert the placement of 2 lines, and have it work
10:21:51kadobangartral: and does it work if you do so?
10:22:00gartrali can't build...
10:22:40linuxstbgartral: It probably just uses something like "BUTTON_NAME", but it should be "BUTTON_NAME | BUTTON_REL" - so it does the short-press action on release, not first press. Or maybe you can switch to using the action API....
10:24:35pixelmaright above the BUTTON_REPEAT one - assuming you are looking at the Sansa keymaps
10:25:49gartraleventually i'll identify all the "offendin values" and change them
10:25:53pixelmalinuxstb: I hope you don't mean PLA...
10:26:17pixelmagartral: as far as I can see there only is one
10:27:25gartralyes, the sansa specific ones are combined, but i meant across the entire file
10:28:56kadobanit looks like the _PRE stuff is only set up for ondio
10:29:01kadobani assume that that's the problem
10:29:08pixelmaonly the Ondio keymap use long press vs. short press of a button - and it already checks BUTTON_REL and BUTTON_REPEAT
10:29:35linuxstbpixelma: No - that's why I just said "Action API" ;)
10:29:49pixelmakadoban: I think it's only needed for button combos but am not sure at the moment
10:30:17kadobanpixelma: sansa uses the long press as well, and the _PRE isn't set least i think that's right
10:33:31 Quit nuonguy ("This computer has gone to sleep")
10:35:08***Saving seen data "./dancer.seen"
10:35:30 Quit petur ("stack overflow")
10:37:16 Quit Thundercloud (Remote closed the connection)
10:37:57pixelmawell as linuxstb pointed out - the other use of the Select button should have a BUTTON_REL , could be all that's needed
10:39:38 Join Xerion_ [0] (
10:39:48gartralthats what im trying now...
10:42:23pixelmaI wonder about the ALT_MAKE_MOVE for the Ondio pad - why it is there. My guess is that someone just copied the sudoku keymaps without thinking (or asking)
10:42:59gartral <−− ok, replaced by a dif
10:43:04 Join DarkSpectrum [0] (
10:45:10 Quit Kohlrabi (Remote closed the connection)
10:53:27 Nick fxb__ is now known as fxb (
10:55:36 Quit Xerion (Connection timed out)
10:55:36 Nick Xerion_ is now known as Xerion (
11:04:05gartralshould i submit a FS for this new patch?
11:08:34 Quit einhirn (Read error: 104 (Connection reset by peer))
11:12:31 Quit SirFunk (Read error: 60 (Operation timed out))
11:14:19 Join SirFunk [0] (
11:18:12Lssjust something random but is it possible to flip the screen so i can use my ipod upsidedown?
11:20:02DarkSpectrumis there a pre-compiled download for the sansa clip i can experiment with?
11:22:44Unhelpfulyou could check the unsupported builds forum, or the port's new port thread, or wiki page, but generally unsupported targets are not really ready for the general public...
11:22:50gartralFS #9961 is my new patch. and Lss yes, upside down mode is under settings>general settings>display>LCD settings>Upside down *IF* the hardware supports it
11:23:44gartralhey, shouldn't that be called Display Settings for consistency?
11:23:45 Quit jordoex (Read error: 110 (Connection timed out))
11:23:46n1skadoban: maybe i am not the target user of playback controls in the plufins, never understood why we want them at all but I *do* value quick exit in simple plugins and would rather not see an extra menu added just to be able to skip to the next track
11:24:09DarkSpectrumty been lookin for hours and couldnt find anything pre-compiled
11:24:51n1sgartral: Settings->Display->Display Settings?
11:25:37n1sbtw _LCD_ settings are seperate from display settings, the lcd settings are for the lcd and display is more general app settings
11:25:50kadobann1s: does it matter that it's pretty easy to exit from? (on e200 << and power both exit from that menu)
11:26:11Lssdang its not there
11:26:23Lssmay i know which ipods actually support this?
11:26:37gartraln1s: thats the path too the upside down mode in my e250
11:26:50n1skadoban: so if you press << in the menu it exits the plugin?
11:27:35kadobann1s: yes. how i have it now is if you don't first go into the Playback Control manaully (select or >>), the normal "get me out of this menu" buttons will exit the plugin
11:28:02n1sgartral: no, my point was that if you changed "LCD settings" to "Display settings" it would be a very weird menu structure
11:28:04kadobanif you go into that menu first, it assumes you just wanted to mess with playback and will resume after
11:28:20Unhelpfulkadoban: that seems somwhat unintuitive to me, shouldn't that return you to the plugin?
11:29:06n1syes, that is not how menus work regularly in rockbox, they tend to return to where they were opened if you exit
11:29:16kadobanUnhelpful: i don't know, it's easy enough to do, but i wanted to make it unobtrusive to exit (another quick button press of the button you probably pressed to get there instead of selecting Quit from the menu)
11:29:52 Quit _lifeless (Remote closed the connection)
11:30:23 Join _lifeless [0] (n=lifeless@
11:30:58gartralwrong place kadoban, i was thinking settings>General settings>Display Settings>LCD Settings, instead of settings>General settings>Display Settings>LCD Settings
11:31:10gartralwrong place kadoban, i was thinking settings>General settings>Display Settings>LCD Settings, instead of settings>General settings>Display>LCD Settings
11:31:19kadobangartral: i have no idea what you're talking about
11:31:41 Join gromit` [0] (
11:32:24kadobanso yeah, i don't know. there's two competing preferences: easy/quick exit of the plugin, and consistency in menus. i don't really know which should win out
11:32:33gartralwrong place n1s, i was thinking settings>General settings>Display Settings>LCD Settings, instead of settings>General settings>Display>LCD Settings, sorry kadoban
11:33:23pixelmagartral: I think you meant Lss - but according to the manual, e.g. the Ipod Video really doesn't have it (it's probably not known how to)
11:33:33kadobanwe could of course stick with the current state of affairs, and just not have playback control in those games, but it makes it kind of suck to use them long term while listening to music
11:33:53n1sgartral: ok, then i misunderstood but note how none of the entries under General Settings contain the word "Settings"
11:33:54pixelmawell, Rockbox is about music ;)
11:35:56kadobanright, that's why i want to add playback control to as many plugins as it makes sense to :)
11:36:00*linuxstb thinks it would be nice if all games saved their state when exiting, and automatically resumed when started
11:37:15kadobanthat would add a decent amount of complexity though
11:38:32 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
11:39:19*n1s agrees with linuxstb
11:39:26 Quit gromit`` (Read error: 110 (Connection timed out))
11:40:08kadobanso is the idea of having a menu instead of directly quitting just fundamentally bad?
11:41:47Unhelpfulkadoban: i favor consistency, in this case, put the playback menu anywhere we can
11:42:20kadobanUnhelpful: alright, that was my feeling as well. i guess i'll just continue, it's easy to change the behavior of the menu itself at a later time anyway
11:43:22*gartral agrees with Unhelpful
11:43:56*n1s already stated his opinion
11:47:46 Join robin0800 [0] (
11:47:57 Quit robin0800 (Remote closed the connection)
11:48:10 Join robin0800 [0] (
11:50:01robin0800jaunty alpha4 won't see my sansa c240 upgraded today with bootloader patch log sayes usb-storage: waiting for device to settle before scanning
12:00:24 Join einhirn [0] (
12:04:22 Join moos [0] (i=Mustapha@rockbox/staff/moos)
12:08:29 Join pyro_maniac [0] (
12:09:51 Join tyfoo [0] (
12:11:28 Join gromit`` [0] (
12:15:33n1srobin0800: do you have "start screen" set to "resume playback" ?
12:16:30 Join Kopfgeldjaeger [0] (
12:19:02 Quit DarkSpectrum ()
12:20:45 Quit gromit` (Read error: 110 (Connection timed out))
12:24:50gartralhow do i tell how big database is without/before committing it too ram?
12:29:11n1srobin0800: theres a bug that prevents usb connect on startup with that setting so either disable it or start rockbox before connecting usb
12:29:41advcomp2019robin0800, you can change that to main menu.. look at FS #9957 if i remember right
12:32:46 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
12:35:09***Saving seen data "./dancer.seen"
12:36:10 Quit Kopfgeldjaeger ("Serverwechsel")
12:36:45 Join Kopfgeldjaeger [0] (
12:38:52 Quit robin0800 (Read error: 104 (Connection reset by peer))
12:45:11 Join daurnimator [0] (n=quae@unaffiliated/daurnimator)
12:49:05 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
12:52:34 Join Llorean [0] (
12:52:38 Join gromit` [0] (
12:58:50gartralhow do i tell how big database is without/before committing it too ram?
12:59:00gartralbig a
12:59:30 Quit gromit`` (Read error: 110 (Connection timed out))
12:59:53 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
13:00:22 Quit domonoky (Read error: 54 (Connection reset by peer))
13:03:15gartralhmm.. setting a songs rating/times played doesn't work with database ready (not in ram) and gather runtime data set to yes
13:08:21 Join gromit`` [0] (
13:10:19gartrali belive i had this problem a while ago, but cant remember when
13:16:01 Quit gromit` (Read error: 110 (Connection timed out))
13:21:26 Part LinusN
13:36:51 Join Hawson_ [0] (
13:37:50 Join Alex00088 [0] (n=chatzill@
13:38:55Alex00088I just loaded a .cfg file, after not doing that for months. Is it just me, or has that been sped up A LOT?
13:40:37LloreanBagder: The way it was explained to me, the host operating system is the problem. We're exporting it as a block device, and it expects to be the only user of the file system. So at best, I guess maybe Rockbox could be in read-only mode during that time, but I seem to recall there were other complications too.
13:42:43LloreanAlex00088: iPod 5G?
13:42:53Alex00088Sansa c240
13:43:16LloreanI don't remember any recent changes to flash access on it, but I could be wrong.
13:43:27Alex00088If it hasn't been sped up, what might I be doing different to get it to load faster?
13:44:28LloreanYou might've been experiencing a bug before, or been doing it during periods of playback when the CPU and flash were busy?
13:45:27Alex00088I don't know. Right now they load in a few seconds, but a few months ago, I think it could take ~1 minute
13:45:44LloreanThat should never have been the case. Whatever caused it, something was wrong.
13:46:03LloreanI thought you were talking maybe dropping from 2-3 seconds to less than 1 second.
13:46:36Alex00088well I'm happy now. Thanks.
13:47:21 Join Casainho [0] (
13:48:19Alex00088btw - it might have been closer to 30 seconds - I'm not sure. Have a good one!
13:48:25 Quit Alex00088 ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
13:52:23Casainhohello :-)
13:53:22gartralhmm.. setting a songs rating/times played doesn't work with database ready (not in ram) and gather runtime data set to yes... any thoughts? its my e250..\
13:55:01 Quit Hawson (Read error: 110 (Connection timed out))
14:03:05gartralis there anyway to tell hoe big the database is without/before committing the whole thing too RAM?
14:04:05LloreanIf it's too big for your tastes, turn it off and reboot again.
14:09:25gartralram lists are in bytes, right? (im 80% sue they are, i'm just making 100% sure)
14:09:50LloreanI don't even know what you mean by "ram lists"
14:10:49gartralwhere it shows how much ram its using versus how much is there too use
14:11:34gartralall rockbox says is "RAM: 134028/167216"
14:12:03gartralthats under Database Info in the debug menu
14:12:15LloreanThat's how much it's using, out of how much is reserved for it.
14:12:23LloreanShould be in bytes, yes.
14:13:26gartral:) thank you
14:14:23LloreanI seem to recall you have an e200, in which case you really shouldn't care about RAM usage.
14:14:42 Quit FlynDice ("Konversation terminated!")
14:17:34 Join Barahir [0] (
14:18:19CasainhoI would like help in understand the audio drivers. I would like to know if pcm_play_data_start() is called and DMA ends the transfer, how do Rockbox firmware knows that should transfer more data using DMA?
14:20:17 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
14:21:17 Quit pyro_maniac (Read error: 110 (Connection timed out))
14:23:00CasainhoThere is a comment "Set up the DMA transfer that kicks in when the audio FIFO gets empty" for pcm_play_dma_start() on pcm-coldfire.c −− I would like to know when firmware knows that FIFO is empty...
14:24:07CasainhoI understand that DMA transfer data from that FIFO to I2S, however don't understand what and when triggers the DMA transfer, when it finish to transfer the FIFO everytime...
14:25:25*Nico_P wonders whether FS #8894 could be included into 3.2
14:26:24linuxstbCasainho: Look in firmware/pcm.c - the dma handler will call the get_more() function passed to pcm_play_data() each time it finishes a transfer.
14:27:47Casainholinuxstb: where is the DMA handler?
14:28:11linuxstbThat's what you have to write - the ISR that is called when DMA is complete.
14:28:43Casainhoah, okok - I think I understand.
14:28:55linuxstbThere should be one in pcm-coldfire.c
14:29:54Casainhoyes, there is one. I will look on it to understand. That code on coldfire port have good comments, it helps me ;-)
14:31:31Casainholinuxstb: I don't understand this: "get_more = pcm_callback_for_more;" −− what is a callback?
14:32:15MTeelinuxstb : how do know all of these stuff ? :)
14:35:10***Saving seen data "./dancer.seen"
14:38:21linuxstbMTee: I've written an audio driver for Rockbox (the first ipod drivers)...
14:39:03 Join pyro_maniac [0] (
14:40:37Casainholinuxstb, Bagder : I think I understand now better the code - I am ready to make it working on this weekend ;-)
14:40:43Casainhothanks ;-)
14:40:44Casainhobye bye
14:40:54 Quit Casainho ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
14:41:02MTeeoh I see
14:42:21 Join Seed [0] (
14:45:40 Join LambdaCalculus37 [0] (i=44a04303@rockbox/staff/LambdaCalculus37)
14:46:50 Join nibbler [0] (
15:09:37 Quit SirFunk (Read error: 60 (Operation timed out))
15:10:57 Join SirFunk [0] (
15:19:32 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
15:35:37 Quit linuxstb (Read error: 113 (No route to host))
15:36:42 Quit kadoban (Excess Flood)
15:37:00 Join jordoex [0] (
15:37:03 Join kadoban [0] (
15:38:01 Quit bmbl (Connection timed out)
15:40:12 Join jaykay [0] (
15:40:53 Quit bombardier ("leaving")
15:43:52 Join slowmoe [0] (
15:44:51 Quit kadoban (Excess Flood)
15:45:12 Join kadoban [0] (
15:46:55LloreanjhMikeS: Is it possible to do CPU boosting on the Gigabeat S like we do on other targets (and if so, have we tried to see if we get battery life gain?)
15:49:29gartralthe e250 v1s only have 32 megs of ram right?
15:50:02LambdaCalculus37The c200 series also have 32MB of RAM.
15:52:10 Quit kadoban (Excess Flood)
15:52:37 Join kadoban [0] (
15:53:59*gartral leans to distinguish bytes from kilobytes
15:55:48ender`32MB RAM is now "only"? :)
15:57:18gartralnah, i missread the size of something a few times and it confused me, i saw what i thought was along the lines 61 megs, and it turns out i was seeing double of a 0
15:57:21 Join rocko [0] (
15:57:40 Quit rocko (Read error: 104 (Connection reset by peer))
15:58:13gartralspecifically the "Limit: 6291456" line of Dircache Info
15:59:32gartrals// 0/9
15:59:36 Quit kadoban (Excess Flood)
16:00:02 Join kadoban [0] (
16:01:04 Join Febs [0] (n=chatzill@
16:03:28 Join evilnick [0] (i=0c140464@gateway/web/ajax/
16:06:00 Join rocko [0] (
16:07:57 Quit kadoban (Excess Flood)
16:08:03 Quit jordoex (Read error: 110 (Connection timed out))
16:08:19 Join kadoban [0] (
16:09:20 Quit daurnimator ("Cyas later...")
16:15:55 Join webguest47 [0] (n=db69e38d@gateway/web/cgi-irc/
16:16:25 Join phill [0] (
16:17:11 Quit webguest47 (Client Quit)
16:32:51 Join vertic23 [0] (
16:33:29vertic23I'm sorry to bother you again ... ipod video 5.5 <−− this one - is just great for rockbox, right?
16:33:58 Join toffe82 [0] (
16:34:04evilnickvertic23: Looks like it
16:34:25vertic23\o/ 150€ is quite ok, isn't it?
16:34:41evilnickWhat capacity hard drive?
16:34:53evilnick(and what currency was that?)
16:35:12***Saving seen data "./dancer.seen"
16:35:51evilnickYeah, that sounds okay
16:38:10 Join MethoS- [0] (
16:38:20 Quit domonoky (Read error: 104 (Connection reset by peer))
16:40:04 Join kugel [0] (n=kugel@rockbox/developer/kugel)
16:40:41 Join jgarvey [0] (
16:41:35vertic23finally ... had to use a samsung mp3 player the last month ... and the sound was so terrible
16:41:43vertic23now back to rockbox at last!
16:41:56 Quit slowmoe ()
16:44:28 Quit jaykay (Read error: 110 (Connection timed out))
16:44:29 Quit phill ()
16:44:35 Part pyro_maniac ("Leaving.")
16:51:09 Join Esterbrook [0] (n=carlosgu@
16:51:24 Part Esterbrook
16:52:40gartralfs 9960 and 9961 are two reversi patches by me, the first is asthetic, the second attempts to make the game recognise the difference between a press and a hold (or long press)
16:53:28LambdaCalculus37Someone on the forums posted that Toshiba has just released a firmware update for the Gigabeat T series (no WiFi); here's the link to the firmware update:
16:57:14gartralhuh... the info screen screen of rbutil shows an MD5 hash of the bootloader file, instead of the version string...
16:57:36rashergevaerts: Would logf provide any insight in the usb-at-boot failure?
16:57:54rashergevaerts: Since it doesn't crash, there should be a reasonable chance of capturing stuff
17:00:17toffe82LambdaCalculus37: thanks
17:01:15*gartral wonders if his patches will work, as he can't build :|
17:01:51pixelmakugel: do you know that not all targets that have a scrollwheel (physically) define HAVE_SCROLLWHEEL ?
17:02:23kugelpixelma: You mean older ipods?
17:02:29kugelNo I didn't know that
17:02:35pixelmayes, and the Mini
17:02:44kugelwhy is that?
17:03:20gevaertsrasher: it could, yes. I don't really know what to look for though. I have some suspicions though, but I need to think about them a bit more
17:03:39pixelmaa short read of the SVN logs last time, it seems that this define was introduced for wheel accelaration which has not been adapted to non-"clickwheel" Ipods yet in the driver
17:03:44 Part B4gder
17:05:08kugelwell, that's a bug, imo
17:05:27kugeli.e. my commit was OK
17:06:06LambdaCalculus37saratoga: (for the logs) There's a Gigabeat T firmware update released by Toshiba for non-WiFi models (i.e. the US models); we should check it out to see if there's a digital signature exploit like on the S.
17:06:11*LambdaCalculus37 wonders if that was correct
17:06:26pixelmaunfortunately I'm not sure what exactly happens now but potentially breaking a plugin is not ok, IMO
17:06:41kugelif scrollwheel targets don't define scrollwheel
17:06:41 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
17:07:02kugelthat's a bug. Plugins should define their own scrollwheel just because of that
17:07:40kugelpixelma: and no, it's not broken. I know that because e200/fuze didn't use the code within SCROLL_WHEEL, and the scrollwheel worked anyway
17:07:43pixelmawell, that global define has a misleading name... and I guess the plugin was there before this wheel accelaration stuff
17:08:26*LambdaCalculus37 goes to put the link to the firmware updater on the wiki
17:08:31gartralkugel: but could it be thats what's causing doom to not pick uo some scroll wheel action on the E200s?
17:08:47kugelgartral: No, that's something entirely different
17:08:55kugelthis is only about rockblox
17:10:45Lloreankugel: Even if a define is badly named, you shouldn't use it until you know what it does (don't trust the name).
17:12:39kugelI know what it does. I just didn't know some scrollwheel targets don't define it
17:12:58pixelmait looks like it could work differently now on thos Ipods, don't know
17:13:12pixelmaeven if not completely broken
17:13:25Lloreankugel: I don't understand that sentence. If you knew what it does, that means you should know why it's defined, and by which targets...
17:13:39Llorean"What it does" includes what uses it.
17:14:11kugelI know that it ( is supposed to) identify scrollwheel targets AND activates wheel acceleration.
17:14:20kugelI messed with HAVE_SCROLLWHEEL on the fuze
17:14:52Llorean"What it's supposed to do" is different from "what it actually does" though.
17:14:54LloreanThat was my point.
17:14:57kugelknowing what it does doesn't include that which targets don't define it (wrongly, imo)
17:15:25pixelmathey can't just define it
17:15:33kugelLlorean: It's supposed to identify scrollwheel targets, and activates wheel acceleration code. That IS what it does
17:15:59Lloreankugel: Clearly it's NOT what it does, since some scrollwheel targets don't have the define.
17:16:03LloreanIt's what it *should* do, not what it *does*
17:16:24LloreanIt identifies some, but not all, targets with a wheel.
17:16:26kugelLlorean: it activates wheel acceleration code, it DOES that
17:16:30Lloreanyes, it does that
17:16:50LloreanBut in every previous line, you've also claimed it does something else.
17:17:01kugelI did not
17:17:11pixelmaif I understood correctly it needs some work on the driver(s)
17:17:31kugelpixelma: either this, or it wheel acceleration code should be based off a different define
17:17:41Lloreankugel: If you were only using it to activate acceleration, why'd you remove the other define?
17:17:51kugelthat acceleration thing got me into trouble on the fuze as well
17:18:47 Join miepchen^schlaf [0] (
17:18:49 Quit MethoS- (Read error: 104 (Connection reset by peer))
17:19:09kugelLlorean: it's not only, as I said, it's also to identify scrollwheel targets. And i only added the "supposed to", since it apparently isn't as determining as I thought
17:19:14Lloreankugel: My point was, you didn't know the differences in how SCROLL_WHEEL and HAVE_SCROLLWHEEL work before swapping them. So you didn't know what one of them did.
17:19:20 Join MethoS- [0] (
17:19:27kugelvarious other places of rockbox use HAVE_SCROLLWHEEL, without relation to acceleration
17:19:31LloreanIt *doesn't* identify scroll wheel targets
17:19:39LloreanIt may be supposed to, but that's NOT what it does right now
17:19:53pixelmaI have to be more precise though: only the first gen Mini does not have it (as the second generation has a so-called "clickwheel" which was adapted)
17:19:53 Quit tmzt (Operation timed out)
17:20:23 Quit Zagor ("Client exiting")
17:20:42vertic23is it possible to make the ipod wheel less sensitive?
17:22:03 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
17:24:50kugelso, the plan is to define HAVE_SCROLLWHEEL for *all* scrollwheel targets, and define another HAVE_WHEEL_ACCELERATION for the actual acceleration code
17:25:52n1srasher: i would suspect some kind of conflict between usb trying to grab the audio buffer while playback is starting
17:25:57 Nick agaffney is now known as agaffney|phone (n=agaffney@gentoo/developer/agaffney)
17:26:16rashern1s: Yeah, that's the assumption
17:26:49 Nick agaffney|phone is now known as agaffney (n=agaffney@gentoo/developer/agaffney)
17:28:01Lloreankugel: I'm not sure, but I think the wheel need some distinction because some allow us to get the absolute position, and others don't.
17:28:20vertic23is this discussion right now about 'my' problem?
17:28:44Lloreanvertic23: No.
17:28:54kugelLlorean: which allows to get the absolute position?
17:29:02vertic23is there an answer anyways for my problem? :P
17:29:16Lloreankugel: Newer iPods, *probably* the ones that currently use HAVE_SCROLLWHEEL but I'm not sure.
17:29:31LloreanIt's one of those "click wheel" vs "touch wheel" vs "scroll wheel" vs whatever things.
17:29:38n1svertic23: it's probably possible, dependign on what you mean
17:29:38kugelNo, those do not
17:29:58 Join tmzt [0] (
17:29:59vertic23just slow down the speed
17:30:09pixelmaLlorean: there already is HAVE_WHEEL_POSITION for this
17:30:43pixelmabut wheel accelaration is not dependent on this, it was even introduced for the e200 wheel
17:30:45kugelor wait, seems they do
17:31:00Lloreanpixelma: I thought the iPod wheels used a different sort of acceleration though.
17:31:23 Quit GodEater_ (" ajax IRC Client")
17:31:27LloreanAnd the e200 wheel gives us absolute positioning too, doesn't it? We have deltas.
17:31:44LloreanSomething like 96 values that it scrolls through?
17:31:47kugelthe e200 wheel doesn't give absolute position
17:32:06Lloreankugel: It gives absolute position on a scale, rather than just "scrolling left" or "scrolling right"
17:32:25LloreanMy understanding was that the old wheels were like having a "left" button and a "right" button, effectively.
17:32:35kugelthe e200 scrolling is read by 2 bits
17:32:37LloreanYou didn't have the information to get a delta for speed.
17:32:51 Join flydutch [0] (
17:33:10kugelwhich do some 01 -> 00 -> 10 -> 11 rotation
17:33:30pixelmaLlorean: it uses the same define though
17:33:51Lloreankugel: That seems awful small, it seems like you'd be able to easily trigger accidental scrolls in the other direction with that.
17:33:55kugelLlorean: you have the information, because you never miss a change
17:34:04 Join grndslm [0] (
17:34:22kugelNo, that's not aweful, and works perfectly with the use of a interrupt handler
17:34:47Lloreankugel: I was still talking about the iPod wheels when I talked abuot the information. That's why I didn't say "kugel:" before it indicating it was a response to your newer statements, rather than a continuation of my current topic.
17:36:56kugelanyway, on the e200 (and fuze) you have deltas, not sure about other wheels
17:37:07Lloreankugel: e200v1 or v2?
17:38:32LloreanAh, I see. We do have absolute positioning within a range, as I said. You have the 00, 01, 10, 11, but they allow us to keep track of exactly where the wheel is and how fast it moved.
17:38:41 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
17:39:00LambdaCalculus37saratoga: Check out the logs.
17:39:15Lloreankugel: So... why did you tell me we didn't have the ability to position it within a range?
17:39:48kugel"And the e200 wheel gives us absolute positioning too, doesn't it?" that was the question
17:40:03kugeland I answered that it doesn't give the absolute position
17:40:20LloreanIt does, though, as far as "absolute" can go with a physically rotating wheel with no reference mark.
17:40:31LloreanOnce we've started tracking it, we know exactly how far rotated it is from its original position.
17:40:32 Join low_light [0] (i=c730190a@gateway/web/ajax/
17:40:48kugelit's not tracked though
17:40:51saratogai thought i had but it seems the log script stopped working a couple hours back
17:40:53kugelonly the delta is tracked
17:41:02LloreanIt's tracked temporarily for the purpose of determining the delta.
17:41:05LloreanWe could track it wholly.
17:41:11LloreanWe don't have a need to, though.
17:41:15kugeland the time between the calls of the clickwheel function determines the acceleration
17:41:29kugelI don't see how my statement was wrong
17:42:42LloreanThe e200 wheel gives enough information for absolute positioning. In total effect, it gives absolute positioning. My point was to contrast this with the early iPod wheels
17:42:43 Quit tvelocity (Read error: 104 (Connection reset by peer))
17:42:53pixelmasaratoga: there is still the raw text log
17:43:39Lloreankugel: On the c200, for example, we have no wheel. Just an up button and a down button. If you hold "up" you can't get a delta, because it's just "true" for up, or not. That's how the early iPod wheels were explained to me. You just know if they're scrolling or not, not how far or how fast.
17:44:05LloreanI don't know if this explanation is definitely how they work, just what I was told about a limitation of them.
17:44:08kugelLlorean: my point was: just by reading the wheel values, you cannot determine the absolute position. You need exact tracking, from the point of when the wheel was powered on.
17:44:22 Join tvelocity [0] (
17:44:55 Quit avis (Remote closed the connection)
17:45:09rasherNico_P: any thoughts on FS #9957?
17:45:11LambdaCalculus37saratoga: Look at 17:06:06 on the raw text feed.
17:45:22LloreanI never said you could get it *just* by reading the wheel values. I said some "allow us to get the absolute position"
17:45:44 Quit ender` (" Washing your car to make it rain doesn't work.")
17:46:08 Quit miepchen^schlaf ()
17:47:16saratogaLambdaCalculus37: ok misread that post in the forums
17:47:29saratogado you have a T?
17:47:40LambdaCalculus37saratoga: Nope. Got an F and an S.
17:47:55LambdaCalculus37I wanted to point it to you because toffe82 mentioned that you had a T.
17:48:05saratogayeah i picked one up cheap
17:48:33saratogai'm not really familar with the exploits on the S, but if someone wants to suggest how to check for the one on the T, I don't mind trying
17:48:37saratogai never use the thing anyway
17:49:20LambdaCalculus37saratoga: I don't remember who it was that discovered the exploit on the S in the first place, but maybe you could ask jhMikeS; he might know something.
17:50:33LloreanIf we use the Gigabeat S updater (official) to update to 1.2, is Rockbox reinstallable now? I never did find out if we solved that.
17:51:12 Join MethoS-- [0] (
17:54:09toffe82Llorean: I installed rockbox on a S with 1.2 and it was working, but using sendfirm, no the beastpatcher
17:55:25Lloreantoffe82: Did you install 1.2 using the exe installer?
17:55:46toffe82Llorean: yes using the installe
17:56:06toffe82it works fine and you still can go back to 1.1 if you want
18:00:04 Quit MTee (Read error: 60 (Operation timed out))
18:03:07toffe82Llorean: I use the 1.2 updater to go to 1.2 and the hacked updater to go back to 1.1
18:05:49 Quit MethoS- (Read error: 110 (Connection timed out))
18:12:24 Quit MethoS-- (Read error: 110 (Connection timed out))
18:16:59 Quit saratoga ("CGI:IRC (Ping timeout)")
18:17:36 Join nuonguy [0] (
18:19:48 Part domonoky
18:22:13 Join CaptainKwel [0] (i=2669ecc2@gateway/web/ajax/
18:34:56 Join woyciesjes [0] (
18:35:14***Saving seen data "./dancer.seen"
18:36:02 Join gregzx [0] (
18:37:35 Quit grndslm (Client Quit)
18:37:46 Join gromit` [0] (
18:44:00 Quit gromit`` (Read error: 110 (Connection timed out))
18:47:27 Join SirFunk_ [0] (
18:48:26 Join newnick [0] (n=d0ff9c9a@gateway/web/cgi-irc/
18:48:54 Nick newnick is now known as MaxHork (n=d0ff9c9a@gateway/web/cgi-irc/
18:52:05*MaxHork echoes
18:52:17LloreanMaxHork: Did you have a question?
18:52:32MaxHorkAh sorry was just reading, i'll head over to the community.
18:52:55 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
18:52:55 Quit MaxHork (Client Quit)
18:54:05 Join MaxHork [0] (n=d0ff9c9a@gateway/web/cgi-irc/
18:56:25MaxHorkOk now I have a question!
18:56:42MaxHorkHow do you switch over to the rockbox community IRC server?
18:57:48Nico_Prasher: no thoughts so far, but I'll take a look now
18:59:11MaxHorkNevermind, been awhile abit rust on the commands
19:00:40 Quit SirFunk (Read error: 110 (Connection timed out))
19:02:05rasherNico_P: great
19:03:13 Join __lifeless [0] (n=lifeless@
19:10:13 Join phill [0] (
19:13:10 Quit courtc (Read error: 113 (No route to host))
19:13:39 Join midijunkie [0] (
19:13:54MaxHorkOk got a question
19:14:06MaxHorkI'm doing the manual installation on a sansa e250
19:14:18MaxHorkusing the PDF instructions at
19:14:47gartralwhats the question?
19:14:53MaxHorkCurrent Build. The current build is built at each source code change to the Rockbox
19:15:04MaxHorkugh hld on stupid computer
19:15:14MaxHorknow im at the fonts section
19:15:30MaxHorkwhen I downloaded the latest fonts and extraced to the sansa root
19:15:36LloreanPlease don't paste a bunch of lines from our webpages and docs.
19:15:44MaxHorkI didnt mean too
19:15:58MaxHorkbut lets continue at the question at hand shall we
19:15:59LloreanAre you doing the automatic install using RBUTil as recommended by the manual?
19:16:08LloreanWhy not?
19:16:09 Quit phill (Read error: 60 (Operation timed out))
19:16:18gartralno sansa found bug, i would assume
19:16:34MaxHorkThats not the question.
19:16:37gartralhe did mention he was doing this on an e250
19:16:47LloreanMaxHork: That's my question, to you.
19:17:00MaxHorkWell let me finish my question and I'll answer yours
19:17:04LloreanThe automatic install is a much better installation method, and really should be used to prevent user error.
19:17:07 Join faemir [0] (
19:17:24MaxHorkIts not an error, its a point of uncertainity
19:17:29MaxHorkinstalling fonts
19:17:30LloreanIf your question is about the manual install, I'm probably going to answer it by saying "use the automatic install" anyway
19:17:47gartralrbutil can install fonts too
19:17:49MaxHorkok well thats fine, let me WRITE OUT MY QUESTION
19:17:49rasherLlorean: Still, if the manual install instructions are wrong or unclear, we want to know, surely?
19:18:03LloreanMaxHork: Nothing I've done prevents you from typing...
19:18:16MaxHorkinstalling fonts
19:18:21MaxHorkextracted to sansa
19:18:29*evilnick has yet to see a question
19:18:32LloreanPlease, type in complete sentences, as per the channel guidelines.
19:18:41MaxHorkstats destination already contains folder .rockbox
19:18:58MaxHorkdo you still want to merge this folder
19:19:03rasherAny Iriver H10 (5GB) users around?
19:19:04 Join phill [0] (
19:19:33*kugel uploaded a new version for disabling wps updates
19:19:34MaxHorkdate modified 12/23/2008 with date modified 2/27/2009
19:19:43 Quit _lifeless (Read error: 113 (No route to host))
19:19:55*gevaerts waits for the question
19:19:58MaxHorkand I see nothing about accepting a merger in the pdf
19:20:01MaxHorkshould I merge?
19:20:10gartralthis is why we recommend the rbutil... but your case, yes, merge
19:20:12LloreanMaxHork: Is your question "Should I tell it to merge the folders?". This is basic file system use... if you have one folder, and you're trying to merge its contents with the other, then you should say "yes"
19:20:20LloreanBut RBUtil will install the fonts for you, without asking you questions like that.
19:21:05kugelany ipod 5g user?
19:21:07MaxHorkRight I understand that now, but I tend to read in a weird way and do things the hard way.
19:21:07LloreanThe PDF assumes you know computing basics, and the questions you will be asked from your OS will vary across OS versions, etc. Extracting or copying one folder into another is not Rockbox-specific at all.
19:21:30evilnickkugel: Yes
19:21:35gartraland irc is not a linear conversation like a telephone call, we can scroll back and read something past the visible field...
19:21:36woyciesjeskugel: Yep, I have a iPod Video, 5.5G
19:21:41kugelcan you check this patch out?
19:22:21kugelit possibly gives upto 1h more runtime (depending on the wps)
19:22:24evilnickkugel: Sorry, can't from work.
19:22:49kugelshould work for beast too.
19:22:49 Quit MaxHork ("CGI:IRC (EOF)")
19:23:44kugelwoyciesjes: can you?
19:23:56kugelA battery bench would be interesting
19:24:35woyciesjeskugel: With instructions... Yep, I'm new here, and not a programmer....
19:24:57kugelwell, I could just send you a build
19:25:36woyciesjesK. Whatever is easiest.
19:25:41gartralspeaking of battery benches, does anyone know the specifics of the stock sandisk headphones that come with an e200? i want to do one, but i have no idea what the range/response is of these things..
19:25:42 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:26:03kugelwho cares about the stock headphones?
19:26:15gartralthe fact that i have no other ones
19:26:53gevaertsin the context of battery benches?
19:27:34gartralthings = headphones
19:28:01kugela) headphones hardly matter w.r.t. runtime (at least the in-ear ones) b) you can do the bench without
19:28:27Lloreankugel: They make a significant difference on runtime...
19:29:14woyciesjeskugel: I presume you can find my email?
19:29:23LloreanPlayers run significantly longer without headphones attached, and I've got some open headphones with lower impedance than some of my earbuds have.
19:29:33kugelwell, I mean the differences between the different headphones. headphones vs no headphones make a difference. Sorry for not being clear enough
19:30:01kugelin-ear ones
19:30:08gevaertsYou need to do the bench with the same headphones for all cases you want to compare, but that's about it I'd say
19:30:52kugelwoyciesjes: wait a second
19:31:22Lloreangevaerts: Maybe we should ask all future benches be done without headphones attached. They're already very artificial circumstances since they don't include skipping, etc. At least this way we'd round out one of the user-specific factors.
19:31:31 Quit balou (Read error: 104 (Connection reset by peer))
19:32:39gevaertsLlorean: if we assume that benches from different people can be compared anyway, that sounds like a good idea
19:33:22gartralgevaerts: i wonder how many sansa owners don't have the original headphones...
19:33:27kugelOur benches are meaningless anyway, since they aren't scientific who cares!
19:33:37Lloreangevaerts: Well, we could provide test files, maybe. They're only comparable under similar circumstances.
19:33:46kugelthat at least 1 guy once claimed
19:34:16kugel(that's not my opinion!)
19:34:24gevaertsLlorean: I was thinking more about the state of the battery, but yes, files matter as well
19:34:28kugelevilnick: if you can at least install a build from work, 5 lines up
19:34:29*freqmod_gq think battery benches will never be simular as batteries have different wear-level
19:34:58gevaertsI think benches are really only comparable if you only change a single variable at a time
19:35:05*gartral knows his battery doesn't quite last a full 24 hours
19:35:14Lloreangevaerts: True. Honestly I think the only "interesting" number is the ratio of Rockbox time to OF time, given the same playlist and headphones and player.
19:35:34woyciesjeskugel: Got it.
19:35:36gevaertsLlorean: that, and for testing specific patches or features of course
19:35:45LloreanYes, rockbox to rockbox ratios
19:36:06 Join balou [0] (
19:36:24woyciesjeskugel: any specific test to perform, or just drop the files on, and let it play?
19:37:12kugelwoyciesjes: which 5g do you have?
19:37:25woyciesjesVideo, 30GB. 5.5G
19:37:43evilnickkugel: is this the iPod build?
19:37:44kugelit would be ideal if you could try to match one of the benches here
19:37:49kugela recent one
19:37:54kugelevilnick: yes
19:38:09evilnickkugel: Damnit, no USB cable! Sorry!
19:38:23*gevaerts points kugel to the discussion we just had :)
19:39:30kugelgevaerts: no need
19:39:31 Quit midijunkie (Read error: 104 (Connection reset by peer))
19:41:20pondlifeAnybody know what #ifdef __PCTOOL__ refers to, and how I can do a test build of the "PC tool" ?
19:41:34rasherpondlife: make in tools/database iirc
19:41:43 Part gartral
19:42:08rasherOr something like that. It refers to the db builder
19:42:09pondlifeThe sim is currently broken if you use the −−root option (like I do)... it has a couple of hard-coded "simdisk"s...
19:42:54pondlifeHmm, didn't get far... cc1: error: unrecognized command line option "-Wno-pointer-sign"
19:43:40linuxstbpondlife: checkwps also uses that PCTOOL define.
19:43:46pondlifeAh, ok
19:44:49gevaertsI think it basically means compilation outside the regular rockbox code
19:45:04 Join Zoxc [0] (
19:45:08kugelwoyciesjes: first thing: does the patch even work?
19:45:36 Join MaxHork [0] (n=d0ff9c9a@gateway/web/cgi-irc/
19:45:42kugelwell, probably hard to judge, but you may see that it didn't update while the lcd was off if you turn it on
19:45:48pondlifeCan someone else try to build checkwps or the database tool?
19:45:58MaxHorkHave a question regarding the sansa e250 and a microSD card
19:46:02pondlifeNeither compiles here, maybe it's a Cygwin thing?
19:46:18linuxstbYes, cygwin isn't supported here ;)
19:46:28pondlifecheckwps' buildall gives lots of "../../firmware/export/debug.h:27: error: parse error before "ATTRIBUTE_PRINTF"" errors
19:46:30gevaertsthe database tool builds fine. Checkwps doesn't
19:46:53*linuxstb checks checkwps
19:47:01pondlifegevaerts: Under Cygwin?
19:47:11gevaertspondlife: linux
19:47:28pondlifeHmm, my GCC doesn't like -Wno-pointer-sign at all
19:47:42pondlifegcc 3.4.4
19:48:08linuxstbWorks fine for me - all targets build without warning.
19:48:22linuxstb$ gcc −−version
19:48:22linuxstbgcc (Ubuntu 4.3.2-1ubuntu11) 4.3.2
19:48:32woyciesjeskugel: K. Gimme a few moments...
19:48:36pondlife3.4.4 is rather old...
19:48:48*pondlife looks for a newer gcc
19:48:49rasherI'd expect cygwin to have something newer available
19:49:37gevaertslinuxstb: checkwps.c:74: error: expected identifier or ‘(’ before ‘unsigned’
19:50:03gevaertsI expect some header defining htole32 already
19:51:11linuxstbNothing in my /usr/include defines it...
19:52:42gevaertsAnd 3.4 doesn't build the database tool here either. I agree with having all possible warnings, but I'm not sure if it's worth it for host-side tools if the price is supporting fewer compilers
19:53:58gevaertslinuxstb: endian.h has it here. Debian unstable amd64.
19:55:55MaxHorkIs there a possible way to make the information from the files>microsd1 card show up in the Database for the music on the sansa e250?
19:56:09linuxstbgevaerts: Hmm, and my Ubuntu 8.10 doesn't have it anywhere...
19:56:51rasherMaxHork: It will be there if the card is inserted when you create/update the databse
19:57:01rasherAnd nothing breaks.
19:57:47gevaertslinuxstb: libc6-dev is 2.9-3 here
19:59:04MaxHorkThanks rasher :D
19:59:05linuxstbgevaerts: 2.8~20080505-0ubuntu7
20:00:29gevaertsMaybe that's the difference
20:00:43gevaertsA few #ifndefs around it?
20:00:59MaxHorkI have another question about the size of the MicroSD card. In the Sansa e250 it said the limit was a 2GB card. Since installing Rockbox is the restriction different or unchanged?
20:01:16BagderMaxHork: any size you can find!
20:01:23pondlifeOK, if anyone can build either checkwps or the database tool, perhaps they'd like to check out before I commit it :)
20:01:54woyciesjeskugel: tested. Not working right for me. Chose song, let it play to next. Looked at WPS, had updated.
20:02:20woyciesjesThen, skipped to next track, while screen was active, did not update. Skipped two more, still no update.
20:02:41MaxHorkAwesome. Excellent job guys. Great work, fast intialize on the device no horrible waiting to load times really quality work. Thank you for your efforts and continued support.
20:03:13woyciesjesWent to main menu, then tried to go to Now Playing. Did not switch to WPS, but the controls acted as if I were on the WPS. Let the screen bacl out, during the song, hit a button to bring it back, and there it was.
20:03:24CaptainKwelkugel: doesn't seem to ever refresh until the screen wakes up (after going to sleep), and then only once at that instant.
20:03:47woyciesjesBasically, seems like you achieved the opposite. WPS updates while the screen is off, and not if it's on.
20:03:59woyciesjesGotta run to a meeting. back in about an hour....
20:04:07 Part woyciesjes ("Ex-Chat")
20:04:48 Quit MaxHork ("CGI:IRC")
20:05:02kugelI'm stupid
20:05:25kugelbut that at least shows that the patch is working
20:05:26pondlifekugel: You can't be stupid, you've successfully modified playback.c in the past...
20:05:35kugelI just need to swap the truth values
20:05:52pondlife(on second thoughts, maybe touching playback.c is a sign of stupidity)
20:06:06rasherkugel: This is obvoiusly some strange use of the word "working" that I was prevoiusly unaware of
20:06:12kugelpondlife: :) well, this time I was stupid. I had it working correctly, but that was confused by the name of my own function and changed it to a non-working version ;)
20:06:35kugelrasher: yea, well, I mean the concept is working
20:09:22gevaertsrasher: what do you need tested?
20:09:43rashergevaerts: If it can boot the OF
20:09:56kugelCaptainKwel: re-download, I updated, should be good now
20:10:09rasherApparently the current one can't, according to barrywardell, the last known working one was r17357
20:10:42 Quit vertic23 (Read error: 104 (Connection reset by peer))
20:10:43rasherBut I'd test that one as well, just to make sure
20:10:45 Join vertic39 [0] (
20:12:49pondlifeNah, looks like I'm stuck with gcc 3.3...
20:14:47 Join robin0800 [0] (
20:16:17CaptainKwelkugel: looks like it's still refreshing with the backlight off.
20:17:30 Join gartral [0] (
20:17:45 Nick fxb is now known as fxb__ (
20:17:54kugelCaptainKwel: with backlight != lcd off, in the ipod video case
20:18:06gartralmight i ask how we managed to completely skip a whole revision number?
20:18:12kugelthis is intended, since the ipod video is transflective
20:18:24kugelit should only not update if the lcd is really off
20:18:46kugels/with backlight/backlight off/
20:19:22linuxstbgartral: Not every commit is shown on the front page
20:19:35kugelCaptainKwel: You might notice that it didn't update while the lcd was off
20:19:41kugelupon turning it on again
20:19:48CaptainKwelyes, backlight != lcd off
20:20:27kugelthe updating solely depends on lcd_enabled(), and lcd_asleep() if the former isn't available
20:20:49CaptainKweloh, wait I see what you're saying
20:20:53kugeland the ipod sleeps the lcd with a little (configurable) delay after backlight went off
20:21:00CaptainKwelSo we can't ever know if it's really working?
20:21:22kugelnot really. On my sansa though, for a split second I see the old wps
20:22:01kugele.g. if the backlight turned off at 2:04, and 10s later I turn it on again, I see for a very split second 2:04, then 2:14
20:22:27CaptainKwelcause it needs to refresh, yeah?
20:22:42kugelCaptainKwel: well, the non-working version actually showed that the concept works, so I expect it to work now
20:22:55CaptainKwelI'm seeing that now.
20:23:05 Quit chrippa (Read error: 60 (Operation timed out))
20:23:10CaptainKwelIn any event, it doesn't appear to have broken anything in any way.
20:23:22kugelthat's good
20:24:36kugelwith a heavy wps (multiple conditional viewports, multiple progressbars, album art, etc) it gave +1:15h on a sansa
20:24:45 Join jones_ [0] (
20:24:59*rasher is sceptical about having multiple progressbars
20:25:04toffe82linuxstb: I will try this weekend the different configuration for installing the S ,.... if I have time;)
20:25:21kugelrasher: complains are redirected to JdGordon :)
20:25:28jones_Would anyone be able to help me out with a battery issue?
20:26:12gartralmultiple progress bars? WHHHYYYY!?
20:26:17 Nick JdGordon|zzz is now known as JdGordon (
20:26:21kugelSo, it works on my e200 too. I'd like to commit that soon if it breaks nothing
20:26:28 Quit moos ("Rockbox rules the DAP world")
20:26:35rashergartral: for testing, I assume.
20:26:47kugelI'd like to have a test on a h300 with remote though
20:27:09gartraltesting what, how fast we can make the DAPs battery die?
20:27:21kugeldoes scrolling work if the main unit is without backlight, but the remote is used?
20:27:29kugelfrom the code, it appears not
20:28:29 Join chrippa [0] (
20:28:38kugelpondlife: do you have a h300 with remote?
20:28:48pondlifeYes, a non-LCD remote though
20:29:11 Join mirak [0] (
20:29:27kugelhm, but the remote code is active anyway. would be nice if you could give a test
20:30:14pondlifeI can do, but not right now - got two other patches brewing
20:30:19jones_My battery doesn't seem to want to stay on for very long unless I'm plugged in
20:30:23jones_Is there a patch for this?
20:31:08gartraljones_: it would help, mildly, if you told us which DAP your using :P
20:31:26jones_Oh, sorry
20:31:39jones_I have a 5.5g 30GB
20:31:55jones_Beyond that, I'm not entirely sure
20:31:56kugelwhat's your runtime?
20:32:02jones_I haven't updated in a long time
20:32:08jones_About 6 months or so
20:32:26kugelI mean, how long does the battery last?
20:32:40*gartral wondres if kugel meant runtime of battery before shutdown
20:32:49jones_I'm dual booting, so switching over to Rockbox takes about a third of the battery out
20:32:55jones_And then it lasts for maybe 2 minutes
20:33:20kugeland you get normal runtime in the apple os?
20:33:47jones_That's what I keep saying
20:33:55gartralfirst, update, then see if thats any improvement...
20:34:07jones_I was just about to do that
20:34:14jones_I have to go to class soon, but I'll just work on it then
20:34:55gartralthe latest build has working USB, meaning you shouldnt have to unplug at all anymore, unless you use itunes to sync
20:35:13jones_I'm on linux, so I don't even have iTunes
20:35:15***Saving seen data "./dancer.seen"
20:35:43*gartral highly recommends and update
20:37:21*LambdaCalculus37 also recommends an update
20:42:39linuxstbI'm intending to commit my FS #9948 in the next few days (switching english.lang to sentence case). Are there any objections left? (Llorean has told me he's no longer objecting)
20:44:00*gevaerts has a terminology objection :)
20:44:53gevaertsTitles are always in title case by definition! Only some people disagree about what title case should be :)
20:45:06JdGordonlinuxstb: no objections from me.. are you going to remove deperacted strings also? or does it not invalidate old .lng files?
20:46:16rasherlinuxstb: Just committed most of the language updates.
20:46:34pixelmamultiple progressbars came with conditional viewports - to be able to have a "full line" progressbar in differently wide viewports shown on different conditions (e.g. my c200 wps does to have a little more space for album art and fill the screen a bit better if there is none)
20:46:41jones_Thanks everyone
20:46:46jones_I'll be back if it doesn't work
20:46:48 Quit jones_ ("Ex-Chat")
20:46:52pixelmaonly one is shown at a time
20:47:08rasherlinuxstb: Need a fullname for the Chinese ones.
20:47:21*pondlife wants linuxstb to resurrect the WPS-as-a-statusbar stuff
20:47:30pondlife(After 3.2, of course.)
20:47:51pondlifeThen we might have a better use for multiple progressbars
20:53:45pixelmaI think it is of good use already ;)
20:54:21 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
20:56:33 Join FlynDice [0] (
20:57:50 Join Rhodderz [0] (n=chatzill@
20:58:20Rhodderzhi is there any way to "mod" or do anything with an ipod nano 3rd gen
20:59:19Rhodderzis there any way to get free games for it?
20:59:43bluebrotherplus, we are not interested in modding the nano 3g OF
20:59:47pondlifeYou could perhaps see how far you could throw one?
20:59:52*gevaerts grumbles about whoever declared the H10 5GB MTP a supported model
20:59:58pondlifeMore of a sport than a game though
21:00:35Rhodderzwhy not the nano 3rd gen it has to be hackable if that is the right phrase
21:00:53kugelI'm off. Just as a reminder, I'm still keen on opinions about the sorting patch (see -dev ml)
21:00:53BigBambiwhy does it have to be?
21:00:56 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner]")
21:01:23BigBambiRhodderz: It has encrypted firmware and unknown hardware. That makes it as of yet unhackable
21:01:23bluebrothersearch the forums, mailing list or web. The reasons have been given over and over again.
21:01:40bluebrotherfeel free to work on hacking the nano 3g.
21:02:58Rhodderzok ill have a go. but first what was changed in the "hackable" firmware to make it play free games "illegal" or have linux on it?
21:03:21bluebrotherchanged in what hackable firmware?
21:03:29BigBambiwhat is illegal?
21:03:38BigBambiand Rockbox is nothing to do with linux
21:04:10Rhodderzwhen i was searching around rockblox changed the firmware on the ipod
21:04:18Rhodderzhang ill paste a link in a min
21:04:35BigBambiRockbox? And yes, Rockbox is a replacement firmware for some ipods
21:04:35Rhodderzif i find it again thats why i came here.
21:04:53BigBambiAnd clearly we are aware of the Rockbox website - hint, you are talking to Rockbox developers
21:04:53bluebrotherRockbox doesn't change the original firmware. It's a replacement.
21:05:02Rhodderzah that might be where i got confused then sry
21:05:15bluebrothersame with Ipodlinux, which runs linux on the Ipod but does NOT modify the original firmware.
21:05:26 Join avis [0] (n=ident@pdpc/supporter/student/avis)
21:06:06 Quit jfc (Read error: 104 (Connection reset by peer))
21:07:30Rhodderzso in easy terms you "add" your content to the ipod so the ipod thinks its genuine
21:08:08BigBambiRhodderz: No, we replace the ipod operating system with our own
21:08:25BigBambiRhodderz: Like replacing windows with linux for instance
21:08:58Rhodderzright cool got ya. cheers for being "newbie". so where is it usually place?
21:09:09BigBambiwhat useual place?
21:09:43BigBambiand where is what?
21:09:45Rhodderzlike windows os is in C:/WIndows where is the ipod and how did you get to it
21:10:01BigBambiIt is on a separate firmware partition
21:10:17BigBambiHave a look at the Rockbox website for more details on how it works
21:10:31bluebrotherit's on c:\Ipod ...
21:10:41Rhodderzso there has to be a way to it
21:10:48BigBambibluebrother: Don't be mean :)
21:10:59bluebrotherBigBambi: oops :)
21:11:06BigBambiRhodderz: Why? That is on the ipods that Rockbox works with, not the new ones
21:11:17BigBambiRhodderz: And do you know what "encrypted" means?
21:11:51Rhodderzyes but every encryption can be broken with time and patience with a little help of a mini fridge
21:12:02BigBambiRhodderz: Then please do
21:12:15BigBambiRhodderz: I'm sure you can even though nobody has managed to yet
21:12:20bluebrotherwell, depends on how little "little time" is.
21:13:03bluebrotheran encryption is usually considered safe if it can't be brute-force broken within reasonable time −− that is several years of calculation
21:13:36Rhodderzthe best way to do it is find a program which works out "similarities" that might get you somewhere
21:13:53 Join z35 [0] (
21:13:54gevaertsSure. Go ahead
21:13:55bluebrothersimilarities to what?
21:14:01BigBambiRhodderz: Don't talk bollocks about a subject you clearly know nothing about
21:17:11*bluebrother considers bumping rbutil to 1.1 −− we now don't need to update rbutil on a new Rockbox release ...
21:17:40bluebrothertoo bad the OS X issue still persists :(
21:18:13rasherbluebrother: How do we handle bootloaders? Do we use the same bootloader when installing release and current builds?
21:18:16BigBambiWhat do people think to a more major version number bump of rbutil with each Rockbox release, and more minor ones for any intermediate fix versions released?
21:18:26BigBambiOnly if needed
21:18:32Rhodderzan encryption works is mainly a password and that "encrypts" the file/message with that password, so if you find simularities and work out each sequence in time( in know it may be long) but you might just find the password. anywa it might be something stupid like the serial number or it may evan be in the firmware update file you download when you update or restore it
21:18:37bluebrotherrasher: we aleays install the most recent bootloader from the download server
21:18:40Rhodderzsry had to go to toileyt
21:18:56rasherbluebrother: Couldn't that potentially be problematic?
21:18:57BigBambiIf the new Rbutil doesn't need updating with each RB release anymore then ignore me :)
21:19:03kadobanRhodderz: that is word soup, sorry to say
21:19:12gevaertsRhodderz: you're about three centuries out of date. Modern encryption is slightly more advanced
21:19:17bluebrotherRhodderz: you know about asymetric encryption?
21:19:18BigBambiRhodderz: That is still stupid and still shows a lack of understanding
21:19:19 Join calman_ [0] (n=caleb@
21:19:32*bluebrother throws in the word "entropy"
21:19:37 Join woyciesjes [0] (
21:19:46Rhodderzso is the ipod "hidden" partition like a shadow partition?
21:20:14bluebrotheron the older Ipods it's simply a partition that is not visible in Explorer
21:20:26Rhodderzwhere you can not see it at all as it is run by the machine and the only way to get to it is "change" the machine's bios
21:20:45bluebrotherBigBambi: that's the point: rbutil now gets information about the most recent release from the download server, so no update needed with a Rockbox release :)
21:20:46LambdaCalculus37Rhodderz: That clearly shows that you haven't got a single idea of how this works.
21:21:11BigBambibluebrother: Ah, cool. That sounds better :)
21:21:25rasherbluebrother: I think the bootloader situation isn't terribly good though. A release should be static.
21:21:35bluebrotherrasher: I don't think that could be dangerous. In the last years I've only seen lack of forward compatibility ;-)
21:21:37woyciesjeskugel: one more hitch. Came back from the meeting, and the iPod wouldn't wake up from when Rockbox put the display to sleep for the hour. I had to give it the 2finger to bring it back. Not sure if it's related to the Apple firmware bug (mentioned this morning on the dev mailing list)
21:21:37 Join jfc [0] (
21:21:48bluebrotherwell, we could add that (like in 1.2 ;-)
21:22:01Rhodderzwhat happens if you take the hard drive out and find a way to plug that into the computer without the ipod there at all
21:22:09rasherbluebrother: Sure, but think about the current issue with booting Rockbox on usb insert. If we updated the bootloaders now, it'd cause confusion for people installing Rockbox 3.1
21:22:14BigBambiYou mean use a caddy!
21:22:15gevaertsrasher: r17357 doesn't work
21:22:24rashergevaerts: oh darn
21:22:31BigBambiRhodderz: You can see the drive, as you would expect
21:22:42bluebrotherrasher: true. Still it would only cause annoyances and / or confusion, no breakage.
21:22:46BigBambiHowever - ENCRYPTION
21:22:49 Join p3tur [50] (n=petur@rockbox/developer/petur)
21:23:13Rhodderzyes but it could be in the ipod instead not the hdd
21:23:16bluebrotherI do agree that the bootloaders for a release could be considered static. Though this doesn't go in line with the way Rockbox currently installs −− build and bootloader are treated separately
21:23:29BigBambiRhodderz: The data is encrypted
21:23:49rasherbluebrother: I'm pretty firmly of the opinion that when installing a release, a "release" bootloader should be installed alongside it
21:23:50 Nick p3tur is now known as petur (n=petur@rockbox/developer/petur)
21:23:54BigBambiRhodderz: The whole point of encryption is that you cant just put the drive in another computer
21:23:55rasherrather than "whatever's current"
21:23:56gevaertsrasher: oh wait...
21:24:21rashergevaerts: well, if you want me to build some more, prod me. I have a loop set up
21:24:42LambdaCalculus37Rhodderz: If it were as easy as you suggested it to be, someone would've done it already. It isn't. It seriously isn't. You're just blindly spouting things without thinking about how this stuff works.
21:24:54bluebrotherrasher: well, I agree that this makes sense (though I'm not too sure if we should do it that way). Still, the current rbutil doesn't allow that easily
21:25:05Nico_Ppondlife: do you think FS #8894 could make it into 3.2?
21:25:12pondlifeI'd like it to
21:25:20pondlifeUpdate about to arrive
21:25:41Rhodderzokay so the only way is if someone from apple leaked the encryption
21:25:45*Nico_P adds it to ReleaseTodo
21:25:51pondlifeI wonder what impact it has on DSP CPU usage though (when disabled)...
21:26:04pondlifeWould be nice if test_codec could include the DSP.
21:26:52rasherRhodderz: There are several other possible ways. None of them easy. None of them have been done.
21:26:57pondlifeOK, last call for a checkwps user to test
21:27:27rasherpondlife: just checking if checkwps compiles with it?
21:27:29Rhodderzoh well sry for wasting your time
21:27:34pondliferasher: Thanks
21:27:52rasherpondlife: I mean, is that all you need?
21:27:54gartralis anyone intrested in testing fs 9960 and/or 9961 for me? i cant build
21:28:04pondliferasher: It's a good start!
21:28:28pondlifeWould be good to try some basic operations - any file opening, renaming, deleting basically
21:28:29rasherpondlife: It compiles..
21:28:38Rhodderzall we need is apple to have a leak again owell cya
21:28:43 Quit Rhodderz ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
21:28:59*LambdaCalculus37 grumbles about clueless kids with new nanos
21:30:28rasherpondlife: seems to work
21:30:43rashersuccesfully checked cabbiev2 for e200
21:33:12pondlifeNico_P: New patch on give it a whirl!
21:33:18*pondlife goes for food
21:33:30pondlifeWill commit #9964 soon though
21:33:35 Quit robin0800 (Remote closed the connection)
21:34:25 Join Aurix_Lexico [0] (
21:36:17 Join jones_ [0] (
21:36:26jones_Hey there
21:36:27jones_I'm back
21:37:28jones_I just updated to 3.1
21:37:37jones_And my battery still doesn't work
21:38:29*rasher bugs pixelma about FS #9754
21:45:25 Join robin0800 [0] (
21:48:13 Join {phoenix} [0] (
21:49:29gevaertsjones_: my guess is that your battery is just going bad
21:49:53Nico_Ppondlife: I got a prefetch abort with the patch
21:50:19jones_I'll see what I can do
21:50:23jones_Gotta run
21:50:25 Quit jones_ ("Ex-Chat")
21:50:31gevaertsrasher: I've managed to boot the OF with r17357 once or twice, but I can't get it to work anymore (about 10 tries). None of the other builds worked either
21:54:13 Quit FOAD ("I'll be back")
21:57:22rashergevaerts: Sounds awful. I guess we need to go further back, then. Want me to build some?
21:57:27 Join FOAD [0] (
21:57:56 Join kv [0] (i=kosss@
21:58:11Nico_Pam I the only one getting prefetch aborts on the beast?
21:58:15 Quit LambdaCalculus37 (" ajax IRC Client")
21:58:27gevaertsrasher: if you feel like it
21:59:13gevaertsSince it hangs with the bootloader output showing I know I was running the right version
22:01:17 Quit vertic39 ()
22:03:42BigBambiNico_P: Doing what? I haven't experienced it
22:03:52Nico_PBigBambi: just playing an mp3 file
22:04:08rashergevaerts: Building some now
22:04:23BigBambiNico_P: Well, I use it daily and haven't had that - I lat updated 2 or 3 days ago
22:04:49Nico_PBigBambi: it was working for me too until I updated just today
22:04:58Nico_Pright now even
22:05:25BigBambiNico_P: OK, I'll try in a bit and see if I can too (watching the Rugby right now :) )
22:08:39 Quit kv ()
22:11:36 Join gotthardt [0] (
22:12:38pondlifenico_P: Do you get the prefetch abort without even enabling timestretch?
22:12:52Nico_Ppondlife: yes, it's not the patch
22:12:58pondlifePhew ;)
22:13:05gartralcould someone with an e200 please test my patch/es? i want to know if it works before i go making more modifications for other targets...
22:13:58gotthardtyo - anyone with a gigabeat F test the latest build? I can no longer play tracks - seems to reboot ??
22:15:12rashergevaerts: earlier bootloaders uploaded
22:15:23Nico_Pgotthardt: on the beast I get prefetch aborts
22:15:48Nico_Ppondlife: I tested the patch on the ipod video, it appears to be working beatifully :)
22:15:52gotthardtNico_p thx - so its not just me
22:16:44 Quit perrikwp (" ajax IRC Client")
22:16:45 Quit phill ()
22:16:48pondlifeNico_P: If you look at dsp.c, we are now doing buffer size calculations (i.e. a variable, not a constant). Do you think that will have too much of an adverse affect?
22:17:01linuxstbrasher: Thinking about FS #9948, what genlang magic do I need to do?
22:17:14Nico_Pgotthardt: do you think you could try finding the revision that introduced the issue?
22:17:40 Quit rocko ("Leaving")
22:17:50gotthardti can later today -
22:18:25pondlifeI'm tempted to precalculate the various sizes at least... but having the redundant variables goes against my normal practice.
22:19:05rasherlinuxstb: I don't think genlang will be of much help, really. You basically need to loop through all phrases and see if the source matches the current english.lang. If it does, change it to match your new english.lang. If not, leave that phrase unchanged.
22:19:33rasherI'm not sure langtool will be of much help either, to be honest
22:19:37*gevaerts confirms
22:19:45gevaertsGigabeat F has issues today
22:19:46linuxstbrasher: Hmm... Or I could just leave it to translators? ;)
22:20:10linuxstbI guess not then ;)
22:20:35rasherWe could make genlang case-insensitive, but I don't know if that's a great idea
22:20:35pondlifeHmm, some of my last commit went astray, I think...
22:20:41 Quit avis (Remote closed the connection)
22:20:52linuxstbrasher: No, I don't think that's a solution.
22:21:34rasherIt'd really put a rather heavy load on translators though, having to check whether it's a case-only change, or a phrase needing actual attention
22:21:44*linuxstb thinks awk may be his friend here, as he knows it better than perl
22:22:11rasherThat might not be a bad tool for this job
22:22:33rasherCollect the lines between <source> and </source>, and compare
22:23:04rasherDoes awk do associative arrays?
22:24:38rasherSo load current and sentence-case english.lang in BEGIN, start collecting data on <source>, do the compare/possible replace in </source>, and unconditionally print everything else.
22:24:50rasherOr something like that
22:25:38linuxstbYes, that's what I was thinking.
22:25:56*pondlife has no idea how half his patch vanished..
22:26:00rasherI think it's acceptable if you just do a "dumb" compare of the source strings as a single string (no need to compare the target names and values etc)
22:26:20rasherThat way we might even eliminate some syntax issues
22:26:33rasherIf there are any.
22:26:54gevaertsrasher: I tested r16837 and r17062, both don't work
22:27:10rashergevaerts: oh dear
22:27:16 Quit robin0800 ("No Ping reply in 30 seconds.")
22:27:20*rasher wonders if there was ever actual dual boot on H10 5GB
22:27:40 Join robin0800 [0] (
22:27:42rashergevaerts: Building some more.
22:27:44linuxstbrasher: If you're bored, I won't mind if you do it ;) I'm probably not going to get chance until tomorrow evening or Sunday.
22:28:11rasherI might. No promises.
22:29:28Nico_Ppondlife: about your question, I don't really see what you mean
22:30:19pondlifeIf you look in dsp.c, we used to have a fixed RESAMPLE_BUF_RIGHT_CHANNEL - but now it's a calculation.
22:31:08pondlifeAs is SAMPLE_BUF_RIGHT_CHANNEL
22:32:00rashergevaerts: put up a few even older ones
22:32:28Nico_Ppondlife: I don't know, the calculation doesn't seem too complex but how often is it done?
22:32:43pondlifeNot sure.. quite often , I think
22:32:58pondlifeA redundant variable would probably be worthwhile
22:33:52pondlifeI just don't like redundancy
22:35:19pondlifeAlso, do you see the weird way that the splash appears when you first enable timestretch?
22:35:20***Saving seen data "./dancer.seen"
22:35:28pondlifeI have no idea why that happens
22:36:21*linuxstb senses 3.2 could have quite a nice changelog...
22:36:45gevaertsrasher: r14837 doesn't work either... I'm leaving this to someone who knows the H10
22:36:49*pondlife only wishes an H300 bootloader was included...
22:37:02rashergevaerts: And you're sure you're doing it right?
22:37:18rasherThe guy in the bug report is willing to help also
22:38:09pondlifeNico_P: Was it you who suggested #8894 should be in 3.2?
22:38:38Nico_Ppondlife: yes
22:38:47pondlifeI'm more of the opinion it should wait until after, and have the UI improved (i.e. merged with the pitchscreen if possible)...
22:38:58pondlifeBut I suppose it's nearly ready
22:39:40gevaertsrasher: it works with r12862 (the released bootloader)
22:39:50gevaertsSo yes, I think I'm doing it right
22:40:26rasherAh, so we probably just need to go further back, then
22:41:07gevaertsMaybe, but I'm too tired right now...
22:41:11rasherOh wait!
22:41:16rasherI wasn't doing it right, was I?
22:41:34rasherI never corrected that...
22:41:36 Quit FlynDice (Read error: 110 (Connection timed out))
22:42:05*gartral isnt used too seeing rasher flood the room
22:42:05gevaertsThe latest daily build works on gigabeat F, so the bug was introduced after 20122
22:42:22Nico_Ppondlife: I agree the UI isn't quite ready
22:43:55pondlifeToday's version is better, but still I'd hope to work it into the pitch screen. Perhaps UP/DOWN would change pitch in the current way and left/right would speed up/slow down with timestretch. Then we need to reallocate the +/-2% jog functions...
22:44:46gartralmaybe add a return to 100 thing? or is there already an easy way of doing that?
22:44:56rashergevaerts: Okay, I now inserted the required ../tools/scramble -mi4v3 bootloader/bootloader.bin H10.mi4 in my loop. Building new bootloaders now.
22:45:13gevaertsrasher: that's not the issue. This H10 is now a UMS one...
22:45:32gevaertsAnd the wrong scramble line just makes it not boot at all
22:45:32pondlifeMaybe the jog mode should be a third option (pitch up/down, semitone up/down, jog up/down)?
22:45:39Nico_Pgartral: thetre's already a return to 100 button, on the ipod at least
22:45:55Nico_Ppondlife: I don't undersand the use of the jog mode
22:46:04pondlifeIt's for beat matching
22:46:18gartralNico_P: the e200s dont, that im aware of
22:46:26pondlifeSay you're almost in sync, but you want a little "pulse" to sync better...
22:46:29Nico_Pgartral: it's the menu button
22:46:56Nico_Ppondlife: ah, I see
22:47:15rashergevaerts: Alright, building both types now
22:47:34pondlifeI know it's doable on H300, but that is blessed with buttons
22:47:35gartralNico_P: no, that returns to wps while keeping the scale
22:47:42pondlifeIt's SELECT on H300
22:47:49pondlife(aka NAVI)
22:48:07pondlifeMENU toggles the semitone mode
22:48:14Nico_Pinteresting. on ipod select returns to the WPS
22:48:14gartraloh, im just f*ck all retarded, nvm!
22:48:54pondlifeOn H300, PLAY and STOP both return to the WPS
22:49:00Nico_Ppondlife: the ipod lacks a button. maybe timestretch could be a third mode (along with normal and semitone)
22:49:17pondlifeNo, you probably want to use both timestretch and pitch together.
22:49:37pondlifeTo change key whilst slowing it down, say for guitar practice?
22:49:37Nico_Pthen they could simply be two separate screens
22:49:55pondlifeWell, that's what we have now..!
22:50:04 Quit gregzx (Read error: 110 (Connection timed out))
22:50:37Nico_Ppondlife: yes, but the timestretch screen could be similar to the pitch screen, and also be accessible through the WPS context menu as the pitch screen is
22:50:53pondlifeMaybe commit it as is, then start a new (even more controversial) patch for a UI...
22:51:07pondlifeI'm still slightly in fear of the red delta
22:51:13gevaertsmidgey: r20124 broke codecs on gigabeat F (and I guess S). Not sure about others
22:52:00midgeyon target or on the sim?
22:52:05Nico_Ppondlife: IMO the delta shouldn't matter for a feature like that
22:52:38*gartral tests on e200
22:55:21rasherNico_P: Keeping it as low as possible matters though.
22:55:27pondlifeNico_P: I guess it's just the silly splash screen that needs fixing before commit then... Did you see the same problem?
22:55:34*pondlife gets repetitive
22:55:46Nico_Ppondlife: what's the problem?
22:55:56Nico_PI saw the splash screen, but nothing wrong with it
22:56:30pondlifeWhen you enable timestretch for the first time, the splash appears after it exits the menu, and stays until a key is pressed
22:56:46pondlifeIt should appear for 2 seconds BEFORE exiting the menu
22:56:48Nico_Pah yes, it did that
22:56:51gartralcodecs are working A-OK here
22:57:14Nico_Pgartral: the problem seems to only affect gigabeats
22:58:23BigBambigartral: As has been said
22:58:24gevaertsrasher: maybe tomorrow...
22:59:29linuxstbmidgey: IRAM? (Or rather, lack of IRAM) ?
22:59:37 Part woyciesjes ("Ex-Chat")
23:01:11 Quit evilnick (" ajax IRC Client")
23:01:52midgeyseems like a reasonable guess
23:02:05midgeyi'm not sure why that would affect targets at all
23:02:22amiconngevaerts, midgey: It's quite obvious why r20124 breaks codecs (and probably plugins) on target
23:02:28linuxstbLooking at your change to config.h, you've only defined DATA_ATTR for a subset of targets...
23:02:37midgeyplugins work for me
23:02:49amiconnci (and possibly rb) must stay in .data
23:03:23midgeydamn, you're right
23:03:24amiconnIf they are in .bss, at least ci gets overwritten by bss clearing in codec_crt0
23:03:38midgeyi didn't look at the preprocessor stuff
23:04:22pondlifeWhat's holding up FS #8314 - kugel's sorting stuff?
23:04:40amiconnI didn't think of this in the morning, when Unhelpful asked whether that change might cause problems
23:05:26amiconnWhat is the problem re using the .data section on mac?
23:06:24Nico_Pcan I go ahead and remove 3.0 and 3.1 from ReleaseTodo?
23:06:46pondlifeYes, they are NoToDo anymore
23:06:55pondlifeNotToDo, even
23:07:02midgeygcc and the assembler don't agree about the assigned section
23:07:03pondlifeAnd we have wiki history
23:07:05Nico_Ppondlife: about FS #8314, I think it's name issues mainly
23:07:40pondlifeI thought the current wording was uncontroversial.
23:08:04Nico_Poh, maybe it is
23:10:25amiconnmidgey: Hmm, maybe the data section has a different name on macos?
23:10:35amiconnThe assembler seems to complain about the dot
23:10:58amiconnThere might be a better solution though
23:12:09amiconnPlugins are indeed unaffected. There's an inconsistency between plugins and codecs:
23:12:33amiconnFor plugins, the loader does the bss zeroing, while codecs do it themselves
23:12:44midgeythe apple documentation seems to suggest .data is fine
23:13:03amiconnIf we move the bss zeroing for codecs into the core, that would fix both the incosistency and the crashes
23:13:27midgeyprobably a good idea
23:13:30 Quit robin0800 ("No Ping reply in 30 seconds.")
23:14:04 Join robin0800 [0] (
23:14:35amiconnThe crash affects all targets which don't use iram for codecs (i.e. at least Gigabeat F/X, Gigabeat S, and lowmem AMS Sansas)
23:15:25 Quit robin0800 (Client Quit)
23:15:54 Join robin0800 [0] (
23:19:55 Quit low_light (" ajax IRC Client")
23:21:08 Join robin0800_ [0] (
23:21:37Nico_Pany objections to my making the release annoncements on the frontpage link to the release notes?
23:21:47Nico_P(instead of the download page)
23:22:22gevaertsNico_P: not from me
23:23:05 Quit petur (Remote closed the connection)
23:26:47*amiconn prepares a test on the beast
23:27:07midgeyanyone want to test out this patch
23:27:21midgeyseems to fix the gigabeat f problems
23:27:37gartralmidgey: you should probably start a fs for that...
23:27:41midgeyi had to do a make veryclean first
23:28:30midgeygartral: it's a quick fix for the breakage i introduced in r20124. If it fixes the problems for others, i'll just commit it
23:29:30 Join robin0800__ [0] (
23:30:27*amiconn just notices that it is pure luck that plugins are working, and this inconsistency should most probably be fixed the other way round
23:31:27amiconnThe problem is that the iramcopy overlaps bss. The plugin loader doesn't clear all bss, but just the part after the iramcopy (which is part of the loaded binary image)
23:31:42amiconnThe rest of it is cleared when a plugin calls plugin_iram_init()
23:32:24amiconnSo if 'rb' is placed in .bss, and a plugin uses iram, it depends on where the linker places it whether the plugin breaks or not
23:33:18 Quit HellDragon (Client Quit)
23:33:22 Quit CaptainKwel (" ajax IRC Client")
23:33:58 Quit jgarvey ("Leaving")
23:34:32amiconnThe problem with a (not yet existing plugin_crt0 would be that unlike codecs, only some plugin should init iram, not all of them
23:34:40 Quit linuxstb (Remote closed the connection)
23:34:42gevaertsmidgey: that fixes it here
23:35:34 Join mc2739 [0] (
23:35:37midgeygood to hear
23:35:47 Quit Febs (Read error: 60 (Operation timed out))
23:35:52amiconnThat fix should be committed imo
23:35:52 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
23:35:54kadobani'm getting a segfault on starting rockblox in the e200 simulator in r20133. anyone else?
23:36:00*gartral wonders why chessclock tries to re-assume controll after a usb plug/unplug...
23:36:01 Quit robin0800 (Read error: 113 (No route to host))
23:36:03amiconnImproving plugin startup is independent of that
23:36:35midgeyamiconn: i'll probably commit within 5 minutes. im experimenting with the section attributes on mac
23:37:09amiconnYou could check what asm gcc produces
23:38:46 Quit mc2739 (Client Quit)
23:38:47gotthardtNico_P I binary searched the builds down to: r20123 plays and r20124 does not for the gigabeat F40
23:39:20BigBambigotthardt: The fix is about to go in, see above :)
23:39:31 Quit flydutch ("/* empty */")
23:39:36gotthardtheh - cool
23:39:46 Join HellDragon [0] (
23:39:58amiconnHmm, actually it wouldn't make much sense to unify plugin and codec startup
23:40:27amiconnI'm a little undecided
23:41:20 Quit {phoenix} (Remote closed the connection)
23:41:25amiconnWith delayed iram init, plugins could apply special tricks like using iram (and hence stopping playback) only after asking the user, or just optionally if music is stopped.
23:42:52amiconnBut there is a *really nasty* gotcha: Since iramcopy and bss are shared, using bss before initing iram *might* mess up the iram copy, and hence crash the plugin later
23:43:56 Quit robin0800__ (Read error: 104 (Connection reset by peer))
23:44:12amiconnThe solution would be to make the sections disjunct - at the cost of allocating a larger part of the plugin buffer for plugins using iram
23:45:25amiconnIf we don't want to use such tricks, startup could be unified to make things easier for plugin devs.
23:47:15*amiconn just found the solution for the two startup variants (plugins with and without iram usage)
23:47:42amiconnplugin_crt0() could check iramsize. If it is zero, the plugin doesn't use iram -> no playback stopping
23:48:41*midgey might have a slightly more correct version of the fix for mac
23:49:55midgeyusing .section __DATA, __data instead of .data seems to work
23:51:35 Quit calman_ ()
23:51:39amiconnSo #ifdef __APPLE__ / #define DATA_ATTR __attribute__((section("__data"))) / #else / #define DATA_ATTR __attribute__((section(".data"))) / #endif
23:52:24 Quit robin0800_ (Read error: 113 (No route to host))
23:54:54kadobanmidgey: i think you're already looking at it (missed some chatter), but r20124 broke rockblox in simulator on linux
23:55:49linuxstbkadoban: Are you sure that wasn't r20113?
23:56:13kadobanlinuxstb: pretty sure, unless i bisected wrong
23:58:20 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
23:58:42 Quit homielowe ()

Previous day | Next day