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 2010-01-09

00:00:02Spaceghostare different things
00:00:25AlexPSpaceghost: I don't know what you mean by "sports pictureflow"
00:00:54Spaceghostsorry, I used the google translate that time xD
00:01:00AlexPhehe :)
00:01:05Spaceghostwasn't good to trasnlated
00:01:05pamaurySpaceghost: I'm not sure but that's was just a joke to say that the pictureflow plugin was useless
00:01:41*pamaury don't remember who maintain the pictureflow plugin for fun
00:02:12JdGordonamiconn: even if the selection list can be stored on the files (or by the list caller in general), there is still 2 problems
00:02:38JdGordon1) the list needs to know which to draw as "selected" which could be done via the text callback through a function call (yuck!)
00:03:02JdGordonand 2) a way to enumerate the selected list after (might not be needed to be done in a generic way though)
00:03:14Spaceghost* pamaury wonders how many stupid-things-like-the-nike+-ipod-kit the iPod Touch/3GS/... supports ! There are as useless as the pictureflow plugin ;)
00:03:57*JdGordon slaps Spaceghost
00:04:02JdGordonpf is very useful!
00:04:19pamauryah that's you JdGordon who maintain this funny plugin
00:04:29GodEaterJdGordon: he was quoting pamaury
00:04:34AlexPUnhelpful last worked on it IIRC
00:04:35GodEaterhe didn't say that personally
00:04:39 Join ifonefox [0] (n=irchon@
00:04:42SpaceghostI did not say that
00:04:48*pamaury prepares to being slapped
00:04:52*JdGordon sneaks away
00:05:05JdGordonand no, i dont think i've ever touched the pf code
00:05:13*B4gder takes all the blame. sorry guys, my fault
00:05:23AlexPAh, the fall guy is here :)
00:05:52*B4gder wears shades and earpiece
00:05:54Spaceghostpamaury, do you think that the Nike accesory not helpful?
00:06:12 Quit ifonefox (Client Quit)
00:06:29 Join ifonefox [0] (n=irchon@
00:06:45pamauryI'm wondering how useful is it ? I mean, how many people really use this ?
00:07:15 Quit ifonefox (Remote closed the connection)
00:08:24SpaceghostI think it's an interesting idea, pity it is so exclusive and limited
00:08:51JdGordonthere is no correlation between somethings usefuleness and how much its used... *especially* in apples world
00:08:55 Quit evilnick (Ping timeout: 180 seconds)
00:09:18pamauryJdGordon: do you think it would be useful to implement item selection in the listview ? Would other areas (apart from *would be* multi-select) use it ?
00:10:24pamauryit costs money to develop it so apple would sell it only if people buy it, no ?
00:11:25*JdGordon was more thinking about the numerous fart iphone apps...
00:11:38JdGordonbut yes, multiselect is for sure worth implementing
00:12:59pamauryWould it require lots of work to have a listview with selectable items ?
00:16:15bertrikpamaury, I'd say a moderate amount
00:16:59bertrikFirst we'd have to add support to do the multi-select (agree on a way to do this in the UI)
00:17:16 Quit jgarvey ("Leaving")
00:17:55***Saving seen data "./dancer.seen"
00:17:58pamauryHow exactly it works ? I mean, a listview needs a state, when you browse fil you're in a listview with an associated state. But when you do a context menu, there is a another list-view, so is the state of the previous one deleted ? (I know nothing of the gui part)
00:17:59bertrikSeconds, we need to think how for each action to do something meaningful when multiple items are selected and activated
00:18:56 Quit Spaceghost ("güit")
00:21:18 Quit bmbl ("Bye!")
00:25:50JdGordonpamaury: the tree and the context menu use different lists so the state is known
00:25:59JdGordon.. rememebered
00:26:41pamauryOk that's a per-list basis then.
00:29:01JdGordonmost of the work is on the callers side, not the list side
00:32:08 Quit togetic (Read error: 110 (Connection timed out))
00:32:44 Join evilnick [0] (i=457288ab@rockbox/staff/evilnick)
00:34:53 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
00:36:11 Quit Barahir_ (Read error: 104 (Connection reset by peer))
00:37:29 Join Barahir [0] (
00:40:01 Join GeekShado_ [0] (
00:41:05 Quit Utchybann (Read error: 113 (No route to host))
00:42:59 Quit mcuelenaere ()
00:43:20 Quit GeekShadow (Read error: 60 (Operation timed out))
00:44:38 Quit bertrik ("De groeten")
00:44:39 Quit CaptainKewlllll ("Page closed")
00:50:53 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
00:52:16 Join fdinel [0] (
01:00:52Unhelpfuluhoh, where'd pf break?
01:02:03 Quit togetic (Read error: 110 (Connection timed out))
01:06:22 Quit robin0800 (Remote closed the connection)
01:07:04 Quit pamaury ("exit(*(int *)0 / 0);")
01:07:39 Quit GeekShado_ (Read error: 110 (Connection timed out))
01:09:44 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
01:10:30 Part toffe82
01:11:24 Quit n17ikh (Read error: 104 (Connection reset by peer))
01:11:53 Join n17ikh [0] (
01:22:46 Quit petur ("Zzzz")
01:24:21 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
01:52:48 Quit GeekShadow ("The cake is a lie !")
01:55:04 Join CaptainKewl [0] (
02:09:33 Join FlynDice [0] (n=FlynDice@
02:14:29 Nick advcomp2019__ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
02:17:59***Saving seen data "./dancer.seen"
02:20:38 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
02:22:08 Quit GodEater (Read error: 113 (No route to host))
02:34:36 Join fdinel [0] (
02:51:10 Join kramer3d [0] (n=kramer@unaffiliated/kramer3d)
03:00:12 Quit chrism (
03:00:44NJoinchrism [0] (
03:01:55 Join toffe82 [0] (
03:05:17 Join Casainho [0] (
03:17:20 Quit domonoky (Read error: 104 (Connection reset by peer))
03:18:52 Quit fdinel (Read error: 110 (Connection timed out))
03:25:38 Quit evilnick (Ping timeout: 180 seconds)
03:30:15 Quit Casainho (Remote closed the connection)
03:31:30 Join Casainho [0] (
03:36:01 Quit Kopfgeldjaeger (Read error: 104 (Connection reset by peer))
03:36:38 Join Kopfgeldjaeger [0] (
03:40:06 Quit toffe82 (Read error: 60 (Operation timed out))
03:41:26 Join handeh [0] (
03:41:55handehI have an ipod 5.5g will it work with rockbox?
03:43:54 Join fdinel [0] (
03:49:34Unhelpfulthere's a list of supported players on the front page at
03:50:23handehWell I have installed it on the ipod, but it doesnt change when i disconnect the ipod
03:53:43 Quit togetic (Read error: 110 (Connection timed out))
03:54:56 Join Spaceghost [0] (i=quassel@unaffiliated/spaceghost)
03:55:26 Quit Casainho ("ChatZilla 0.9.86 [Firefox 3.5.7/20100106054534]")
03:55:36 Quit handeh ("CGI:IRC (EOF)")
04:01:25 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
04:09:47 Join GodEater [0] (
04:16:58 Quit panni_ (Read error: 104 (Connection reset by peer))
04:18:02***Saving seen data "./dancer.seen"
04:25:52 Quit Oxymis (Read error: 60 (Operation timed out))
04:49:11 Join Space [0] (
04:50:37S_a_i_n_thandeh: have you installed the bootloader?
04:51:46S_a_i_n_twhoops, he's gone....meh :/
04:54:51Spacewhat is the target with the better sound?
04:59:49 Join Barahir_ [0] (
05:00:43Spacethe earlier ipods have a good sound?
05:05:54 Join saratoga_ [0] (i=463f90ed@gateway/web/freenode/x-duzeipiejepacppa)
05:08:33 Quit Spaceghost (Read error: 110 (Connection timed out))
05:12:41 Quit Barahir (Read error: 110 (Connection timed out))
05:17:52S_a_i_n_tearlier iPods?
05:18:06S_a_i_n_tmy nano 1st gen sounds great rockboxed...
05:18:12S_a_i_n_t2nd gen good also..
05:29:35 Join Benjamin_ [0] (
05:31:31 Nick Benjamin_ is now known as Batmanifestdesti (
05:32:12BatmanifestdestiSo, which mp3 player for under $50 would you guys reccomend for rockbox?
05:34:41 Quit saratoga_ (Ping timeout: 180 seconds)
05:34:53advcomp2019Batmanifestdesti, the sansas e200 and c200 series
05:35:05Batmanifestdestiokay, I'll check that out
05:40:57*Unhelpful thinks he's found his umod32 bug, and feels very very very very stupid
05:41:07Unhelpfullsr where i need lsl :/
05:42:14Batmanifestdestithat's happened to me before with semicolons.
05:43:02Unhelpfulworse, it was add pc, pc, offset, lsr #3... instead of lsl #3. amazing it never *crashed*.
05:48:32 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
05:49:09 Nick YPSY is now known as Ypsy (
05:55:23Batmanifestdestiman, it seems like all the mp3 players in my price range available for rockbox are no longer sold in stores.
05:56:45S_a_i_n_t<Batmanifestdesti>: pretty much a given that you'll be getting it second hand...
05:57:24S_a_i_n_tespecially if you're looking at the ipod family
05:57:48BatmanifestdestiIf only my dad would realise that some things are good secondhand. He's had me shopping at Target and stuff for mp3 players, and I want one that supports rockbox.
05:58:35S_a_i_n_tnothing wrong with second hand, you dont need to but it from some dodgy guy out of the bewspaper...there are refurb shops
05:59:26S_a_i_n_tgoogle 'model mp3 you want+refurb+your location
05:59:45S_a_i_n_tshould get *some* hits at least.
06:03:08maraz^/op failbot
06:04:43Unhelpfulhrm, is it worth making an add-to-pc instruction conditional? will it still cost as much as a branch if skipped?
06:09:08 Join ruckus1 [0] (
06:10:10ruckus1hey peoples
06:11:27Unhelpfulbetter still, special-case divisor-is-one and then never do the first iteration...
06:12:05ruckus1i've been having an interesting problem with my 5.5 gen ipod. i managed to rip its ribbon cable that connects the ipod with the hold button and the headphones, and i replaced it with another one. after i did so, the feature "pause on headphone unplug" ceased to work, and instead of pausing the music continues to play
06:12:39ruckus1does anyone have ideas what can i do to get it to work again?
06:13:04Unhelpfulruckus1: figure out what went wrong with your repair, i'd guess. ;)
06:13:22ruckus1Unhelpful: irony ?
06:13:31ruckus1i mean, the ipod works perfectly
06:13:51Unhelpfulexcept that it can no longer detect the headphone disconnecting.
06:14:44ruckus1i'm thinking i bought a compatible ribbon from another ipod generation
06:14:54ruckus1but the ebay page doesn't say anything about it
06:15:29Batmanifestdestiwith the whole "sansa e200 series" support gor rockbox, does that include e260, e240 etc?
06:17:10UnhelpfulBatmanifestdesti: e2xx v1 are supported. later revisions, which (thanks sansa) are not distinguishable from labeling or external features, are in development
06:17:48Batmanifestdestiwow, so I'd be sitting around going "hurr hurr, which one do I have?" until I found out I had the wrong one, eh?
06:18:06***Saving seen data "./dancer.seen"
06:19:32UnhelpfulBatmanifestdesti: well, no, if you can start it up, you can check the vendor firmware version. also certain sites sell refurbs that are reliably v1.
06:19:51Batmanifestdestioh, ok
06:20:08Batmanifestdestiso, with rockbox, you can boot it from the original firmware, right?
06:20:17Batmanifestdestithat's what the unstable port site says
06:20:24Unhelpfulso, if you're getting it used from ebay, you could ask the seller...
06:22:29Batmanifestdestiholy cow the early archos players are large
06:40:50 Part ruckus1 ("Leaving.")
07:02:05 Quit killan (Read error: 60 (Operation timed out))
07:05:28 Nick Space is now known as Spaceghost (
07:07:31 Quit Batmanifestdesti ("bed time")
07:13:49 Join Horschti [0] (n=Horscht2@xbmc/user/horscht)
07:32:06 Quit Horscht (Read error: 110 (Connection timed out))
07:33:35 Quit StealthyXIIGer (Read error: 110 (Connection timed out))
07:54:21 Quit S_a_i_n_t ("S_a_i_n_t [St.] has left")
08:05:15 Quit Strife89 ("Bed.")
08:14:59 Join Llorean [0] (
08:18:08***Saving seen data "./dancer.seen"
08:18:27 Join bertrik [0] (
08:23:47 Join CaptainKwel [0] (
08:23:48 Quit CaptainKewl (Read error: 54 (Connection reset by peer))
08:27:56 Join Space [0] (
08:42:00Unhelpfulhrm, can i assign a register name, etc, to a variable in an asm macro, and use it as i would one of the arguments? i'm trying to figure out how to do this... esentially i need to use a different one of the registers passed as arguments for something based on ARM_ARCH, and it would be much more messy to duplicate the entire macro with the changes :/
08:42:31 Quit Spaceghost (Read error: 110 (Connection timed out))
09:05:34 Join flydutch [0] (
09:10:08 Quit mc2739 (Read error: 110 (Connection timed out))
09:16:16 Quit bertrik ("De groeten")
09:17:26 Quit Space (Read error: 110 (Connection timed out))
09:19:13 Join Spaceghost [0] (i=quassel@unaffiliated/spaceghost)
09:22:28 Join killan [0] (
09:23:15 Nick Barahir_ is now known as Barahir (
09:23:49 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
09:27:55 Join robin0800 [0] (
09:31:58 Quit robin0800 (Remote closed the connection)
09:33:39 Quit BHSPitLappy (Read error: 110 (Connection timed out))
09:36:23 Join einhirn [0] (
09:43:30 Join S_a_i_n_t [0] (i=S_a_i_n_@
09:43:51 Quit einhirn (Read error: 60 (Operation timed out))
09:50:52 Join pamaury [0] (
09:50:56 Join stoffel [0] (
09:53:21 Join Space [0] (
09:53:47amiconnUnhelpful: Did you test on the beast?
09:54:17amiconnOtherwise I would expect the lsr case to crash with a data abort if the address ends up being unaligned
09:56:07amiconnHmm, or maybe not...
09:58:19 Join arbingordon [0] (
09:58:50arbingordonhello, I'm trying to install rockbox on my sisters sansa e250v2
09:59:04arbingordonbut I can't find the usb options under settings
10:01:18Unhelpfulamiconn: well, the lsr was a mistake anyway, and is fixed now. i found more things to debug when i got to the armv5e/v6 version that uses clz to align the inputs instead of a shift/compare sequence. i now have unsigned div, divmod, and two mod implementations working on beast and e200. signed already works on e200, but will need some fixups to parallel the ones i did for unsigned (the divider body inverts the divisor early, and obviously
10:01:18Unhelpful clz needs the positive divisor)
10:03:14Unhelpfuli've been looking for a good place to fit in a 1-divisor or pow2-divisor test... these were very nicely optimized in the APE version but the zero-divisor test already fits to an op that the new divider needs, so calculating divisor - 1 would only be used for these new tests.
10:03:55Unhelpfuland i'm still not sure it's really important to special-case these, although dividing by one of course perform abysmally, probably about 80-100 cycles, i'd think
10:11:13 Quit Spaceghost (Read error: 110 (Connection timed out))
10:16:40 Quit bmbl (Excess Flood)
10:16:57 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
10:17:29arbingordonnvm, i figured it out, outdated, crappy sansa firmware keeps you from having MSC access to it
10:17:34arbingordonthanks i guess
10:17:36 Part arbingordon
10:18:09***Saving seen data "./dancer.seen"
10:24:41Unhelpfulit seems the best way to do things like swapping registers used, etc, in a macro is to wrap it in another macro? at least, this solution seems to work. the issue is that the armv5+ version needs the non-inverted divisor for the clz, but can still do the inversion early by storing to a different register and then using the divisor register as a scratch register.
10:26:06Unhelpfuland it works for signed on beast now, too. the only thing left to do is a signed version of the specialized mod-only function, and of course benchmarking these against each other and against libgcc's functions
10:29:34 Join Utchybann [0] (
10:32:05S_a_i_n_tQuestion: In 'Advanced Rockbox install methods with ipodpatcher' do I just use 'ipodpatcher' as a command-line argument (as in 'ipodpatcher -rf apple_os.ipod')? or is it talking of something entirely different?
10:32:47S_a_i_n_tadditionally, what is the 'best' (subjective) method to use?
10:43:01amiconnUnhelpful: What about this newton-raphson implementation?
10:46:49amiconnI wonder how an approximating loop can be faster than the linear test-and-subtract
10:50:00Spacehow can I do a full backup of the ipod (with all her music) before to try install rockbox?
10:50:13Spacecan I do a ISO image of that in linux to then can be restore?
10:50:49 Quit kramer3d ("Leaving")
10:54:33Unhelpfulamiconn: because trial subtraction takes one iteration per bit of output, while newton-raphson approximation of 1/d, with a good starting estimate, does a good deal better. it appears the sample divider does 2-3 N-R iterations to get a result that is close enough, and gets the last 2 bits or so via trial subtraction.
11:01:15tmztSpace: can't you just copy the mp3 files from the iPod's harddrive?
11:01:47tmztalso, iso is not a generic term for disk image, it's for compact discs, it's the iso9660 filesystem
11:02:03S_a_i_n_tSpace: though it's generally a good idea toback up files, everything *should* be fine as long as you don;t botch the install...
11:02:09S_a_i_n_tand that's pretty hard to do
11:02:36Spacewhat is hard to do?
11:02:50S_a_i_n_tbotch the install of rockbox
11:02:59S_a_i_n_tyour files *should* be fine
11:04:12S_a_i_n_tif you're installing rockbox, it's probably best to start from scratch the iTunes Database is shit.
11:05:48Spacebut I don't have the fonts that all music that I have on it
11:08:15S_a_i_n_tsorry dude, english is my first language, so you may have to try a little harder. that doesn't make any sense.
11:08:41SpaceI have a lot of music only in my ipod
11:08:54Spaceand not in the computer
11:09:14S_a_i_n_tsync your iPod with iTunes...or whatever you used to put the music on there in the first place
11:09:14 Quit grndslm (Read error: 60 (Operation timed out))
11:09:24S_a_i_n_tthen you'll have it backed up as a library
11:10:10Spacewith iTunes can I extract the music from iPod?
11:10:57S_a_i_n_tyes, it's called 'syncing', you just sync the music on the iPod with the computer
11:12:57 Join twick [0] (
11:14:55Spaceohh, I have a problem with him now
11:18:16S_a_i_n_ta problem with what sorry?
11:20:17S_a_i_n_tSpace: are you using a Mac or Windows based PC?
11:20:22SpaceI eject, but then reconnect and disconnect it when it was reconnecting
11:21:12Spacenow for restarting, I can see my bookshelves but only for a few seconds then restarts
11:21:53Spaceand when I connect I can't access
11:22:14Spacewindows says that he need be restaured by iTunes
11:22:34 Quit xavieran (Client Quit)
11:22:54S_a_i_n_twell, if prompted to restart, it's a good idea to do so, so try that first.
11:23:18SpaceI have tried
11:23:37Spacehe continous do the same, restart and then...
11:24:18SpaceI will boot linux, may be can I see the files on hdd on it?
11:24:21Unhelpfulouch, i am *slower* on beast than __aeabi_uidiv... although perhaps that's on account of my test values specifically *not* being uniformly distributed and the divider not special-casing 1 or powers of 2
11:24:50S_a_i_n_tSpace: what type of iPod is it?
11:26:01Spaceipod video
11:26:27Spaceis 5G
11:27:43S_a_i_n_thave you installed the bootloader? ipodpatcher
11:28:01S_a_i_n_tthere you go...
11:28:32S_a_i_n_tare you installing manually, or using rbutil?
11:28:51Spacebut, I didn't do anything only I disconnect badly I think
11:29:27Spacebecause I expulse same a external hdd maybe, and iTunes continous working, but
11:29:53S_a_i_n_tyou need to install the bootloader AS WELL as the .rockbox folder, otherwise rockbox won't run
11:30:46SpaceI didn't start the installation
11:30:58S_a_i_n_tspace: are you installing manually, or using rbutil?
11:31:14Spaceso may be I don't have talking about it here
11:32:14Spaces/don't have/musn't/
11:32:33pixelmaSpace: do you mean that your Ipod isn't working currently?
11:32:43S_a_i_n_twell, if it's not a rockbox problem, then maybe...
11:33:30Spaceyes, he are restarting and restarting
11:33:45Spaceand only can see the interface for a few of seconds
11:34:38pixelmadid you try to hard reset it?
11:35:03S_a_i_n_tif hard reset doesnt work, try putting it in 'disk mode' and using iTunes to restore it. you WILL lose all your music though.
11:35:23Spacebut not restore yet
11:35:45S_a_i_n_tyou may have to
11:35:59SpaceI will try if I can saves of another form, even if the database
11:36:37Spacemay be can I find it and have the names of songs, and artists? even?
11:37:32S_a_i_n_tiTunes doesn't save the Database as you'd expect it to. it splits the database up into its own filesystem.....if you can, good luck.
11:39:23Spaceitunes is crap, this happened because I took over my computer, and automatically starts automatically connects, :@
11:39:33*amiconn thinks that the 2-instruction steps can be extended to handle 32-bit division without special casing
11:40:10S_a_i_n_tSpace: those features can EASILY be disabled in iTunes.
11:41:37SpaceI haven't idea about the second
11:42:30 Join {phoenix} [0] (
11:42:37S_a_i_n_tIf you don;t want iTunes to automatically connect )to the internet?) remove the firewall exception
11:44:08SpaceI only have firewall on access point
11:49:52Spacedon't exist a option to turn off complete, true? to stop the restart, I need to expect your battery runs down?
11:56:28Spacewell, the last try
12:01:45Unhelpfulamiconn: oh? how so? i tried to think of a way, but it doesn't seem possible to eliminate the possibility that the partial remainder will have the high bit set, but still fail the trial subtraction, and then have its high bit shifted out on the next iteration.
12:02:21pixelmaSpace: if you get your Ipod into disk mode you should be able to copy all files off if the filesystem is not damaged I believe. Then the music won't be lost but it will be hard to find and order the files correctly again (I don't have an Ipod so I'm not completely sure, just from what I heard about Ipods here)
12:05:52Unhelpfuland if the high bit shifts out, the next iteration will produce a correct partial remainder, but will not set the carry bit correctly
12:10:08 Join teru [0] (
12:10:45Unhelpfulin a uniform distribution the numerator will have the high bit set half of the time. this is probably not the case very often in practice. for the APE udiv32_arm i optimized rather heavily in favor of the common case, so that it costs only one skipped branch. the one i'm working on uses a few more ops but handles the 32-bit numerator case without resorting to calling part of itself as a function
12:12:15 Join wilfred_ [0] (
12:15:51 Quit Space (Read error: 110 (Connection timed out))
12:18:13***Saving seen data "./dancer.seen"
12:19:45 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
12:23:57wilfred_I've not had any feedback on my patch at, can anyone look over it for me please?
12:31:36 Part S_a_i_n_t
12:31:41 Join S_a_i_n_t [0] (i=S_a_i_n_@
12:40:00 Quit yosafbridge ("Coyote finally caught me")
12:40:11 Join yosafbridge [0] (
12:42:30 Join Sajber^ [0] (
12:44:21 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
12:47:38 Nick Ypsy is now known as YPSY (
12:47:40 Nick Horschti is now known as Horscht (n=Horscht2@xbmc/user/horscht)
12:48:13 Join einhirn [0] (
12:51:52amiconnUnhelpful: I think it will be possible by negating the dividend instead of the divisor
12:52:25amiconn(and then using cmn for comparison)
12:57:23pixelmawilfred_: nice to see some translation updates, will look over it later today maybe. A first question - we have a real name policy for contributions - could you state your full name otherwise the patch couldn't be accepted. You'll get a place in the credits then too ;)
12:57:58amiconnHmm, cmn won't work
12:59:14 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
13:02:56 Join Spaceghost [0] (i=quassel@unaffiliated/spaceghost)
13:08:22 Join Tomis [0] (n=Tomis@
13:09:36 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
13:10:02wilfred_pixelma: my full name is in the patch, are you referring to the tracker?
13:11:12pixelmaoh, didn't see it. I think that should do
13:21:21CIA-9New commit by teru (r24203): Fix FS #10893.
13:38:24SpaceghostS_a_i_n_t, finally I could restore the most of my music put the ipod on manually disk mode
13:41:32 Join stripwax [0] (
13:49:08S_a_i_n_tSpaceghost: !yay!
13:51:49 Join petur [0] (
13:52:06 Quit liar (Client Quit)
13:53:10 Quit flydutch ("/* empty */")
13:57:45SpaceghostS_a_i_n_t, what is the meaning of that expresion?
13:59:53 Nick Spaceghost is now known as Space (i=quassel@unaffiliated/spaceghost)
14:00:05 Nick Space is now known as Spaceghost (i=quassel@unaffiliated/spaceghost)
14:01:17S_a_i_n_tyay = great! woohoo! congratulations etc.
14:09:26 Join fdinel [0] (
14:12:04 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
14:14:24 Join stripwax__ [0] (
14:16:13 Join bertrik [0] (
14:16:14 Quit wilfred_ ("Konversation terminated!")
14:18:11pamaurygevaerts: could you have a look at the patch I posted on FS #10889 if you have time ?
14:18:15***Saving seen data "./dancer.seen"
14:25:11 Quit Tomis ()
14:28:44gevaertspamaury: comment added
14:30:38pamauryYes for 1), for 3) I thought about it but it's only a matter of translating then. Also I don't know how translation work. For 2), yes perhaps, it could save some bytes. I'll continue the work on this to fix all that
14:31:43gevaertshm, on second thoughts forget about the string duplication. In a non-logf build there is no duplicate in the first place, and if you translate it gets even more tricky
14:32:39 Quit stripwax (Read error: 110 (Connection timed out))
14:33:07 Join evert [0] (
14:33:09pamauryHow translation work: you add a D***(lang_id)...something instead of the string and where do you put the lang id and the message ?
14:33:43*gevaerts doesn't know :\
14:34:01S_a_i_n_tpamaury: wiki?
14:34:36pamauryprobably on the wiki, yes, I'll have a look at it
14:35:34S_a_i_n_tpamaury: I wasn't trying to be a smartass, I was wondering if you'd checked or not, 'cos if it isn't there, it *should* be there.
14:36:12pamauryNo I haven't check ;) But it seems that there is a description of .lang files on the wiki, I'm reading...
14:39:12 Join Tomis [0] (n=Tomis@
14:44:53evertwhat files can i safely delete in the .rockbox so my rockbox will generate a fresh and new database?
14:45:11evertthere are a lot of songs 2 to 10 double in the database for some unknown reason
14:45:40everti guess all the .rockbox/database* files but just asking to be sure :)
14:46:12stripwax__evert - you might want to 'export modifications' (from database menu) first (e.g. song play counts, your ratings, etc) if you use that feature
14:46:20stripwax__but yes all the database* files
14:46:42everti don't really use that feature, having the scrobbling for that :)
14:46:48evertok, thanks
14:47:36S_a_i_n_ttheres a bunch numbered 0 - 9 or something
14:47:50evertalready gone :)
14:48:00S_a_i_n_tthose are the db files
14:55:25 Quit S_a_i_n_t ("Leaving")
14:55:39 Join S_a_i_n_t [0] (i=S_a_i_n_@
15:00:03 Part S_a_i_n_t
15:00:32 Join S_a_i_n_t [0] (i=S_a_i_n_@
15:01:07 Quit pamaury (Read error: 113 (No route to host))
15:01:38 Join bluebrot1er [0] (
15:09:12 Nick bluebrot1er is now known as bluebroth3r (
15:09:23 Quit bluebrother (Nick collision from services.)
15:09:29 Nick bluebroth3r is now known as bluebrother (n=dom@rockbox/developer/bluebrother)
15:10:34 Join mt [0] (n=mtee@rockbox/developer/mt)
15:12:19Unhelpfulamiconn: also -(1<<32) is still (1<<32) isn't it? but that leaves one very small special case.
15:12:39Unhelpfulalso remember there are no compares in the loop :)
15:13:02mtWould there be a problem if the front page showed the latest commits in the mdctexp branch ?
15:13:31mtB4gder: ^ ?
15:15:17Unhelpfulamiconn: i'm heading for bead, but if you want to look at the N-R multiplier and the analysis, it's in the ARM System Developer's Guide, which you can browse here:
15:15:46Unhelpfulheading 7.3.2, page 223
15:17:18Unhelpfula lengthy analysis of the path taken through the routine depending on divisor size is also given.
15:21:06B4gdermt: well, it'd require modified scripting...
15:21:18B4gdernot necessarily a problem, just work
15:21:22 Quit stripwax__ (Read error: 104 (Connection reset by peer))
15:22:12 Join liar [0] (n=liar@
15:22:32 Nick SIGSEGV is now known as pjm0616 (n=user@
15:23:12mtB4gder: I'm just looking in /www/ , I could take a shot at a it myself, I was just asking to see if there were any objections to the idea itself. (btw, would that be the only file to edit ?)
15:24:43 Join Oxymis [0] (
15:25:20B4gderafair, it works against a svn checkout so it needs that at least
15:26:43 Join stripwax [0] (
15:27:07 Quit stripwax (Client Quit)
15:27:36 Join stripwax [0] (
15:28:15 Join stooo [0] (
15:28:15mtstripwax: Great work on the mdct, Thanks :)
15:28:57stripwaxmt :) no problem. it's not there yet though..
15:30:07stripwaxin particular, I *really* want to investigate alternative pre/post twiddling steps, to elimate requiring different twiddle tables for different sizes (and/or unifying all tables into one/two tables, like Tremor has)
15:30:54stripwaxThat would also eliminate one of the remaining two "Init" steps (the mdct twiddle factors is one init step, the fft bitreverse/reordering index table is the other init step). Fft no longer has any twiddle factor init, cos that uses the same table as Tremor now
15:31:02stripwax(the same static/fixed table)
15:33:26mtI see. I really want to work on it, but it seems my college is trying to set a new record on how many hours/day they keep us there. (last sat ~13 hours :/ )
15:33:51stripwaxPah, that's nothing . This time last year I was working 42hour days at work! :)
15:34:02B4gderon the hiway!
15:34:14B4gderin a shoebox
15:34:25*B4gder sneaks off into -community again
15:34:25mtAh you win. :)
15:34:55stripwaxOne twiddle idea comes from a pdf I read, I'll dig out the link later on. Basically you can replace the current cos(2PI * (i+1/8)/N) pre twiddle with just cos(2PI*i/N), and replace the cos(2PI*(i+1/8)/N) post twiddle with cos(2PI*(i+1/4)/N)
15:35:08stripwaxNow that would mean the pretwiddle can share the same table as fft/tremor
15:35:42stripwaxthe post twiddle is still a different set of trigs though :( But you could actually reduce the size of the table at the expense of one extra complex mul per sample:
15:36:02mtNice. Could you add that link to the wiki ?
15:36:18stripwaxI.e. post twiddle by cos(2PI*i/N), then post twiddle again by cos(2PI/4N)
15:36:47stripwax(which is a constant dependent only on N, i.e. a single cos/sin factor required for each different N, as opposed to a whole new trig table for different sizes of N)
15:36:53stripwaxmt - yep, will do
15:37:48mtGreat. I must run now. btw, excellent code comments. ;)
15:37:53stripwaxhah, thanks! :)
15:38:08*mt heads off
15:39:00*stripwax wonders which will be better : trig table in ICONST data and four extra muls per sample, or trig table in ram and no extra muls per sample
15:41:02CIA-9New commit by teru (r24204): draw selection bar even if str is empty string (e.g. in text_editor plugin).
15:43:48 Join saratoga_ [0] (i=463f90ed@gateway/web/freenode/x-bfbmserowycpnrmr)
15:44:11bluebrotherdomonoky: I was thinking about adding a new parameter with the rbutil version number when retrieving rbutilqt.php from the theme site. That way the theme site can react to the versions, like displaying a notice that a newer rbutil is required for the theme site.
15:44:16bluebrotherany thoughts on that?
15:46:38domonokyhm, does the theme site then need to know about rbutil versions ? wouldnt it be better the other way round ? ie a version in the themeoutput of rbutil.php ?
15:48:13 Quit stripwax (Read error: 104 (Connection reset by peer))
15:53:16bluebrotherwell, that would mean that rbutil needs to know what to display −− and it doesn't know about this.
15:53:51 Join stripwax [0] (
15:53:54bluebrotherbut we could add both directions :)
15:55:01bluebrothermy idea was to maybe try to submit the svn revision of rbutil, so the theme site can say something like "rbutil svn > r12345 (version 10.20.30) required, your rbutil is r9999"
15:55:28bluebrotheri.e. it doesn't need to know exactly about rbutil but only which is the minimal version required to use the theme site.
15:55:40domonokybut how does the theme site know this ? hardcoded in the php file ?
15:56:08bluebrotherof course we could add more functionality around that later. No idea what useful applications are around, but well ;-)
15:56:45bluebrotherwell, the theme site obviously needs to know about the versions. Maybe a database table with a revision - error string mapping.
15:57:03bluebrotherthe php file needs to access the database anyway ;-)
15:57:24domonokysure it has access to the db, but we dont have this table :-)
15:57:34domonokyand someone has to maintain the table.
15:58:10bluebrothersure, but I guess this shouldn't change often.
15:58:36domonokyi think it would be better it rbutil just requests version x of the rbutil.php output, if it cant deliver that, it gives a error. And we just bump the version, when ever we change the rbutil.php output.
15:59:15bluebrotherwell, it's not only the output format of rbutil.php. It's also the parameters −− like changing the target names.
15:59:46 Quit saratoga_ ("Page closed")
15:59:53bluebrotherright now svn will request a target sansae200 while the last release will request e200. That was a source for that idea ;-)
15:59:58 Quit stripwax (Read error: 54 (Connection reset by peer))
16:00:39domonokyfor the targetnames, we should just rename the themesite targetnames, shortly before/after we release a new rbutil version.
16:00:54bluebrotherso instead of retrieving rbutil.php?target=e200 we could now retrieve rbutil.php?target=sansae200&rbutil=12345. Then the theme site can return an error if that revision number is too low
16:01:07*flyback falls over head
16:01:28bluebrotherand as old rbutil version don't send that number the theme site can simply tell that the version is too old too.
16:03:40bluebrotherthat's at least better than simply to tell that there are no themes (because the target name changed). The rename is still an issue with the build server binaries but that's a different thing.
16:04:19domonokysure, a better error-message would be nice. currently it could also tell you, that it doesnt know this theme :-)
16:04:54 Join stripwax [0] (
16:05:26 Quit stripwax (Client Quit)
16:06:08domonokybut i think we should bind this to revision numbers, a simple version number would also work. like rbutil.php?target=sansae200&version=1
16:07:11bluebrotherok, so we agree that it makes sense to send some rbutil version string / number to the theme site at least? ;-)
16:08:08bluebrotherok. I'll rethink it a bit about what exactly to send. Sending a GET parameter more fortunately doesn't hurt so it's no problem :)
16:10:58 Join stripwax [0] (
16:12:21domonokygood. the less changes this need on the themesite the better :-)
16:13:10 Join pamaury [0] (
16:13:56 Quit teru ("Quit")
16:18:19***Saving seen data "./dancer.seen"
16:26:49 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
16:29:44 Join mc2739 [0] (n=mc2739@rockbox/developer/mc2739)
16:33:54 Quit stoffel (Read error: 113 (No route to host))
16:34:49 Quit killan ("( :: NoNameScript 4.22 :: )")
16:35:03 Join killan [0] (
16:41:20 Join stripwax [0] (
16:44:11bertrikis there something special in tick_funcs[0] ?
16:45:22bertriknever mind, the code looks funny, but is probably correct
16:49:39 Join efyx_ [0] (
17:00:05 Quit Oxymis (Read error: 110 (Connection timed out))
17:01:24 Join kramer3d [0] (n=kramer@unaffiliated/kramer3d)
17:01:43 Quit kramer3d (Read error: 104 (Connection reset by peer))
17:14:39 Join panni_ [0] (
17:15:15bertrikThe as3525 has two PLLs of which we currently use only one (PLL A). This results in a lot of clock frequencies being dependent on each other.
17:15:35bertrikMaybe we could get a bit more accurate by using PLL B just for generating the audio clocks
17:17:13bertrikbut that could have the disadvantage of drawing a little more current
17:18:52LloreanHow much is it likely to impact battery life?
17:20:18LloreanI mean, in my book "introducing as little error into the audio as possible" is a pretty high priority as long as we're still able to compare equally or favorably with the OF.
17:20:27LloreanThat is, once power management over all is done.
17:20:48bertrikI don't know by how much. I vaguely remember someone having done measurements on this (saratoga/buschel maybe?) but I'm not sure
17:21:49soapI assume nothing which could be runtime selectable? The option of "precise timing with less battery" vs "OF timing and OF battery life" ?
17:22:09LloreanDo we even match the OF battery life on Clip yet?
17:22:18LloreanI haven't paid attention but I thought last time I looked it was running short still.
17:22:28bertrikI don't think so yet, I'll have a look at SansaRuntime
17:22:49soapWhile it's a nice rule-of-thumb to try and compare favorably with the OF when it comes to battery life, if the OF "cheats" and degrades battery life at the cost of audio playback it's not really a fair comparison anymore.
17:23:03bertrikhm, there's no Clip runtime yet
17:23:13soapexcuse me. s/degrades/improves/
17:24:31Lloreansoap: Well, I mean if it's a significant (I mean 10% or more kind of numbers) reduction in battery life for something a majority of users probably can't hear, and it can't be user configurable, and it'll mean we'll probably always measure less than the OF in battery life we probably shouldn't do it.
17:24:42LloreanI mean optimizations for all of our compressed audio are accuracy vs performance.
17:25:04LloreanBut it really depends on how badly it affects things (which we won't know without some sort of testing anyway)
17:25:44bertrikI guess significantly less than 10% impact
17:26:13LloreanThen we should probably just go for it if it means fixing a notorious problem the OF has.
17:27:04 Quit seani__ (Read error: 110 (Connection timed out))
17:27:08 Join seani__ [0] (n=seani@
17:28:22 Quit killan ("( :: NoNameScript 4.22 :: )")
17:28:29bertrikI'm not sure we can do it with 0% error anyway
17:28:42 Join killan [0] (
17:28:44 Join Grahack [0] (
17:28:51LloreanDo we know the error on our other targets?
17:29:19bertrikAt least I don't know it, or know where it is documented .. :P
17:31:47bertrikWe can configure PLL B to 16 x Fcrystal at 384 MHz, this would give 44117.6 Hz, or a 0.04% error
17:32:55 Join Oxymis [0] (
17:33:37 Quit Oxymis (Read error: 54 (Connection reset by peer))
17:34:31 Join Oxymis [0] (
17:35:03 Quit Oxymis (Read error: 104 (Connection reset by peer))
17:36:04 Join Oxymis [0] (
17:36:04bertrikI guess this takes 2 hours (by me or by funman for example) or so to implement as a proof-of-concept
17:40:22 Quit stooo (Read error: 104 (Connection reset by peer))
17:40:24 Join stooo1 [0] (
17:41:01FlynDicebertrik: We could just change the whole clock scheme to run off of plla @ 384, the main thing we lose is the higher fclk @ 248, we could use fclk 192, pclk 64
17:42:21bertrikFlynDice, ah, that's just changing a #define in clock-target.h, right?
17:42:23 Join mc2739_ [0] (n=mc2739@rockbox/developer/mc2739)
17:42:55bertrikFlynDice, do you remember anyone doing current measurements while changing the various clocks?
17:43:09 Quit mc2739 (Nick collision from services.)
17:43:09 Nick mc2739_ is now known as mc2739 (n=mc2739@rockbox/developer/mc2739)
17:43:09FlynDiceyes , a couple of defines
17:44:09FlynDiceno current measurements, saratoga mentioned something along those lines but then we kind of settled in to the current scheme as being the one to go with.
17:45:55FlynDiceI can say mpegplayer runs fine at 192 fclk, is there something else that we do that requires more horsepower?
17:46:26bertrikI think one of the lossless codecs needs a lot
17:47:19 Join robin0800 [0] (
17:48:42 Join stoffel [0] (n=quassel@
17:48:43 Join shonn [0] (
17:49:27bertrikthe ape codec is very CPU intensive
17:50:13 Quit shonn (Client Quit)
17:53:11FlynDiceso the tradeoff there is better ape performance for reduced runtime(using pllb for audio)?
17:53:49bertrikI think the options are: 1) keep the current scheme with about 0.15% pitch error (still better than OF) 2) use a PLLA frequency of 384 MHz (0.04% error@44.1 kHz), but sacrifice some boosted CPU speed (248MHz -> 192 MHz) and some battery life or 3) use a separate PLL (PLL B) at 384 MHz (0.04% error @ 44.1 kHz) for the audio clock and sacrifice some battery life
17:54:27FlynDiceyes that seems correct to me
17:54:41bertrikmaybe option 2) is even a bit better w.r.t. battery life because of the reduced boosted CPU speed
17:55:47 Join kugel [0] (n=kugel@rockbox/developer/kugel)
17:55:51soapis there an option which splits the difference?
17:55:56FlynDiceyes, that would mean no switching for voltage scaling, just a constant lower voltage should work
17:56:11*kugel thinks we should keep what we have now
17:56:34*FlynDice knew that's what kugel would say ;-)
17:56:40soap0.04% is below what Wikipedia is calling the threshold of perception. Could a bit more error be allowed to safe any battery life - or are there really only two good clocks to be at?
17:57:07kugelI doubt anyone notices 0.15%, the OF's error is >2% AFAIK
17:58:29kugelhm, maybe the >2% was on the v2 AMSes then (fuzev2, clipv2)
18:01:13 Join DerPapst [0] (n=DerPapst@
18:01:40 Join saratoga_ [0] (i=9803c57f@gateway/web/freenode/x-juveldxmhzzwmbid)
18:01:45 Quit stooo1 (Read error: 104 (Connection reset by peer))
18:01:54saratoga_FWIW I tend to think > 200MHz clocks on AMS aren't really worth it since they require higher voltage
18:01:55kugelthe samsas are known for great sound quality, even with the .247%, so I don't think .15% is enough to sacrifice battery like or performance
18:01:55bertriksoap, the wikipedia page I'm reading mentions 0.36 Hz in the octave of 1000-2000 Hz, that's 0.036% at 1000 Hz and quite close to the 0.04% we can reach
18:01:59 Quit seani__ ("Leaving")
18:01:59 Join stooo [0] (
18:02:03saratoga_we'd probably get better battery life without them
18:02:12kugelwe need the voltage anyway
18:02:27kugelmany microsd cards don't work with lower voltage
18:02:49saratoga_is that actually related to the cpu core voltage or just because of how we switch it?
18:03:07FlynDiceI would tend to think its the switching
18:03:10bertrikI still don't really get the low-voltage / sd-card problem by the way, AFAIK sd cards work at 3.3V, not at the internal processor voltage
18:03:36kugelbefore we boosted for microsd access many more cards didn't work
18:03:54saratoga_i always assumed it was the SD controller on the AS chip that had either a problem with lower voltage or a problem with switching voltages (perhaps it ends up in some undefined state occasionally on voltage change)
18:04:53FlynDicewe're using 3.0 v for the sdcard voltage now which matches the of
18:05:06saratoga_i wonder if its worth getting an m200v4 and hooking up jtag to it to see what clocks the OF is using
18:05:11saratoga_does anyone have one of those?
18:05:23saratoga_iirc it had widely spaced jtag pins that could be readily used
18:05:24bertrikdomonoky has one
18:05:55bertrik(an m200v4 I mean, not sure about any jtag adapter)
18:06:09saratoga_serial to jtag adapters are a couple dollars IIRC
18:06:24saratoga_though maybe just reverse engineering the OF more is easier
18:06:35*domonoky could send the m200v4 to someone todo this...
18:07:10kugelsaratoga: the current problem is the switching probably, but number of non working cards was way higher when we accessed them while being at lower voltage (that wasn't a switching problem)
18:07:40saratoga_i still want to know why we can't clock the AMS memory controller at 2x PCLK
18:07:50saratoga_the controller and the DRAM both say they can do it
18:09:17 Quit Oxymis (Remote closed the connection)
18:12:04kugeldomonoky: I see the admin work needed text, but I can't find the actual job
18:13:00kugel(i logged in)
18:14:09 Quit saratoga_ ("Page closed")
18:15:27domonokykugel: use the link to view all themes, and then limit selection to reported themes.
18:15:55kugelthere are a few errors at the top
18:16:23domonokycan you show them to me (screen shot) i dont see errors.
18:16:38domonokyoh, i see them now too.
18:17:09domonokyah, it look at the wrong place for the file.
18:18:04kugelpeople seem to abuse the report function
18:18:22***Saving seen data "./dancer.seen"
18:21:22kugelare we going to delete themes that are marked with "Theme was replaced by newer version."
18:21:54*domonoky doesnt understand why it looks at the wrong place for filestat. Maybe someone else can take a look to spot the error ? (private/themesite.class.php listthemes(), approved should be 2 if a theme is reported.)
18:22:05 Quit stooo ("Leaving.")
18:22:50domonokykugel: that are themes which are replaced with the update function from the users.. i didnt want to delete them, so we can revert if something goes wrong.
18:25:17 Quit stoffel (Read error: 60 (Operation timed out))
18:29:02*domonoky sees the error... the are moved the changestatus().. *goes fixing* :-/
18:30:27CIA-9New commit by Domonoky (r24205): make sure reported themes are not moved in the private dir.
18:32:08bertrikI somehow remember the build system only starting a build round for changes in the actual rockbox firmware (not for the manual and not for rbutil changes), but I see my client is building nevertheless
18:40:03kugelit builds on any commit since the remake iiuc
18:41:26domonokykugel: the report about the "Too Much Information-Terminal" theme is funny. the theme actually uses %cw instead of %cd so it has very short months :-)
18:42:56CIA-9New commit by bertrik (r24206): Make a few global variables static instead of global where possible
18:43:32kugelthat theme clearly catched the unique ubuntu terminal look :D
18:43:55 Quit Grahack ("Tu m'as vu ?")
18:50:56 Join Oxymis [0] (
18:54:49 Join einhirn [0] (n=Miranda@
18:59:32 Join Tomis2 [0] (n=Tomis@
19:11:37 Quit Tomis (Read error: 110 (Connection timed out))
19:11:38 Nick Tomis2 is now known as Tomis (n=Tomis@
19:17:00pamaurywhat is the "user:" field in lang files ? It's not described in the wiki.
19:21:45 Quit Oxymis (Read error: 110 (Connection timed out))
19:21:49kugelpamaury: it's planned for plugin localization
19:22:16kugelso, there could be strings with "user: bubbles" or the like in the future
19:25:44 Quit robin0800 (Remote closed the connection)
19:27:21 Join hebz0rl [0] (
19:32:33 Quit alexbobp (Read error: 104 (Connection reset by peer))
19:32:38 Join alexbobp [0] (n=alex@
19:41:18 Join moronic44 [0] (
19:41:41moronic44hi! anybody here..?
19:42:27AlexPIf you have a question, just ask
19:42:37 Quit einhirn (Read error: 104 (Connection reset by peer))
19:43:25moronic44alright.. im a novice to ubuntu..and rockbox, and i was thinking of investing in an ipod and wondering which one works most superbly with rockbox? my first choice was an ipod touch/classic but it appears neither of those models work with rockbox or anything on linux really...
19:44:07AlexPlinux/windows/whatever is irrelevent to Rockbox
19:44:39AlexPAnd if you really want an ipod, then any from 1st - 5.5 gen (video) are fine, or 1st (and semi- 2nd) gen nano
19:45:08moronic44alrighty...thank you sir
19:45:12 Part moronic44
19:45:15AlexPBut which player is best depends on your needs - storage/radio/recording/battery life/etc. etc. may help somewhat, but I've a feeling it is somewhat outdated
19:46:02AlexPoh, he left
19:48:11 Quit amiconn (Nick collision from services.)
19:48:13 Join amiconn_ [0] (i=quassel@rockbox/developer/amiconn)
19:48:30 Quit pixelma (Nick collision from services.)
19:48:32 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
19:48:33 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
19:48:49 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
19:58:11 Join flydutch [0] (
20:00:20 Join Strife89 [0] (
20:01:44 Join toffe82 [0] (
20:04:54CIA-9New commit by torne (r24207): FS #10107: work around the issue on some ipods where after shutdown they do not power on. ...
20:07:06 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
20:13:18kugelFlynDice: maybe we could add a micro delay after polling the core voltage before switching?
20:14:17pamaurygevaerts: new version of the patch for FS #10889, it includes translation, I hope I did it correctly (it seem to works for me). Also remove LOGF_ENABLE
20:14:58kugeldo we need to translate debug-only stuff?
20:15:11pamaurythis is a gui message
20:15:15pamauryso yes, I think
20:15:54pamauryBut this can be discussed. It's just that it would be strange to have it in english only.
20:16:04kugelit seems a bit awkward to ask translators to translate a string that you should ideally never see and that is eventually disabled in normal builds
20:16:55 Quit bmbl (Read error: 60 (Operation timed out))
20:18:25***Saving seen data "./dancer.seen"
20:18:50kugelisn't the debug menu also english only?
20:19:08pamauryIndeed but because then then is no problem in allowing it to be translated, no ? Don't know for the debug menu
20:19:24pamaury*-because -then +it's
20:22:44 Quit alexbobp (Read error: 54 (Connection reset by peer))
20:24:36pamauryAnyway, it's not definitive, we can change this, if you strongly oppose translation then insult me on FS #10889 ;)
20:24:45kugelpamaury: the problem is also that it increases language sizes even if disabled
20:25:49bertrikI noticed that for archos recorder a file is created (instead of a file)
20:26:06bertrikThis is not intentional, is it?
20:27:05pamaurykugel: really ?
20:27:33pamauryThat logical in some way, much simpler this way
20:27:58pamauryIsn't there a #ifdef-like mechanism for lang files ?
20:28:05pixelmabertrik: where and what .link file?
20:28:35gevaertspamaury: there is, but it's a bit problematic. It changes the string ids, which makes both lang files and voice files incompatible
20:29:06bertrikpixelma, in apps/plugins/pictureflow/pictureflow.make, line 22
20:29:48pamauryAh, indeed. And isn't it possible to set the string to none ? Is the id skipped in that case ?
20:31:30bertrikHow about just using the "plugin returned error" splash? That one is already present so we don't need to add a new translation. Possibly we could show a number to differentiate between various plugin errors.
20:32:51pamauryThat's a bit weird but why not
20:33:28gevaertsseems reasonable
20:33:50kugelthat sounds ok indeed
20:34:31 Join munk [0] (
20:36:00pamaurySo display somethong like "Plugin returned error xx". I'll come back in a minute with a new patch ;)
20:36:41 Quit hebz0rl ("Ex-Chat")
20:39:32 Quit munk ("CGI:IRC (Ping timeout)")
20:39:53pamaurywhat is the difference between ID2P() and str() ?
20:40:30bertrikpixelma, oh, as far as I understand, this is just a intermediate file, the name does not really matter and it does not end up in the install zip
20:41:42pamaurydo we display a number or a little text ? (#00 for example vs leak-file-handles ?).
20:42:54kugelpamaury: I think ID2P is for voicing
20:42:59kugelbut I'm not sure
20:42:59 Join munk [0] (
20:43:35munkHello. I would like to ask if there is any version of rockbox for Sansa Fuze v2? Are there any plans for supportins this device?
20:43:52 Quit kaniini ("BRB! IRL!")
20:44:08*flyback goes out for a redbull to clear his headache, bbl
20:45:02pamauryBut ID2P is use also for strings in the code
20:47:02bertrikmunk, no there is not a rockbox version for the fuze v2, and no plans that I'm aware of
20:48:09pamaurygevaerts: new version(5) of patch in FS ! I think well will succeed in commiting it before version 99 ;)
20:49:22munkthanks for the answer, bertrik
20:52:46gevaertspamaury: looks good. There are two things that I might do differently, but I think they're more a matter of taste: I'd still go for a normal splash, and I'd put the MAX_OPEN_FILES test in config.h
20:54:11pamauryfor MAX_OPEN_FILES I disagree because it's clearly implementation specific and thus has nothing to do with config.h. In could even depend on other things. You mean normal splash instead of yesno ?
20:54:52pixelmabertrik: I see. It still doesn't sound like intended but I don't know
20:55:32*gevaerts said it was a matter of taste :)
20:56:04*pamaury don't like splash :) It not only my taste. If everyone prefer splash, I'll go for splash.
20:56:35pamaury(but if nobody complain I will go for yesno niark niark)
20:58:59pixelmaspeaking about pictureflow and "plugin returned error" reminds me of that weird bug in pictureflow I experienced and don't know if it is fixed yet. If there are no album art pictures at all, pictureflow exited after trying to build the album art cache with an error. I found that weird because you usually get a default picture showing a question mark for missing art - why should that not work if those would have to be used for all albums?
21:03:04 Join p3tur [50] (n=petur@rockbox/developer/petur)
21:06:23 Quit munk ("CGI:IRC (EOF)")
21:08:48 Nick krazykit` is now known as krazykit (n=kkit@
21:30:48soapTorne, "fixed" or "workaround commited" re:10107?
21:30:59soapDid I miss a conversation on this?
21:32:04 Join webgue [0] (
21:33:58 Quit kugel (Read error: 104 (Connection reset by peer))
21:34:08 Quit webgue (Client Quit)
21:55:08 Quit stripwax (Read error: 104 (Connection reset by peer))
21:58:26Unhelpfulamiconn: i'm not sure you'd need to invert the division loop "test" at all if you invert the numerator instead of the divisor... but i'm also not certain that it solves the problem of being able to shift out bits that are needed. i already have a framework for emulating adds with carry etc in C, i'll give it a try when i get a chance.
22:10:29 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
22:16:52 Part froggyman
22:17:06 Join gill0r [0] (
22:18:26***Saving seen data "./dancer.seen"
22:38:43B4gderRockbox talk at Fosdem 2010 more or less decided now
22:44:29AlexPIn the linux section? :P
22:45:32B4gderthere's an "embedded dev room" for which I'm aiming
22:45:44AlexPSeems more reasonable :)
22:46:16B4gderI think its already more or less accepted
22:46:35B4gder"already" being fosdem time schedule, I would otherwise say its rather late
22:46:45AlexPyeah, it is in Feb right?
22:46:51B4gderyes, feb 6-7
22:49:26 Join stoffel [0] (
22:50:25 Quit stoffel (Remote closed the connection)
22:51:23bertrikI should go to FOSDEM
22:51:49B4gderthere will be rockbox silliness!
22:52:26 Join shaika-dzari [0] (
22:56:00shaika-dzarihello everyone
22:56:31shaika-dzariCan I ask a question related to rockbox utility and sansa e200v2 ?
22:56:33pixelmaB4gder: while you are here, could you check why the 5/6GB H10 links on the daily manuals page link to the 20GB H10 manual? As far as I remember those are separate manuals due to the different screen resolution, haven't checked if the former still builds though yet
22:56:43 Quit {phoenix} (Remote closed the connection)
22:58:23 Quit faemir (Read error: 104 (Connection reset by peer))
23:00:13 Quit bmbl ("Bye!")
23:00:45 Join rhodan [0] (
23:00:55pixelmashaika-dzari: don't ask to ask...
23:01:37shaika-dzariok, I want to install rockbox on my sansa e200v2 with rockbox utility 1.2.3
23:02:01AlexPI don't think rbutil release versions supports the v2
23:02:47AlexPalthough I could well be wrong :)
23:02:49shaika-dzariWith 1.2.3 I can select sansa e200v2 series (Unstable)
23:02:54AlexPah, OK
23:03:10AlexPThen I was wrong :)
23:03:11shaika-dzariThis is the first time I try to install rockbox
23:03:57shaika-dzariSo, if I understand correctly, I should install the bootloader and rockbox in installation Tab.
23:05:12pixelmaB4gder: the small H10 manual still builds locally
23:05:17shaika-dzari>Ok, I select default bootlaoder and current build of rockbox and it should be fine?
23:05:37shaika-dzariOk, I try right now
23:06:17Unhelpfulamiconn: i've gotten another cycle off of the large-numerator handling. i think that what i'm doing now is likely as cheap as inverting the numerator, if that would work... it costs only 4c from the small-numerator case, one to unconditionally save the old numerator (with flag set), one to shift the numerator right if that save has a negative result, one for an asr one on the stored numerator, and one for a bmi to the fixup code. the same
23:06:17Unhelpful asr 1 also puts the bit that was lost into C, and the fixup can avoid other flag-setting ops until it needs that bit.
23:06:48shaika-dzarihe, 300 mo free space should be enought for rockbox?
23:07:28AlexPRockbox is about 10/15 MB if you include all the fonts etc., less without
23:07:40Unhelpfulhere's the code as it is now:
23:07:45AlexPIncidently, install the fonts too :)
23:12:27shaika-dzaricool, installation done.
23:12:52shaika-dzariI'll umount my player and try to boot.
23:12:58B4gderpixelma: I don't even understand how the builds are done these days... :-O
23:14:40Unhelpfulit probably still loses on beast. the shifted operand in the divide loop causes a 1c interlock, and i'm not certain but it may do so even if the subcc is skipped... so we're as slow as the libgcc divider inside the main loop, and have extra fixup.
23:16:33shaika-dzariwow, work great! Thanks AlexP!
23:16:43AlexPshaika-dzari: No worries, enjoy :)
23:18:23 Nick fxb__ is now known as fxb (
23:19:17pixelmaB4gder: a pity. I thought it was some symlink problem on the server or so
23:19:58pixelmamaybe something to do with the target rename
23:23:28B4gderoh I think I found it
23:25:33 Join mc2739_ [0] (n=mc2739@rockbox/developer/mc2739)
23:28:02 Quit p3tur (Remote closed the connection)
23:28:12CIA-9New commit by mc2739 (r24208): iriver H10 5GB has it's own manual
23:29:14B4gderhaha, 10 seconds before I tried to commit that fix ;-)
23:29:46mc2739_too slow :)
23:29:55*B4gder hangs head
23:29:59 Quit mc2739 (Nick collision from services.)
23:30:00 Nick mc2739_ is now known as mc2739 (n=mc2739@rockbox/developer/mc2739)
23:31:16 Quit krazykit (Read error: 110 (Connection timed out))
23:34:50pamauryI have a question: I've just built a e200 simulator and I found that the screen lacks brightness, nearly difficult to read. Am I the only one to complain about that ?
23:35:02AlexPpamaury: yep :)
23:36:13pamaury:( That's really strange
23:36:24AlexPpamaury: I'm just building a current one to see
23:36:33pixelmaB4gder, mc2739: thanks to both of you ;) :)
23:39:21 Quit efyx_ (Remote closed the connection)
23:39:43amiconnWeird poweroff method...
23:40:01AlexPpamaury: Seems OK here
23:41:21pamauryPerhaps I'm just used to backlight on my device and it seems weird to :)
23:41:37 Join bluebrot1er [0] (
23:42:31AlexPpamaury: It isn't blindingly bright, but not dim either
23:42:50 Join faemir [0] (n=faemir@
23:43:15pamauryYes it's between. I prefer my real device for sure ;)
23:43:29 Quit bluebrother (Read error: 60 (Operation timed out))
23:45:11mc2739pamaury, AlexP: It is dimmer than it used to be
23:46:25 Quit petur ("Zzzzz")
23:46:27AlexPWhy do all the lang files have the same line "# $Id: portugues.lang 23888 2009-12-07 13:53:55Z zagor $" with everything identical except the language name?
23:52:52bluebrot1erperhaps all of them were changed in r23888?
23:53:25bluebrot1eror someone copied the string and adjusted the filename and svn:keywords isn't set.
23:55:00AlexPbluebrot1er: I'm assuming the latter, as lang files are updated periodically
23:55:18AlexPSo for a new one, I'll set keywords and delete that line
23:55:20 Nick bluebrot1er is now known as bluebrother (
23:55:24AlexPI guess :)
23:55:34*bluebrother is about to check it
23:55:40AlexPcheers :)
23:55:59bluebrotherall lang files show r24208 as last changed revision here.
23:56:37AlexPWhich is current svn
23:57:00AlexPWhich didn't touch lang files
23:57:08bluebrotherthe last change is "Corrected ondio language wildcards." Seems like a good commit to touch all lang files
23:57:31AlexPah, good point
23:57:53AlexPI'm looking at the slovak.lang on flyspray
23:58:11AlexPWhich looks to have been based on a previous one
23:58:29AlexPSo for that, I delete that line from the file and set keywords
23:58:33bluebrotherhmm, svn info returns the latest revision, not the last revision that touched the file. Good to know.
23:58:36AlexPbluebrother: What rev was that?
23:58:58AlexPSo that makes sense then :)

Previous day | Next day