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 2010-03-24

00:00:02 Quit domonoky (Read error: Connection reset by peer)
00:00:33 Part froggyman
00:02:16 Join blairb [0] (
00:03:43 Quit kolka (Quit: CGI:IRC (Ping timeout))
00:03:54RoronoaIn the Project 'New WMA audio codecs' on the Rockbox site, i did not exactly get what do is meant by floating point code.(Explain?)
00:04:13 Join Casainho [0] (
00:04:35 Quit jgarvey (Quit: Leaving)
00:04:44AlexPRoronoa: Code that uses non-integer numbers
00:05:17 Quit rvvs89 (Ping timeout: 246 seconds)
00:05:30gevaertsOr rather, code that uses floating point numbers :) See
00:05:44Roronoait might be because i have not gone through the code of 2008 GSoC
00:05:59AlexPgevaerts: Yes, I was trying to avoid explaining it by saying the same thing again :)
00:06:20gevaertsAlexP: well yes, but fixed point numbers are also not necessarily integer numbers
00:06:30AlexPRoronoa: floating point/integer is a pretty basic thing common to pretty much everything
00:06:36AlexPgevaerts: true
00:06:48RoronoaI understand that much...I have experience in both C and Assembly
00:07:13 Quit kolkakolka (Quit: CGI:IRC (EOF))
00:07:19 Join rvvs89 [0] (
00:07:26Roronoabut just i thought that it might be some thing in Audio Codec Theory
00:07:45gevaertsThe point is that most rockbox targets have no hardware support for floating point operations (FPU), so fixed point is much more efficient
00:08:05gevaertsno, it's the same floating point as everywhere else
00:08:19 Quit Schmogel (Quit: Miranda IM! Smaller, Faster, Easier.
00:08:59Roronoaso, one of the parts of the project is to remove the use of floating point numbers
00:09:35gevaertsas I understand it, yes. (I'm not a codec expert...)
00:09:43AlexPAnd yes, me saying integer was also bad (lazy :))
00:10:34 Join Tomis [0] (
00:10:42Roronoai need some link from where i can get information on 'Audio Codec Theory' as i know almost every thing else specified
00:10:43soaptorne - I thought the "fear" over the remaining patch in FS #9708 was that unboosted buffering might cause system lag. I know Soap_Hotel misspoke when he ;) called it "DMA changes".
00:10:55bieberSo on targets without an FPU, is the compiler doing floating point ops by generating normal arithmetic code to do it?
00:11:07 Quit pamaury (Quit: Quitte)
00:16:51gevaertsb0hoon: if the code uses floating point, yes. It's slow though, and therefore to be avoided, especially in codecs. (Last answer of the day. It's bedtime)
00:17:00gevaertshm, wrong person
00:17:03gevaertsbieber: ^^
00:17:42 Join m3dlg [0] (
00:18:32*linuxstb wonders about r25310 and FS #11134 - wasn't that just the usual issue of a checkout using something creating DOS line endings, and a cygwin setup in Unix line-ending mode?
00:19:37 Quit naag (Quit: Leaving.)
00:20:02linuxstb(in which case, wouldnt' the commit break cygwin configured to use DOS line-endings?)
00:20:23 Join naag [0] (~harish@
00:20:52Tornecygwin configured to use dos line endings is broken in and of itself
00:21:09Tornea large number of the cygwin tools do not work properly like that :)
00:21:18Tornewhich is why it's been recommended against by cygwin for a very long time
00:22:24linuxstbYes, I agree. But that commit just forced one file to have LF line-endings - I would expect that same user to have many other problems compiling Rockbox...
00:24:16 Part naag
00:24:52 Join naag [0] (~harish@
00:27:30 Join Casainho_ [0] (~chatzilla@
00:29:06 Quit Casainho (Ping timeout: 245 seconds)
00:29:09 Nick Casainho_ is now known as Casainho (~chatzilla@
00:29:40 Join mt_ [0] (~mtee@
00:29:59 Part naag
00:30:32 Join naag [0] (~harish@
00:30:55 Part naag
00:31:02 Join stripwax [0] (
00:31:09 Quit mt (Ping timeout: 246 seconds)
00:31:55 Quit anewuser ()
00:32:11 Quit Adubb (Ping timeout: 258 seconds)
00:32:14 Quit TopyMobile_ (Ping timeout: 276 seconds)
00:32:57 Quit einhirn (Read error: Connection reset by peer)
00:34:19 Quit Luca_S (Quit: CGI:IRC)
00:36:10stripwaxHas anyone else observed buffering problems recently (e.g. permanent CPU boost and buffer containing only the current playing track and no buffer refilling takes place until that current playing track ends)?
00:38:09 Join Adubb [0] (~aldubuc@
00:38:57 Join RadicalR [0] (
00:39:41 Quit DataGhost (Ping timeout: 258 seconds)
00:42:04 Quit ender` (Quit: I was in the grocery store. I saw a sign that said "pet supplies." So I did. Then I went outside and saw a sign that said "compact cars"...)
00:42:47 Join I3uckwheat [0] (~chatzilla@
00:42:53 Part b0hoon ("GTG. Bye.")
00:43:03I3uckwheati have a quick question
00:43:12saratogastripwax: do you have any theory why the new mdct lib doesn't seem to perform much better in libtremor then the old code?
00:43:25saratogaeven on arm9 the performance increase seems minimal
00:43:33saratogabut in rockbox it was huge
00:43:35 Join TopyMobile_ [0] (
00:43:42I3uckwheatdoes anybody know if the rockbox team is working on the sansa fuzev2
00:44:01saratogaI3uckwheat: i think a few people have, check the wiki page for the fuzev2
00:44:12I3uckwheatk will do
00:45:01 Quit Adubb (Ping timeout: 240 seconds)
00:45:11saratogaRoronoa: fixed point math isn't really a codec issue, it was very common for all sorts of things in the days before common CPUs could afford FPUs
00:45:21stripwaxsaratoga - yes (I think I told you my theory when I was last on irc :-) )
00:45:24 Quit planetbeing (Quit: planetbeing)
00:45:34saratogafor instance, doom on the PC used fixed point math so it could run on PCs without an i387 unit
00:45:44Unhelpfulbresenham's predates *common* FP hardware, doesn't it? also the related circle algo?
00:45:47saratogastripwax: yeah but someone else on the tremor list benchmarked it and saw no real benefit
00:45:52stripwaxi.e. that those processors perform better on gcc code, and that the optimisations are mostly loop unrolling / avoidance of branches / ldm / stm
00:46:02I3uckwheatwow they need to go quite a way before sansa fuze v2 comes out
00:46:21stripwaxif they could benchmark on arm7tdmi they hopefully ought to see a bigger improvement
00:46:31saratogastripwax: funman's ARM9 results weren't very impressive either
00:46:46saratogathe improvement was marginal, whereas in rockbox arm9 got significantly faster
00:46:49stripwaxmy other guess would be that the complexity of the indexing offsets the reduced number of multiplies especially on cores where multiplication requires fewer cycles than arm7tdmi
00:46:51 Quit I3uckwheat (Client Quit)
00:46:57stripwaxwhich targets are arm9?
00:46:57saratogaunless our ASM version of the old MDCT is actually slower on arm9 :)
00:47:09stripwaxdoes the libtremor patch include icode/idata usage?
00:47:15stripwaxit may well be
00:47:19saratogastripwax: AMS, Nano2G, Gigabeat F
00:47:42saratogastripwax: it doesn't, but afaik, no one has tested on hardware with IRAM anyway
00:47:49stripwaxi thought improvement was minimal on nano2g. for vorbis anyway. big improvement for wma. i might be misremembering though.
00:47:52saratogaof our arm9 targets, only the nano2G has fast IRAM
00:48:20saratogawell we're comparing against the ASM optimized old mdct in rockbox, whereas libtremor uses a dumb c version
00:48:29saratogaat least on PP the speed up for ASM was fairly large
00:48:36Unhelpfulsaratoga: arm9tdmi is *mostly* arm7tdmi in terms of execution times. i think the major change is 1-cycle (with delayed result) loads, and that loads can retire out-of-order at all. arm7tdmi loads are 3 cycles. more if they block on RAM.
00:48:52 Join moos [0] (moos@rockbox/staff/moos)
00:49:07Unhelpfulldm/stm timing changes a bit too... would have to check ASDG for details
00:50:54saratogause the load/store improvement is much smaller on arm9
00:50:54stripwaxif the dumb c mdct is faster on arm9 than either the old asm mdct or the new asm mdct then we should use the dumb c, really.
00:50:54 Join Adubb [0] (~aldubuc@
00:50:54saratogai guess i need to port the old mdct's asm to libtremor and see if its faster
00:51:56stripwaxquite why the improvement was so big on nano2g for wma, I simply don't know. caching maybe. no idea.
00:51:56Unhelpfulsaratoga: well, on arm9tdmi you can actually benefit from trying to load constants early, though. arm7 loads don't retire early. pretty much nothing does, to my knowledge, on arm7.
00:53:20stripwaxalso, for all i know, my hand-coded arm asm is *terrible* for pipelining on anything other than arm7tdmi .
00:54:30stripwaxthe mag/ang stuff might perform worse on targets that have good branch prediction.
00:54:44Unhelpfulsaratoga: yeah, arm9tdmi ldm is 1c/reg, 1c extra delay for availability of last register loaded, 1c extra delay for using ldm on 1 register
00:55:08stripwaxsaratoga - let me know if you want any more ideas of why performance improvement might not have been as great as expected! :-)
00:55:18Unhelpfulldr is 1c, result not available for 2c more
00:55:27Unhelpfulso difference is mainly codesize
01:03:28 Quit DerPapst (Quit: Leaving.)
01:03:51 Join JdGordon [0] (~7bf38c1f@gateway/web/freenode/x-iugbzmhbqfdgtotk)
01:04:00 Quit liar (Ping timeout: 258 seconds)
01:08:32 Join planetbeing [0] (~planetbei@
01:10:54 Quit Farthen (Ping timeout: 258 seconds)
01:15:45stripwaxon non-arm7tdmi, you could get better performance by interleaving loads and calculations, as opposed to doing a single ldm load followed by calculations that depend on the loaded registers.
01:17:33***Saving seen data "./dancer.seen"
01:22:11JdGordonwasthere a nodo feature talked about recently I missed?
01:23:17 Join xiainx [0] (xiainx@wpa106013.Wireless.McGill.CA)
01:23:31saratogaJdGordon: configurable button and menu in the nano2g bootloader
01:24:43stripwaxsaratoga - wait, asm_mcf5249.h is commented out from fft-ffmpeg.c and mdct-ffmpeg.c in your patch? so you're losing out on the whole MULT31/MULT31/X(N)PROD31 asm code?
01:25:07saratogastripwax: yeah i didn't commit anything specific to coldfire
01:25:24 Join yawny_ [0] (
01:25:34saratogaerr submit
01:26:01stripwaxer I mean asm_arm.h
01:26:24stripwaxunless I'm missing how that is #included via something else perhaps?
01:26:53stripwaxjust from reading that patch; I haven't tried building or anything. (although it should be pretty clear from the disassembled output whether it uses the asm_arm code or not)
01:27:07saratogai hope its this simple!
01:27:15stripwaxasm_arm.h doesn't appear in a grep of that patch file ...
01:27:59stripwaxoh wait, it appears in the makefile. does that work? doesn't the .c actually have to #include it?
01:28:08saratogastripwax: misc.h includes it
01:28:47 Join alexbobp_ [0] (~alex@
01:28:56 Quit JdGordon (Ping timeout: 252 seconds)
01:28:59stripwaxoh - right, there was already an asm_arm.h ? yeah ok so in that case ignore me...
01:29:16stripwaxyeah sorry..
01:29:29saratogait would have been nice if that was the problem!
01:29:34stripwaxhehehe. yeah.
01:29:47stripwaxand definitely compiling with _ARM_ASSEM_ defined too ... ?
01:30:01 Quit xiainx (Ping timeout: 240 seconds)
01:30:02 Join rasher_ [0] (~rasher@rockbox/developer/rasher)
01:30:04 Join JdGordon [0] (~7bf38c1f@gateway/web/freenode/x-vlxqurebpwazmmdp)
01:30:04saratogawell thats up to whatever make tool they use, but i hope so
01:30:10saratogafunman's definitely defined it
01:30:41 Join Strife1989 [0] (
01:30:42saratogaerr gevarts I mean
01:30:49 Join sevard_ [0] (sev@
01:30:54stripwaxwould be worth asking funman to check a disassembled output.
01:30:56stripwaxor gevaerts even
01:31:16 Quit alexbobp (Ping timeout: 260 seconds)
01:31:16 Quit elcan (Ping timeout: 260 seconds)
01:31:16 Quit bieber (Ping timeout: 260 seconds)
01:31:17 Quit SirFunk (Ping timeout: 260 seconds)
01:31:17 Quit rasher (Ping timeout: 260 seconds)
01:31:17 Quit sevard (Ping timeout: 260 seconds)
01:31:17 Quit Xerion (Ping timeout: 260 seconds)
01:31:18 Nick rasher_ is now known as rasher (~rasher@rockbox/developer/rasher)
01:31:57 Join xiainx [0] (
01:31:57 Join Xerion [0] (
01:32:51 Join SirFunk [0] (
01:33:02 Quit Strife89 (Read error: Connection reset by peer)
01:33:10 Nick Strife1989 is now known as Strife89 (
01:33:14saratogastripwax: i can probably get it from my qemu machine later
01:34:36 Join bieber [0] (~bieber@
01:41:55Unhelpfulldm on arm9e is basically the same as on arm9tdmi, arm11 (only the beast) makes ldm a *huge* improvement over ldr again as ldm becomes one cycle and ldr 1 or 2 depending on address mode
01:45:08stripwaxsaratoga - just looking at FasterMDCT - there must be more to building the tremor example than just 'make example' .. I'm not even seeing a makefile.
01:46:31 Join mt [0] (~mtee@rockbox/developer/mt)
01:47:18 Quit mt_ (Ping timeout: 258 seconds)
01:48:30 Join CaptainKewl [0] (
01:53:17 Join xooch [0] (
01:55:06stripwaxsaratoga - hm, on a default compile (on an arm platform - here, nslu2) it doesn't seem to define _ARM_ASSEM_ . That might just be me though. But that's not a particularly exotic platform (i think, anyway)
01:55:07 Join robin0800 [0] (
01:57:18stripwax(also warning: fft.h no newline at end of file)
02:00:54saratogastripwax: well that would explain the results for the nslu2 on the tremor mailing list
02:01:03saratogacan you force it and see how much difference it makes?
02:01:55stripwaxsaratoga - by the way, there was a previous problem with arm_assem not being defined in Tremor, but which I had assumed was fixed in Trunk.. ? a quick google for ARM_ASSEM and armv5teb should find it
02:02:33stripwaxbasically the configure file matches uname against arm-*-* which just doesn't work.
02:03:03stripwaxbut I see that same problem is still there in the trunk. sigh. Maybe the nslu2 guys fixed it in their libvorbisidec package, only, without upstreaming it.
02:03:42saratogastripwax: if theres a fix we should send it to the list
02:04:08saratogamaybe if we bug monte enough we can get some rockbox people commit access for tremor :)
02:04:35saratogathen fix tremor enough that rockbox CAN run stock tremor
02:04:36 Join froggyman [0] (
02:05:32stripwaxwhat are you using to time the difference in decode speed? i'm not going to get time to do that now (must sleep..)
02:05:45 Join planetbeing_ [0] (
02:06:10saratogastripwax: just "time ./vorbis ...
02:06:37saratogathen running maybe 2-3 times and averaging just to make sure
02:06:38stripwaxok. might be able to get to that some time this week... no promises...
02:06:43saratogaok cool
02:06:50saratogano hurry, tremor has sat there for years already
02:06:54stripwax:) gnight
02:06:55saratogabut i do want to fix it up this spring
02:07:47 Quit stripwax (Quit:
02:10:37 Join nofish [0] (
02:13:29 Quit ps-auxw (Ping timeout: 248 seconds)
02:14:05 Quit perfectdrug (Ping timeout: 245 seconds)
02:15:17 Quit mt (Read error: Connection reset by peer)
02:15:27 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
02:15:59 Nick alexbobp_ is now known as alexbobp (~alex@
02:17:11 Join perfectdrug [0] (
02:17:18 Quit perfectdrug (Client Quit)
02:18:36 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
02:20:41 Quit nofish (Quit: CGI:IRC)
02:25:21 Quit TheSeven (Read error: Connection reset by peer)
02:25:46 Join TheSeven [0] (~theseven@rockbox/developer/TheSeven)
02:25:48 Join saratoga_lab [0] (~9803c20d@gateway/web/freenode/x-bjmlgmvndtlxwpmi)
02:28:35xoochHi, I'm a student and have been looking at the rockbox gSoC projects for this year, and although a few of the projects look good, I'm quite interested in improving video playback
02:29:28linuxstbxooch: Great! No need to wait until the summer though, you can get involved with Rockbox now...
02:30:41xoochI'm more than happy to, I just need to talk to people about the project more in depth, or places to get started
02:30:57JdGordonwell you've come tot he right place :)
02:31:03saratoga_labsure, which part looked interesting?
02:32:05xoochit was mainly trying to getting more video codecs working and integrating them into core playback
02:32:49linuxstbWhich codecs?
02:33:55 Quit planetbeing_ (Quit: planetbeing_)
02:35:06xoochon the site it mentions integrating mpeg4 and aac into normal playback, but getting new codecs to run such as maybe xvid would be good (although I don't know how difficult this would be)
02:35:33saratoga_labwell xvid is mpeg4 :)
02:36:15saratoga_labwell its an implementation of parts of the MPEG4 spec
02:36:32saratoga_labbut you should probably define what you want to decode, not what encoder you will use to encode it
02:36:42saratoga_labxvid can actually produce all sorts of different types of mpeg4 video
02:36:51saratoga_labmany of which will not be practical to support
02:37:23 Quit komputes (Quit: I haven't slept for ten days, because that would be too long.)
02:37:42saratoga_labour you familar with video codecs?
02:38:17xoochnot into any real depth
02:40:01xoochbut it's something I'd want to learn about and know better
02:40:38 Part froggyman
02:41:18saratoga_labhave you done any DSP or image processing?
02:41:35saratoga_labi don't think we have any video codec people here exactly, though many of us have a rough idea how they work
02:41:49saratoga_labthats not required, just curious what your background was
02:42:05saratoga_labsome people are more interested in the math, other the implementation
02:42:10saratoga_labboth are quire useful . . .
02:42:30saratoga_labquite useful
02:44:15 Quit Casainho (Ping timeout: 258 seconds)
02:44:19saratoga_labxooch: do you have a rockbox device?
02:44:46xoochi'm just trying to find a project i'd be suitable for, and finding something i could do well would be good since i'd like to be involved outside of gsoc because it's something i'm interested in
02:45:33xoochi got frustrated with having to constantly covert my flac to mp3 to put on my ipod
02:45:40saratoga_labif you're interested in mpegplayer, looking through it and discussing how it can be improved in your application will make us much more likely to believe you're the right man/woman for the job
02:45:41xooch*flac files
02:47:08saratoga_labwhen i applied to port the wma decoder, I started by looking at the code and making a good argument in my application that the project should be feasible based on the work to be done
02:47:37 Quit Strife89 (Quit: Night.)
02:47:54xoochok, i'll start having a look through that now
02:48:32saratoga_labdiscussing things before hand on IRC also probably helps you come up with good ideas
02:54:36xoochsince i haven't done any coding involving video, would a different project be more suitable or do you think this would be ok?
02:55:22saratoga_labyou don't have to have done something like this before
02:55:35saratoga_labbut you do have to convince us that you're likely to finish the project :)
02:56:09saratoga_labare any other ones interesting to you?
02:56:25saratoga_labkeep in mind you can propose other projects not on the official list, though i suggest talking those over with us first
02:58:51 Join Blue_Dude [0] (
02:59:11Blue_Dudegevaerts: Thanks for starting the NoDo discussion.
02:59:50saratoga_labBlue_Dude: what about you, any interest in helping out with GSOC this year?
03:00:19Blue_DudeUm, maybe. My schedule is month to month. No way of knowing if/when I'll have time to commit to it.
03:00:59Blue_DudeI'm not really a pro developer. I just implement cool stuff I happen to want. :)
03:01:25saratoga_labsame here
03:02:17Blue_DudeThat makes it fun instead of work.
03:02:57saratoga_labmentoring is mostly done over email or gtalk, so you may want to keep an eye on things and maybe be a backup mentor if someone does a project in your area
03:04:00Blue_DudeOK. I can probably assist with DSP stuff since I spent most of my time there. I'll keep an eye out.
03:04:09Roronoasaratoga: I would like to discuss with you the project New WMA audio codec which i left off yesterday before sleep
03:04:31saratoga_labsure go ahead, i'm doing lab work but i can reply
03:06:25 Quit robin0800 (Remote host closed the connection)
03:06:29Roronoabasically i would like to know where do i start reading for Audio Codec Theory
03:06:52Roronoahow much am i supposed to know before starting the project
03:07:21saratoga_labwell i knew next to nothing about it aside from some generic DSP stuff when I applied for the original wma project
03:07:31saratoga_labi would hope you have a basic idea of what an FFT does
03:07:39Roronoaas far as the coding goes i know C and have worked on assembly in x86 and PIC's DSP
03:08:12saratoga_labthats mostly what you'll need
03:08:27saratoga_labporting a codec doesn't have to involve understanding how it works, although it may be easier if you do
03:11:26Roronoais Assembly of ARM quite diff from what i have done earlier or is it nearly the same
03:11:49saratoga_labi haven't done any x86 or PIC, but arm isn't too bad in general
03:12:03saratoga_labits very compact RISC with 16 registers
03:12:18Blue_DudeRoronoa: the Rockbox codec API is pretty straightforward. You could start with a shell codec that has placeholder components and fill in the blanks as you develop the decoding engine. You probably won't have to do much if any assembly code.
03:12:21saratoga_labvaguely like MIPS if you've done that, although somewhat less readable
03:14:29saratoga_labRoronoa: are you familiar with the fourier family of transforms?
03:14:39RoronoaRISC with 16 registers seems similar i think i may be able to do it
03:14:50saratoga_labnot at all?
03:15:11RoronoaI mean i have done in Mathematics not implement it
03:15:29saratoga_labah ok
03:15:33Roronoawe had fourier transform of complex functions
03:15:50saratoga_lablet me put it this way, if I asked you the difference between a fourier transform and an FFT could you tell me ?
03:16:09saratoga_lab(without google)
03:16:22 Join arbingordon [0] (
03:16:25saratoga_labok thats fine
03:16:44Roronoashall i look into that
03:16:46saratoga_labif you've done them on paper you can pick up the rest in practice, but learning the calculus from scratch would have been hard
03:17:28saratoga_lababout codecs, the wikipedia page on JPEG compression actually lays out the general theory that all lossy codecs use, although the DCT in that codec will be replaced with the closely related MDCT in WMA Pro
03:17:34***Saving seen data "./dancer.seen"
03:18:53Roronoaok i will look into that
03:20:34saratoga_labbut the basic idea is to transform to some domain where useless detail can be extracted, and then just compress the remainder using something like huffman coding or whatever lossless strategy you want
03:21:12saratoga_labfor video, a DCT is used, for audio, an MDCT is used
03:22:54Roronoaok. i currently don't know much about that but i will look into it and tell you soon (may be by today)
03:23:03saratoga_labi described wma decoding a little bit here, pro is very similar:
03:23:11saratoga_labok sure
03:23:19saratoga_labdon't mean to pile on information mid semester
03:24:16Roronoano it is not a problem
03:26:35Roronoathese days mostly i am involved in microcontroller coding on MSP430 but i think i can take out some time till the application date for GSoC the rest i will start when my vacations start by april end
03:33:21 Join naag [0] (~harish@
03:36:35 Quit avacore^ (Ping timeout: 245 seconds)
03:38:12 Join avacore [0] (
03:39:05 Quit adnyxo (Ping timeout: 246 seconds)
03:40:22 Join panni__ [0] (
03:40:42xoochsaratoga: to answer a previous question the other projects i am interested in are making multiple-volumes look like a single volume in the files menu or improving/adding online services
03:41:19 Quit panni_ (Read error: Connection reset by peer)
03:41:48 Quit Blue_Dude (Quit: ChatZilla 0.9.86 [Firefox 3.6/20100115144158])
03:42:49 Quit planetbeing (Quit: Poof.)
03:48:41CIA-5New commit by FlynDice (r25314): sd-as3525v2.c Adjust comments for MCI_INT bits to show if they are related to CMD, DATA, or a termination signal. ...
03:53:33 Quit naag (Quit: Leaving.)
03:58:32 Part Roronoa ("Leaving")
03:58:53 Quit JdGordon (Ping timeout: 252 seconds)
04:16:47 Join phanboy4 [0] (
04:21:52 Join Barahir [0] (
04:22:06amrhi, I came with a fix to the diacritics problem in Arabic scripts mentioned here
04:22:53amrthe patch is here
04:23:19 Join CGL [0] (~CGL@
04:24:10 Quit TheSeven (Disconnected by services)
04:24:22 Join The_Seven [0] (~theseven@rockbox/developer/TheSeven)
04:24:32 Nick The_Seven is now known as TheSeven (~theseven@rockbox/developer/TheSeven)
04:25:28 Quit Barahir_ (Ping timeout: 265 seconds)
04:26:36amrI need a confirmation it's ok with other languages and approved to be commited until making the suggested rework of eliminating the need to change drawmode completely
04:27:44 Quit Darkknight512 (Quit: ChatZilla 0.9.86 [Firefox 3.5.8/20100202165920])
04:28:02saratoga_labamr: for what its worth, at this hour IRC has few developers and most are in the USA where fewer people are aware of diacritics :)
04:28:16saratoga_labthough people generally do read the logs so they'll probably see what you wrote when they wake up
04:28:45amrok, thanks :)
04:29:30 Part xooch ("sleep")
04:32:26 Join tromik [0] (
04:37:33 Join Guest23293 [0] (
04:37:42 Quit Guest23293 (Client Quit)
04:37:59tromikhey, just got an e200v2 to replace my e200v1, I downloaded the latest rockbox utility, hit complete install, was warned I needed to download the OF which I got (03.01.16a), I pointed the installer to it and then the installer went through its process, finishes, I safely remove the device, and it boots into the sansa firmware...did I miss something? I've followed the manual and gone through the process a few times, not s
04:41:08 Quit amr (Quit: - Chat comfortably. Anywhere.)
04:46:03 Quit scorche (Ping timeout: 268 seconds)
04:46:38 Join Rob2222 [0] (
04:47:01 Join Gartral [0] (~Gareth@unaffiliated/gartral)
04:49:13Gartralso what are build points? (i seem to generate 183~ish)
04:50:06 Quit Rob2223 (Ping timeout: 265 seconds)
04:51:29bieberIf anyone's still around and/or interested, I'm working on a parser for WPS files (working towards a theme editor), and I've gotten as far as a parser that can properly parse comments, plaintext, newlines, and viewport definitions
04:52:47bieberThe source is on github, "git clone" will grab it (you'll need to use an IDE or roll your own makefile to compile, I just uploaded the source straight from an Eclipse project directory)
04:55:05bieberComments and criticism on the code would be much appreciated (sorry if my C++ isn't the best, it's my newest language), email is bieber@knights.[nospam, please]
05:01:50 Part tromik
05:03:11 Quit evilnick (Remote host closed the connection)
05:03:34 Join evilnick [0] (
05:09:01S_a_i_n_tTheSeven: *way* late (regarding the OF/RB seeking thing...) what I meant by "I've gotten used to it by now" (or whatever it was) was that I don't necessarily expect something I like in the OF to bo present in RB, so I live with it, however...if you *could* do it, it would be great. However, short select goes to the last screen accesed in the filebrowser/database, so I'm not sure *how* you could do it.
05:09:50S_a_i_n_tThough...losing that wouldn;t be *so* bad, as you can just press menu anyway...its kinda 50/50 as to what would be "preferred operation"
05:17:35***Saving seen data "./dancer.seen"
05:18:56 Quit Tomis (Quit: Tomis)
05:20:14 Quit xiainx (Quit: Good Bye)
05:25:27S_a_i_n_tDoes anybody have any ideas regarding FS #10981? I've tried all sorts of combinations with viewers.config /"iconset".icons files with no/similar/worse effects...I'm really quite stumped as to where the problem actually is. RB seems to toss a coin to secide what gets an icon and what doesn't >:(
05:39:47 Quit CaptainKewl (Remote host closed the connection)
05:39:54 Quit Horscht (Quit: Verlassend)
05:41:54 Quit CGL (Quit: Saliendo)
05:42:13 Quit bieber (Quit: Leaving)
05:43:24 Quit panni__ (Read error: Connection reset by peer)
05:45:41 Quit bluebrother (Disconnected by services)
05:45:43 Join bluebroth3r [0] (~dom@rockbox/developer/bluebrother)
05:57:13beta_Is there any way to edit the menu in rockbox?
05:57:56saratoga_labyeah i think its in menu.c
05:58:02beta_I want to give my MIL a Gigabeat F with RB, but I'd like to remove some of the menu items she has no business playing in :)
05:58:51beta_So the menu is hard coded then?
06:00:01S_a_i_n_thas *most* of the things you'd care to remove
06:00:19saratoga_laboh yeah root_menu.c
06:00:31saratoga_labwhich oddly enough is not in the menus folder
06:00:53S_a_i_n_tand also onplay.c (for the labrynthine contect menus...)
06:01:22*S_a_i_n_t has carved his menus back to the bone...
06:03:58S_a_i_n_tAll the changes I made (IIRC) were in root_menu.c, onplay.c and settings.c
06:05:47S_a_i_n_t*they're all in apps I think
06:20:13 Join amr [0] (
06:21:38 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
06:24:45 Quit amr (Client Quit)
06:24:49 Join amr [0] (
06:24:57amrplease ignore my previous message, the last 'fix' caused problems with the moving text of mp3 meta info in the wps, but at least the first version of the patch (bidi.c change only) is woking well - i hope.., thanks
06:25:51 Quit amr (Client Quit)
06:26:47 Join angelwolf71885 [0] (
06:34:17 Join naag [0] (~harish@
06:38:41 Quit naag (Ping timeout: 260 seconds)
06:41:27CIA-5New commit by FlynDice (r25315): sd-as3525v2.c Don't reset controller on every pass through the transfer loop. ...
06:41:35CIA-5New commit by FlynDice (r25316): sd-as3525v2.c Implement HS timings for those cards that are capable. ...
06:46:21 Join ankur [0] (~ankurlnmi@
06:47:02 Quit JdGordon (Read error: No route to host)
06:47:13 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
07:05:16 Quit JdGordon (Ping timeout: 276 seconds)
07:05:23 Quit saratoga_lab (Quit: Page closed)
07:08:20 Quit saratoga (Quit: Page closed)
07:17:37***Saving seen data "./dancer.seen"
07:24:45 Quit mc2739 (Ping timeout: 265 seconds)
07:25:36 Quit BHSPitMonkey (Remote host closed the connection)
07:26:15 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
07:29:50 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
07:31:43 Part ankur
07:34:03 Quit JdGordon (Ping timeout: 246 seconds)
07:47:21 Quit ObsidianX (Ping timeout: 246 seconds)
07:52:56 Join ObsidianX [0] (
07:57:10 Quit ObsidianX (Ping timeout: 240 seconds)
08:00:54 Join B4gder [0] (~daniel@rockbox/developer/bagder)
08:01:59 Join ObsidianX [0] (
08:08:46 Join wodz [0] (
08:13:09 Join bimbel [0] (~Miranda@unaffiliated/bmbl)
08:13:33 Nick bimbel is now known as bmbl (~Miranda@unaffiliated/bmbl)
08:15:00 Join LinusN [0] (~linus@rockbox/developer/LinusN)
08:17:39Gartralwoops, build server went down, sorry
08:23:38 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
08:24:41 Join einhirn [0] (
08:28:45 Join ender` [0] (
08:34:26 Quit jae (Quit: leaving)
08:34:49 Join Zagor [0] (
08:34:49 Quit Zagor (Changing host)
08:34:49 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
08:35:50 Join jae [0] (
08:36:21 Join flydutch [0] (
08:36:36 Join hebz0rl [0] (
08:41:55wodzfunman: ping
08:43:28 Join petur [0] (~petur@rockbox/developer/petur)
08:44:30 Join mitk [0] (~mitk@
08:47:06 Quit JdGordon (Ping timeout: 245 seconds)
08:54:58 Quit arbingordon (Quit: `)
09:17:06 Quit angelwolf71885 (Quit: ChatZilla 0.9.86 [Firefox 3.6.2/20100316074819])
09:17:39***Saving seen data "./dancer.seen"
09:18:05amiconnTheSeven: Sure you get a speedup from dma lcd update. The thread calling lcd_update() will have to wait of course, but other threads can run in the meantime
09:19:27 Join DerPapst [0] (
09:19:53amiconnWe're actually doing dma lcd updates on the H300
09:22:00TheSevenhmm, how much effort would it be to split the lcd stuff into kickoff and completion functions like i did for the ftl?
09:22:41TheSeventhis would mean that even the same thread could continue to run, but needs to call the completion function before it writes to the lcd buffer the next time
09:22:52scorche|shif you are planning on attending, please update the DevCon2010 wiki page with your available dates
09:23:26TheSevenscorche|sh: should that maybe go into the topic?
09:23:59scorche|shperhaps, but i am on my phone at the moment
09:26:23Mode"#rockbox +o TheSeven" by ChanServ (ChanServ@services.)
09:26:44Topic"Please read before speaking: | Please direct offtopic/social chat to #rockbox-community | Rockbox has been accepted for GSoC 2010! Potential students see | If you are planning on attending, please update the DevCon2010 wiki page with your available dates!" by TheSeven (~theseven@rockbox/developer/TheSeven)
09:26:51Mode"#rockbox -o TheSeven" by ChanServ (ChanServ@services.)
09:28:42 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
09:31:56 Join JdGordon [0] (~7bf38c1f@gateway/web/freenode/x-wthposnnjvpzeiwp)
09:43:59*JdGordon decides to ruffle some feathers...
09:44:17JdGordonconsidering the aomunt of requests and patches for some sort of customisable buttons and menus, they REALLY shuoldnt be nodo
09:44:42JdGordonalso, flat out rejecting a menu-ed bootloader is just stupid also. There is no reason we couldnt offer more than one bootloader for each target
09:47:20Tornegiven the number of people who manage not to understand how the *one* bootloader works I'd be dubious about having two
09:48:33 Join phanboy_iv [0] (
09:51:43 Quit phanboy4 (Ping timeout: 248 seconds)
09:52:22 Join xavieran [0] (
09:56:02wodzwhat does it mean menu-ed bootloader?
09:57:38Zagordisplaying a boot menu?
10:00:38wodzdoesn't h100 bootloader display menu?
10:00:59 Join kunal [0] (~kunal@
10:02:46 Join swilde [0] (
10:03:45wodzbootloaders/iriver_h1x0.c void failsafe_menu(void)
10:05:24linuxstbI think that's some kind of recovery mode when Rockbox is flashed to NOR.
10:05:35Zagorwodz: it only does that if the hold switch is engaged
10:06:00Zagorand if it can't detect the original firmware
10:07:20TheSevenwe have that on nano2g for the "FTL borked" case, too
10:07:32ZagorI would call those error screens rather than boot menus
10:08:32wodzWhats wrong with boot menus? Just add timeout and there is not problem at all
10:09:14wodzalmost all 'linux' bootloaders I am aware of do this that way
10:09:20TheSevenprobably as always: some added complexity, more possible problems and a bit more boot time
10:09:26*S_a_i_n_t agrees with this...but never wanted to be the one to open this can of worms in IRC
10:09:35ZagorS_a_i_n_t: :-)
10:09:37JdGordonthe boot menus people are asking for are to let you choose which OS to boot into, i.e ipodlinux, OF, us
10:09:55TheSevenS_a_i_n_t: we've already opened that can several times, and so far, all the worms have always hopped back into the can in the end :-P
10:10:18S_a_i_n_tyes, I've seen worms fleeing this channel :P
10:11:10JdGordonthe support "reason" against any sort of button fiddling is just nonesense
10:11:42S_a_i_n_tI can't remember who, but someone mentioned earlier there being a physical, and an unwitten no-do list...
10:11:50*S_a_i_n_t demands to see the unwritten one :P
10:12:36TheSevenwell, judging from the recent mailing list traffic, those will need to be written, too, or they'll just be killed
10:14:22TheSevenand yes, fancy bootloaders would definitely cause some additional support traffic, but we could just make them unsupported (as they currently are), maybe besides some FAQ list
10:14:34JdGordonthe problem is there really isnt such a listing, it entirely depends on who is around to complain (change is bad remember? )
10:15:24TheSevenbut yes, that's way incomplete
10:15:29linuxstbTheSeven: I can't see how they would cause *more* support traffic. Almost every day someone asks "how can I boot the original firmware?"
10:16:09scorche|shZagor: do you have any clue why the wiki doesnt show certain column separations in tables as seen on the DevCon2010 wiki page?...the WYSIWYG page and the preview both show the separations...
10:16:30TheSevenwell, yes, if a fancy bootloader based on the current rockbox bootloader framework would probably not cause much support issues, but something like iloader will
10:16:43Zagorscorche|sh: which separations?
10:17:13TheSevenand in the end, one of the reasons for this is to get rid of those annoying FTL screwups, which will force us to use nor-based bootloaders, which are likely to cause additional trouble
10:17:44scorche|shZagor: at the moment i see no separation between any of the first 2 columns on
10:17:54JdGordonTheSeven: what I'd suggest is offering 2 bootloaders for those targets, it's not like it would add much extra work once the initial gui is done
10:18:03Topic"Please read before speaking: | Please direct offtopic/social chat to #rockbox-community | Rockbox has been accepted for GSoC 2010! Potential students see | If you are planning on attending, please update with your available dates!" by ChanServ (ChanServ@services.)
10:18:06Zagorscorche|sh: really? such as between "name" and "location"?
10:18:20scorche|shZagor: errr...i take that back...firefox misses them, but chrome shows them
10:18:44Zagormy iceweasel 3.5.8 shows them fine
10:18:46TheSevenscorche|sh: my FF3.6 (windows) shows them
10:19:33JdGordonff 3.5.8 here looks fine also
10:19:47*scorche|sh updates firefox from 3.5.7 (windows)
10:20:44TheSevenJdGordon: basically there are 4 combinations that we could have:
10:20:45TheSeven- NAND-based trivial bootloader (the current RB bootloader)
10:20:47TheSeven- NAND-based fancy bootloader (the old iLoader, deprecated)
10:20:48TheSeven- NOR-based trivial bootloader (not implemented yet, and not much point in it)
10:20:50TheSeven- NOR-based fancy bootloader (the current iLoader)
10:21:30TheSevenhaving them NOR-based allows to get rid of those FTL problems, why it doesn't neccessarily mean that the bootloader needs to be fancy
10:21:37JdGordonok, but in general, there is no real reason why we couldnt have a fancy and boring bootloader for each target
10:21:55TheSevenon the other hand, NOR-based bootloaders are likely to cause installation trouble, depending on the exact way of installation and NOR backup creation
10:22:18linuxstbTheSeven: Why is there not much point in a NOR-based trivial bootloader?
10:22:23 Join xsteadfastx [0] (~spectrum@
10:22:59TheSevenone point why fancy bootloaders is opening a can of worms, is that people will ask for more and more questions, themeability etc. (roughly like what iLoader can do now)
10:23:57TheSevenlinuxstb: because the existing fancy one is working just fine, and can simulate the trivial one's behavior with no noticable boot time overhead
10:24:25TheSevenbut yes, we could try to port the rockbox bootloader to NOR
10:24:33JdGordonTheSeven: only stupid people want themeability in the bootloader! so let them do it on their own
10:25:26S_a_i_n_t<feels stupid>s wants his bootloader to look pretty too! <.feels stupid>
10:25:32linuxstbTheSeven: That's not a reason IMO - a NOR bootloader of some kind still needs to be implemented in Rockbox.
10:25:34JdGordonso you shold!
10:25:42JdGordon(feel stupid that is)
10:25:58TheSevenpeople will always like if their bootloader look and feel matches their rockbox theme
10:26:06TheSevenand it isn't hard to implement either
10:26:18S_a_i_n_t*that* what I should have said...
10:26:45linuxstbTheSeven: But I'm not strongly against a boot menu either, but given the time that Rockbox boots, maybe implementing a new "boot screen" in Rockbox would be a better way to do it. That way, you get full theming, the existing UI widgets, even a voice UI...
10:27:17TheSevenyou mean a boot menu plugin run through autorock? :-P
10:27:38S_a_i_n_tbut it would take longer to boot....say those that desire no fancy boot options.
10:27:39linuxstbMaybe, yes. And users can set their "start screen" to that if they want.
10:27:48S_a_i_n_tIs *that* why its "no-do"?
10:28:16TheSevenS_a_i_n_t: rockbox would boot equally fast, because it would just need to terminate the boot menu if you choose it
10:28:28JdGordonlinuxstb: can we rolo the OF on any targets?
10:28:31TheSevenhowever, all other firmware would boot like 3-5 seconds slowe
10:28:46TheSevenJdGordon: we can rolo about everything on nano2g
10:28:49ZagorS_a_i_n_t: it's not a nodo just because nobody wants to do it
10:29:23TheSevenwhat would really bug me with that rolo approach is that it will sit there for several seconds to do an FTL sync
10:29:25linuxstbJdGordon: If not, we would need to try to implement it... I've always thought a "start OF" option in the main Rockbox menu would be nice.
10:30:21 Quit kunal (Ping timeout: 260 seconds)
10:30:44TheSevenhm, we can rolo things like ibugger, but neither the OF nor disk mode right now, because we trash the sysconfig during boot
10:30:52TheSevenso we would either need to keep or recreate it
10:31:23TheSeventhat wouldn't be hard of course, but it would mean even more target-specific code in rolo
10:31:45JdGordonrolo SHOULD be target specific thoguh?
10:32:08TheSevenprocessor-specific, rather
10:32:26 Join will__ [0] (
10:33:09linuxstbTheSeven: The sysconfig stuff should be saved for older ipods as well - so it's target specific, not processor.
10:33:23TheSevendo they even do it the same way on the older ones?
10:35:00linuxstbI don't know how the Nano2G works, but the NOR bootloaders on older ipods store a sysconfig struct at the end of IRAM.
10:35:47will__I accidentally created an "Active Undelete" (it's a commerical program) recovery USB device out of my e250v1. I haven't shut it off yet, but I'm pretty sure the partition table and MBR got messed up. Since it is still booted, is there an easy way to fix this?
10:35:53TheSevenon nano2g, it's just a pointer at the end of IRAM and the sysconfig struct is somewhere in the IBSS of norboot, so it could be overwritten by our IRAM copying code
10:36:52linuxstbI can't remember the details on the older ipods, but I would be surprised if Apple changed it. But yes, it needs saving by (probably) the crt0.S code for the bootloader.
10:36:58TheSevenso i'd vote for re-creating it during rolo
10:37:07linuxstbYes, or recreating...
10:37:35TheSeventhe code for that is already in iloader, to be able to boot the OF, so that would be a matter of copy'n'paste
10:37:55linuxstbHow much code is that? Wouldnt' it be simpler to just save/restore?
10:38:15TheSevenlinuxstb: around 50 lines
10:38:25TheSevenso yes, save and restore might be easier
10:39:23linuxstbAlthough that would mean people would have to update their bootloaders.... So maybe recreating it in Rockbox would be a better way to do it.
10:39:36linuxstbOr do you save any parts of it?
10:40:40TheSeventhe current rockbox bootloader will trash it, iloader will re-create it at the end of iram (immediately before the pointer to it)
10:41:45linuxstbYes, what I meant was whether iloader requires any of the original values when recreating it? I've forgotten what it actually contains.
10:42:08gevaertswill__: I think the easiest way to fix it will be using either recovery or manufacturer mode, depending on how messed up it is exactly
10:42:26JdGordonhas anyone tried the mr500 or d2 sims recently?
10:42:28TheSevenno, it doesn't, I currently take everything from the nor flash's SCfg list and it works just fine
10:43:27linuxstbTheSeven: OK. Then I think your current approach would be best for Rolo - so it's independent of the bootloader.
10:44:09 Join kunal [0] (~kunal@
10:44:20JdGordonsomeone broke the touchscreen sims :(
10:48:10TheSevenlinuxstb: would you prefer a list-style or button-style menu?
10:48:50linuxstbAre you talking about a menu in the bootloader, or my suggestion of a "boot screen" in Rockbox itself?
10:49:32TheSeventhey would probably serve similar purposes ;-)
10:50:19S_a_i_n_tlist-style would fit in better with RB would it not?
10:50:26*S_a_i_n_t 's two sents
10:50:30TheSevenmore points we need to take care of when using nor-based bootloaders are a) creating a backup of the old one, b) a way to easily update it and c) how to run disk mode etc.
10:50:31linuxstbThe answer would be different for technical reasons. If it's the bootloader, then I don't really care. If it's in Rockbox itself, it should use the existing menu widget, and hence you don't have a choice.
10:51:27linuxstbAt the moment at least, the bootloaders can't use any code from apps/, where all the Rockbox UI code is...
10:51:40TheSeventhey can :-)
10:51:45TheSeventhey just shouldn't...
10:53:46 Quit will__ (Quit: ChatZilla 0.9.86 [Firefox 3.6.2/20100316074819])
10:54:41linuxstbTheSeven: My main question about the NOR bootloader, is whether we would want to make it compulsory - i.e. drop support for a NAND-based bootloader.
10:55:34linuxstbIf it's optional, then it can simply be managed by a Rockbox plugin, the same way flashing things to NOR happens for other targets.
10:56:31TheSevenactually the only way to bootstrap a nor-based setup would be through a nand-based thing, whatever it is
10:57:14TheSevenso we could of course just boot rockbox directly from NAND (that doesn't even need a bootloader stage in between) and flash the NOR bootloader from there
10:57:37TheSeveni wouldn't really advise anyone to use NAND-bootloaders for everyday use though, because of that FTL issue
10:59:34TheSeven("white screen of death" :-P )
11:00:24S_a_i_n_t"Please Restore Using..." ...*fuck!*
11:01:18TheSevenS_a_i_n_t: I even had some almost-bricked situations caused by that dumb bootloader, linuxstb also had one once
11:02:18TheSevenif you manage to put the FTL into a state where every try to write to the firmware partition just results in an I/O error, you have a *big* problem
11:03:17TheSeventhose were of course caused by rockbox bugs, but such things shouldn't *ever* be able to happen.
11:03:48TheSevenoh wait, nowadays we could just have used NOR DFU to recover from that :-)
11:04:00TheSevenbut we hadn't investigated that back then
11:06:25linuxstbTheSeven: So you think we should go NOR-only for Nano2G?
11:06:38S_a_i_n_tMy worst situation that I got into was "white screen of death" caused by simply inseting the USB cable that I couldn't get out of...the PC couldn;t see it and neither could ipodpatcher. But the battery diesd and after that itunes magically found it.
11:06:42TheSevenif we want to get rid of those WSODs, yes
11:07:29TheSevenhowever that means a more complex way than just "use itunes to restore" to go back to the factory state
11:07:55linuxstb"use itunes to restore" isn't our documented uninstallation method anyway.
11:08:03S_a_i_n_t*thats* one thing I think users could be dubious about...
11:08:09TheSevenbut it's the only failsafe one right now
11:08:49TheSevenNOR based bootloader + serious FTL screwup is a bit evil, too
11:09:43TheSevenyou would need to trigger a FTL reinit (in most cases this will happen when you just run disk mode twice), then manually reformat it the right way and extract rockbox again
11:10:11TheSevenit isn't really nice that you need to have a working rockbox installation to remove the bootloader
11:11:59 Join liar [0] (
11:12:19linuxstbHow much NOR do you need? Could a ucl-compressed version of the original contents of the NOR be re-written to the NOR?
11:12:37linuxstbSo the NOR bootloader itself has everything it needs to restore, without reading anything from NAND.
11:14:05TheSevenit's all encrypted, so it won't be compressible
11:14:42TheSevenwe might fit a rockbox bootloader into the free space on the nor though (and just relocate things a bit)
11:14:42linuxstbCan't we decrypt it?
11:14:55linuxstb(and then re-encrypt) ?
11:15:19 Quit mrigesh (Ping timeout: 264 seconds)
11:17:42***Saving seen data "./dancer.seen"
11:19:08TheSevenwe could, and that's what iloader is doing
11:19:15TheSevenon-device ucl is dog slow though
11:20:07 Join TheSeven|Mobile [0] (~theseven@rockbox/developer/TheSeven)
11:21:31linuxstbYou mean on-device ucl compression, or decompression?
11:21:46TheSeventhat's around 20K/sec or something
11:21:51 Quit TopyMobile_ (Quit: TopyMobile_)
11:22:17linuxstbOuch... Maybe that can be optimised... Did you experiment with different ucl compression options?
11:22:42TheSeveni do NRV2E-10-singleblk like the decompressor expects :-/
11:24:47TheSeven|Mobilenot to even mention the 8MB of malloc space it needs to compress a 200K file
11:25:19TheSeven|Mobileif you tell it to allocate buffers on the stack, the 32MB the nano has don't even seem to be sufficient
11:25:54 Join will__ [0] (
11:26:30will__gevaerts: does C200ERASE work on e200v1?
11:26:51gevaertsno, it's not needed
11:27:23 Quit TheSeven (Ping timeout: 258 seconds)
11:27:30gevaertsc200erase only exists because the sansa.fmt thing doesn't work on c200
11:30:00 Quit antil33t (Read error: Connection reset by peer)
11:31:41 Join antil33t [0] (
11:32:36TheSeven|Mobilelinuxstb: i'll probably have a look at recovering an ipod from an ipsw file
11:32:36wodzwhat partition types rolo suport?
11:32:57TheSeven|Mobilewhat does rolo have to do with partition types?
11:33:17TheSeven|Mobileit will support everything that rockbox supports
11:33:39linuxstbwodz: Do you mean rolo, or the Rockbox bootloader ?
11:33:52wodzRockbox bootloader
11:34:02wodzI thought it is the same
11:34:09will__does e200tool work on Ubuntu x64? If so, why do I get this when I try to run it: ./e200tool: No such file or directory
11:34:12linuxstbrolo is the feature in Rockbox itself that can load and run a firmware file.
11:34:34linuxstbwill__: "chmod +x e200tool" ?
11:34:41linuxstbi.e. make it executable.
11:34:57will__i did
11:35:07linuxstbWhat does "ls -l e200tool" show?
11:35:31linuxstbIn fact, ignore me, that simply means you are not in the same directory as e200tool.
11:35:43will__-rwxr-xr-x 1 will will 29073 2010-03-24 05:03 e200tool
11:36:17linuxstbOK, then that makes no sense to me...
11:37:07will__do x86 binaries run in x64 Linux?
11:37:25gevaertsthis one definitely does
11:37:46gevaertswell, if you have the compatibility stuff installed
11:37:53gevaertsWhat does "ldd e200tool" show?
11:38:27will__not a dynamic executable
11:38:53gevaerts"md5sum e200tool"?
11:39:14will__795cc4e75d38284a1460b840fc14967a e200tool
11:39:29gevaertsok, so the file itself is OK
11:39:56will__Ubuntu 8.04
11:42:55will__If the player is not detected by the OS −−- could that cause "No such file or directory"?
11:43:02 Quit wodz (Quit: Leaving)
11:45:16will__I'm gonna boot into an x86 Ubuntu live CD
11:45:21 Quit will__ (Quit: ChatZilla 0.9.86 [Firefox 3.0.17/2010010604])
11:47:29 Join webguest40 [0] (
11:50:21TheSeven|Mobilewould it be a reasonable way for recovery to have the user reformat the data partition.n
11:50:57TheSeven|Mobilecopy some files to it, download an ipsw at phobos, rename and unzip it
11:51:34TheSeven|Mobileplace the firmware file on the ipod and reboot it?
11:52:22TheSeven|Mobileor would it be better to just let them use itunes to restore, then copy some files to it and reboot?
11:52:30TheSeven|Mobileor maybe support both?
11:52:59 Quit webguest40 (Quit: CGI:IRC (Ping timeout))
11:53:40S_a_i_n_tboth my be harder...but covers more bases.
11:53:48S_a_i_n_tie. more people will be happy.
11:53:57S_a_i_n_twhich in the end, is a good thing I guess.
11:54:32 Join _silentAssassin [0] (
11:55:04S_a_i_n_tiTunes to restore, copy some files, reboot...would *probably* be the better option for those with little to no technical background.
11:55:26 Nick krazykit` is now known as krazykit (
11:56:04TheSeven|Mobilethe linux users could just feed it the ipsw or restore it using a dd image
11:57:18TheSeven|Mobilei'll probably implement the itunes way first
11:57:42TheSeven|Mobilewell, first i'll have to figure out how to grab norboot from an AUPD
11:57:43S_a_i_n_tIt seems the "safe"
11:58:23 Join ubuntu [0] (~chatzilla@
11:59:20 Nick ubuntu is now known as will__ (~chatzilla@
11:59:39will__e200tool is working in Ubuntu 9.04 x86
12:04:17 Nick _silentAssassin is now known as mrigesh (
12:10:40*S_a_i_n_t has 7 unsecured wireless connections to choose from at his location ;D
12:10:52S_a_i_n_tshit...wrong channel.
12:16:41linuxstbTheSeven|Mobile: What kinds of recovery are you talking about? I wouldn't confuse "uninstallation of rockbox" with "recovery of an unusable ipod".
12:20:52 Quit RadicalR (Quit: Nettalk6 -
12:32:50 Quit bmbl (Ping timeout: 268 seconds)
12:40:25 Quit TheSeven|Mobile (Ping timeout: 258 seconds)
12:50:48 Quit will__ (Quit: ChatZilla 0.9.86 [Firefox 3.0.8/2009033100])
12:54:50 Quit jae (Quit: leaving)
12:55:12 Join jae [0] (
12:57:08 Join Strife89 [0] (~michael@
13:10:02 Quit blairb (Quit: Leaving)
13:17:43***Saving seen data "./dancer.seen"
13:29:26 Join JohannesSM64 [0] (
13:36:53 Join notlistening [0] (
13:37:52notlisteningHi guys have there every been some work on AMLogic chips for rockbox
13:38:11Tornenot that i know of; what player?
13:39:20notlisteningintresting target
13:39:45Tornethe comment there say they used to make rockchip players
13:40:00Torneif it's a similar kind of thing then it's not gonna work
13:40:32notlisteningthey have changed to another type of chip but I just wanted to ask about it
13:40:58 Join jgarvey [0] (
13:41:59Tornewell, no
13:42:37Torneamlogic's page doesn't have anything interesting about what chips exist or what they do
13:43:37notlisteningyeah so thats probably a no go then :(
13:44:52 Join Luca_S [0] (
13:47:23 Join wodz [0] (
13:55:31notlisteningTorne, what is the important elements to getting Rockbox working on such a chip? Are they just dedicated media solutions?
13:57:08Torne"finding out what it is" is kinda the first step :)
14:00:30 Join adnyxo [0] (
14:00:59notlisteningAMLOGIC 8626 H
14:03:12Torneunfortunatley that's not much help, since that doesn't turn up any information about what the chip contains ;)
14:03:46pamaurythe screen is big but it's not a touch screen so it seems pretty akward to use with the buttons on the side
14:05:47notlisteningperfect for rockbox :D
14:09:39 Quit xavieran (Ping timeout: 268 seconds)
14:09:43notlisteningThere is less than anything on the internet out there on it
14:10:49Tornethis is the case with a lot of these random chinese chips
14:11:02TorneMost of them turn out to be very weak processors with DSPs attached to do codecs
14:11:08Tornewhich are almost always useless for rockbox
14:16:05notlisteninghas there been any development on i386 targets?
14:16:34Tornewhat hardware do you have with an x86? :)
14:17:20 Quit Strife89 (Quit: Class.)
14:17:35 Join Guest23487 [0] (
14:19:21pamauryan x86 is completely useless for a dap :)
14:20:28Torne(or anything else except "being able to run Windows", really) :)
14:21:18 Join xavieran [0] (
14:21:21*pamaury imagines a sansa e400, featuring an intel code 16 octo, with 100 DSPs attached, being able to stream hundreds of files at the same time to hundreds of different speakers :) (by wireless so that it can feat in 1cm)
14:22:21gevaertsTorne: I have a psion 3c somewhere. Does that count? :)
14:22:37 Quit Guest23487 (Ping timeout: 252 seconds)
14:22:46notlisteningits a running a but i can not find it online VIA C7-M 772 Esther
14:22:53Tornei think you are thinking of a different ancient crappy palmtop, the pre-ARM psions were NEC V30H
14:23:46TorneThat doesn't count as i386 :)
14:24:13gevaertsI know, but you said x86, not i386!
14:24:13 Join Farthen_ [0] (
14:24:29Tornex86 doesn't normall count older than 386 :)
14:24:41Torneno enhanced mode is a bit of a killer :)
14:25:04ZagorI have a nice HP 200LX too
14:25:08Zagorit can beep!
14:26:16TorneI think we'd have our work cut out trying to run on a <32-bit cpu
14:26:19notlisteningthere it is but i am struggling
14:26:24linuxstbnotlistening: If you can't find docs on a chip, you could look for a firmware update - that can often contain clues.
14:26:55Tornenotlistening: I am pretty sure nobody in their right minds would want to run rockbox as a firmware on a device with 1GB of ram which can boot win7.
14:27:58notlisteningmore so a platford for something like android and rockbox app
14:29:04Torneyah. if we get rockbox as an app there may be vlaue in someone doing x86-specific codec tweaks
14:29:11Tornebut other than that there's no x86 porting :)
14:33:05 Join evilnick_B [0] (~0c140464@rockbox/staff/evilnick)
14:34:10 Quit Farthen_ (Ping timeout: 252 seconds)
14:35:36 Join Farthen_ [0] (
14:39:34 Join Casainho [0] (~chatzilla@
14:43:31 Quit Farthen_ (Ping timeout: 252 seconds)
14:44:58 Join Schmogel [0] (
14:45:12 Join Farthen_ [0] (
14:48:59 Quit leavittx_ (Ping timeout: 258 seconds)
14:49:12 Quit leavittx (Ping timeout: 258 seconds)
14:50:55 Quit notlistening (Quit: Leaving)
15:00:47 Quit JdGordon (Ping timeout: 252 seconds)
15:02:11 Quit Casainho (Quit: ChatZilla 0.9.86 [Firefox 3.6.2/20100316060223])
15:03:56 Quit phanboy_iv (Ping timeout: 258 seconds)
15:04:46 Join phanboy4 [0] (
15:06:03beta_Kudos to the developers
15:06:14 Quit liar (Ping timeout: 245 seconds)
15:07:18beta_I just picked up a Gigabeat F10 for $40 at the pawn shop, no system files on it, five minutes later it was up and running Rockbox :)
15:07:48linuxstbbeta_: Congratulations! The gigabeat is a nice target.
15:08:15beta_I love the Gigabeat, this is my fourth :)
15:08:50beta_This will also be making the switch from F10 to 20 as soon as I can borrow my 1.8" drive enclosure back
15:09:48beta_But in comparson to doing my first gigabeat a few years ago to this one (which wouldn't boot at all with no system on it) it was damn ass easy :)
15:13:13 Join leavittx [0] (~leavittx@
15:13:46 Quit Farthen_ (Ping timeout: 252 seconds)
15:14:38 Join Farthen_ [0] (
15:15:28 Quit Farthen_ (Disconnected by services)
15:15:43 Join _Farthen_ [0] (
15:15:44 Nick _Farthen_ is now known as Farthen_ (
15:17:02 Quit Farthen_ (Disconnected by services)
15:17:20 Join _Farthen_ [0] (
15:17:22 Nick _Farthen_ is now known as Farthen_ (
15:17:46***Saving seen data "./dancer.seen"
15:18:02 Quit adnyxo (Ping timeout: 260 seconds)
15:19:37 Join Roronoa [0] (
15:23:40 Join TheSeven|Mobile [0] (~theseven@rockbox/developer/TheSeven)
15:26:08 Quit Roronoa (Quit: CGI:IRC)
15:26:14 Quit TheSeven|Mobile (Client Quit)
15:27:51 Join TheSeven|Mobile [0] (~theseven@rockbox/developer/TheSeven)
15:28:18 Part TheSeven|Mobile
15:28:19 Quit phanboy4 (Ping timeout: 245 seconds)
15:28:25 Join TheSeven|Mobile [0] (~theseven@rockbox/developer/TheSeven)
15:28:54 Part LinusN
15:31:47 Quit elinenbe (Read error: Connection reset by peer)
15:32:10 Join elinenbe [0] (
15:38:01 Join phanboy4 [0] (
15:47:31 Quit mitk (Quit: Leaving)
15:52:40 Join kaniini [0] (
15:59:58 Join Darkknight512 [0] (
16:00:40 Quit Darkknight512 (Client Quit)
16:00:56 Join Darkknight512 [0] (
16:02:26 Join phanboy_iv [0] (
16:02:38 Join RoronoaZoro [0] (~d2d43703@gateway/web/freenode/x-aepdsmyxmvzgqmbc)
16:03:01 Quit Darkknight512 (Client Quit)
16:03:44 Quit phanboy4 (Ping timeout: 245 seconds)
16:10:20 Nick Farthen_ is now known as Farthen (
16:12:53 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
16:14:29 Quit bzed (Quit: leaving)
16:14:37 Quit Farthen (Read error: Connection reset by peer)
16:14:39 Join bzed [0] (
16:15:22 Join Farthen [0] (
16:17:34 Quit mikroflops (Ping timeout: 240 seconds)
16:18:53 Nick fidencio[AWAY] is now known as fidencio (
16:21:17 Part xsteadfastx
16:23:03 Nick fidencio is now known as fidencio[AWAY] (
16:27:30 Quit kaniini (Ping timeout: 258 seconds)
16:27:50 Quit wodz (Quit: Leaving)
16:37:30 Join amr [0] (~quassel@
16:37:35 Join DataGhost [0] (
16:37:35 Quit DataGhost (Changing host)
16:37:35 Join DataGhost [0] (~dataghost@unaffiliated/dataghost)
16:39:31 Join TheSeven [0] (~theseven@rockbox/developer/TheSeven)
16:43:03 Nick fxb__ is now known as fxb (
16:44:14 Quit rvvs89 (Changing host)
16:44:14 Join rvvs89 [0] (~rvvs89@pdpc/supporter/base/rvvs89)
16:45:30 Join Tomis [0] (~Tomis@
16:47:10 Quit TheSeven|Mobile (Quit: TheSeven|Mobile)
16:51:43 Quit rvvs89 (Quit: leaving)
16:53:03 Join mikroflops [0] (
16:56:37 Join rvvs89 [0] (~rvvs89@pdpc/supporter/base/rvvs89)
16:58:23 Quit Tomis (Quit: Tomis)
17:05:32 Join stoffel [0] (
17:08:39 Join komputes [0] (~komputes@ubuntu/member/komputes)
17:11:46 Join planetbeing_ [0] (
17:16:28 Quit RoronoaZoro (Quit: Page closed)
17:17:47***Saving seen data "./dancer.seen"
17:24:47 Join mt [0] (~mtee@rockbox/developer/mt)
17:25:04 Quit shaggy-h (Ping timeout: 240 seconds)
17:25:33 Join planetbeing__ [0] (
17:28:52 Quit planetbeing_ (Ping timeout: 276 seconds)
17:31:44 Join retroj [0] (~retroj@pdpc/supporter/active/retroj)
17:39:36 Quit Farthen (Read error: Connection reset by peer)
17:40:07 Quit B4gder (Quit: It is time to say moo)
17:40:19 Join Farthen [0] (
17:45:42 Quit Farthen (Disconnected by services)
17:45:47 Join funman [0] (~fun@rockbox/developer/funman)
17:46:04 Join Farthen_ [0] (
17:46:04 Nick Farthen_ is now known as Farthen (
17:46:28 Join Blue_Dude [0] (
17:47:44 Quit planetbeing__ (Quit: planetbeing__)
17:47:59funmanFlynDice: i tested last revision bootloader and it works fine, i will make 1.0-RC bootloaders for clipv2 & clip+ so we can move them to unstable
17:48:34 Join FOAD [0] (
17:48:36FlynDicefunman: sounds great
17:49:02*pamaury starts messing up with kernel.c, definitely not good...
17:49:15funmani can't test clipv2 though, anyone?
17:49:43 Quit FOAD (Client Quit)
17:49:58 Join _flow_ [0] (
17:50:19 Join FOAD [0] (
17:50:45 Part retroj ("Channel buffer killed")
17:51:49 Quit mt (Ping timeout: 258 seconds)
17:53:30 Quit petur (Quit: *plop*)
17:54:14_flow_Could it be that the Cowon D2 SD Card bootloader support is broken? FS #10900 - I have the same issue with a 2GB SD Card. And yes, the ".rockbox" folder exists in the root of the SD Card... ;)
17:57:10Blue_DudeAnother hotkey patch is at FS #11081. The code is now completely transparent unless the HOTKEY define is explicitly stated in the target config file. And hotkey targets now have a lower binsize increase. The e200 is about 900 bytes.
17:59:38 Join mt [0] (~mtee@rockbox/developer/mt)
18:00:31 Join Peak [0] (~Harmeet@
18:00:50funmanFlynDice: in r25316, HS timings are applied unconditionally ?
18:00:52 Quit Farthen (Ping timeout: 246 seconds)
18:01:22funmani get ATA error -10 with FS #11140
18:01:40 Join Farthen [0] (
18:04:32funmaneven if i put the code around if(sd_v2) {} (no µSD plugged in)
18:05:02 Join zOxta [0] (
18:06:19 Join liar [0] (
18:07:38 Quit zOxta (Client Quit)
18:08:03 Quit pamaury (Quit: Quitte)
18:08:28 Quit Blue_Dude (Quit: ChatZilla 0.9.86 [Firefox 3.6.2/20100316074819])
18:09:05 Join Harmeet_ [0] (~Harmeet@
18:09:29FlynDicefunman: Yes I've been exploring that same problem w/ 11140 this morning. The non HS cards should just ignore that command though. Look at the transfer speeds on the disk debug page and you see 25MBits/sec and 50 if it does make the switch. I'm heading out the door right now back later
18:09:54 Part Harmeet_
18:10:15 Join Tomis [0] (~Tomis@
18:11:11funmanFlynDice: right it's 25mbps for internal and 50 for µSD
18:12:04 Quit Peak (Quit: Leaving)
18:15:31 Quit Farthen (Disconnected by services)
18:15:53 Join Farthen_ [0] (
18:15:55 Nick Farthen_ is now known as Farthen (
18:18:30 Join toffe82 [0] (~chatzilla@
18:19:55 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
18:23:26funmanZagor: can i send binaries to you for putting them on the server ? (Bagder looks busy)
18:28:04 Quit feisar_ (Ping timeout: 258 seconds)
18:29:14 Quit phanboy_iv (Read error: Connection reset by peer)
18:32:07 Join CGL [0] (~CGL@
18:34:12mtI have written my first draft of a proposal for GSoC .. Anyone willing to checkout and discuss ?
18:38:58gevaertsmt: sure
18:38:58 Quit MagusG (Read error: Connection reset by peer)
18:39:02 Join MagusG [0] (
18:39:07 Quit MagusG (Client Quit)
18:41:54mtgevaerts: - pass : 3mT33
18:43:14 Join pamaury [0] (~pamaury@rockbox/developer/pamaury)
18:44:23domonokywow encrypted pastebin, with a publically posted password :-)
18:45:19mtdomonoky: I just didn't want it publicly displayed on pastebin :) .. anyone from rockbox is more than welcomed to look. ;)
18:48:55gevaertsmt: looks good
18:50:26mtgevaerts: No but's then .. ? :) (Thanks for taking the time)
18:50:53 Quit swilde (Quit: ERC Version 5.3 (IRC client for Emacs))
18:51:03gevaertsmt: maybe some small language things, but I'm not awake enough for those right now
18:51:26*domonoky thinks this proposal is good. (a few typing errors, etc but those dont matter much).
18:51:29ranmafunman: BTW looking at debug-as3525.c I noticed that I'm having a problem with button_get on my C200v2
18:52:06ranmaI never noticed that there's more that one screen in HW Info because it has trouble recognizing the button presses.
18:52:13mtgevaerts, domonoky : Thanks, I'll look into the typos, just wanted to know the general opinion on the plan.
18:52:44mt(and domonoky, if you're awake enough, I'd be grateful if you could point out some of the typos :) )
18:53:13gevaertsmt: you know a lot more about codecs than me, so I'll believe you for the exact planning :)
18:53:55mt:) Alright then.
18:54:20ranmaI think there is some problem with doing a button_get directly after lcd_update, but I'll have to look into that in more detail tomorrow or so.
18:54:49 Quit Luca_S (Quit: CGI:IRC)
18:55:38domonokymt: for example the WMA voice section misses "voice" in the text. dont know about other typos :-)
18:55:44 Join Kitar|st [0] (
18:56:52mtdomonoky: So you also have no problem with the time plan ? (I'm revising now for the spelling mistakes)
18:58:25 Join MagusG [0] (
18:58:28domonokymt; i also dont know much about our codecs, but if we go from your last years performance, they look realistic.
18:58:54 Quit Kitr88 (Ping timeout: 276 seconds)
18:59:38domonokymaybe extracting the codec from ffmpeg might need more work, and the rockbox codec might need less time. But as you already have done similar work, you should know it best :-)
19:00:45mtdomonoky: Great. Thanks. :)
19:06:26CIA-5New commit by zagor (r25317): Added clip v2 & clip+ to unusable.
19:10:27 Nick YPSY is now known as Ypsy (
19:12:04mtZagor: not unstable ? ( Didn't follow much the few past days, but I thought it was agreed that they were considered unstable now ?)
19:12:15 Part RoBzz
19:12:38mtZagor: And the twitter page is nice. :)
19:12:41 Join scorche [0] (~scorche@rockbox/administrator/scorche)
19:14:29CIA-5New commit by zagor (r25318): Moved clip v2 & clip+ to unstable. :-)
19:14:48*Zagor speaks via commits
19:14:58funmanZagor: how are you ?
19:15:17*TheSeven boots up his build client to join with the second round :-D
19:15:21*funman looks for a "i'm fine" commit
19:15:44 Join feisar [0] (
19:16:12 Nick feisar is now known as Guest3190 (
19:16:15 Join kugel [0] (~kugel@rockbox/developer/kugel)
19:16:40TheSevenZagor: it's dead!
19:16:45TheSeventhe build system is dead!
19:16:53TheSevensince you added that tweeter :-P
19:17:18Zagorno, it doesn't rebuild for www commits
19:17:48***Saving seen data "./dancer.seen"
19:18:53 Join Horscht [0] (~Horscht2@xbmc/user/horscht)
19:19:41*scorche reminds those who havent edited the chart but plan on coming to DevCon to fill out:
19:20:07 Quit CGL (Quit: Saliendo)
19:21:33mtZagor: well said. :) (The commit I mean ..)
19:22:45 Join stripwax [0] (
19:22:57funmanZagor: just sent you clip+ bootloader
19:23:25Zagorfunman: excellent. I will upload it soon.
19:23:42funmani'll update the instructions on the wiki
19:23:52funmanwith a list of not working features
19:25:05topiknice milestone to get moved up to unstable
19:26:00 Join captainwekl [0] (~2669ecc2@gateway/web/freenode/x-mknznijnauyiaggf)
19:29:49 Join robin0800 [0] (
19:30:10bluebroth3rZagor: how about using the icons used by e.g. rbutil for the twitter account as avatar? Cropping the logo looks weird.
19:33:12gevaertsZagor: the stable release download page still lists the e200r
19:43:40 Quit m3dlg (Ping timeout: 248 seconds)
19:44:15 Quit flydutch (Quit: /* empty */)
19:45:10CIA-5New commit by bluebrother (r25319): Remove unnecessary class declaration, this should come from the includes.
19:47:38 Nick fxb is now known as fxb__ (
19:48:18 Join Luca_S [0] (
19:48:22kugelfunman: I sent the fuzev2 yesterday morning, it should arrive by tomorrow or friday (they said)
19:48:35funmanthanks, i'll tell you when it's there
19:50:44 Join b0hoon [0] (
19:54:24 Quit captainwekl (Quit: Page closed)
19:55:21 Quit jae (Quit: leaving)
19:59:46 Quit mt (Read error: Connection reset by peer)
20:00:49 Join Stephen__ [0] (~S@
20:01:21 Join jae [0] (
20:05:29 Join kugel_ [0] (
20:05:42 Join planetbeing_ [0] (
20:05:45 Quit kugel (Disconnected by services)
20:05:49 Nick kugel_ is now known as kugel (
20:05:53 Quit kugel (Changing host)
20:05:53 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:07:25 Join anewuser [0] (anewuser@unaffiliated/anewuser)
20:08:33 Quit stripwax (Quit:
20:14:10Stephen__for the clip+ was does the internal and external storage being read only mean ?
20:14:49 Quit Barahir (Ping timeout: 260 seconds)
20:15:37Schmogelyou cant save your settings or run any programm that needs to write something to the storage
20:15:59Stephen__ah thank you. now to just wait for the bootloader to hit the server
20:16:16Stephen__will that include resuming playback ?
20:16:20 Join adnyxo [0] (
20:16:28 Join Barahir [0] (
20:16:50funmanyes, file and position are saved to disk afaiu
20:17:07Stephen__ah right ah not to worry small price to pay for having rockbox on this
20:21:06funmanStephen__: bootloader is on (same bootloader to be put on the server)
20:21:40Stephen__thanks funman and great work!
20:25:20 Quit leavittx (Read error: Operation timed out)
20:26:14funmanTorne: ping
20:26:46 Join leavittx [0] (~leavittx@
20:26:51 Quit amr (Ping timeout: 248 seconds)
20:33:36 Quit Luca_S (Quit: CGI:IRC (EOF))
20:35:30 Quit funman (Quit: free(random());)
20:36:14 Quit DerPapst (Quit: Leaving.)
20:36:56 Quit BlakeJohnson86 (Read error: Operation timed out)
20:38:10 Join BlakeJohnson86 [0] (~bjohnson@2002:1876:a27b:0:227:13ff:fe65:1262)
20:46:16 Nick fidencio[AWAY] is now known as fidencio (
20:48:01 Nick fidencio is now known as fidencio[AWAY] (
20:48:07 Join Strife89 [0] (
20:51:10 Join m3dlg [0] (~m3dlg@
20:57:06 Join notlistening [0] (
20:57:24notlisteningAny clip v2 and clip + owners here?
20:57:48notlisteningare they good and which one is better if any?
20:57:50 Join petur [0] (~peter@rockbox/developer/petur)
20:57:51Stephen__i have aclip+
20:57:59ThomasAHnotlistening: only "or": clip+
20:59:31 Quit stoffel (Read error: Connection reset by peer)
20:59:42 Join wodz [0] (
21:00:33topikisn't it the same hardware, except the + has a microSD slot?
21:01:12topikso if you dislike microSD, the clip v2 is better
21:03:15notlisteningAh thanks topik umm they are limited without the SD i think so + it is
21:03:44topikthere are some other hardware differences, but i don't think the matter as much as the microSD slot
21:04:29notlisteningThe SD is the deal breaker for me
21:08:23 Join DerPapst [0] (
21:08:26Stephen__having just rockboxed my clip+ its so much better! now just to wait for write support :o)
21:09:16Stephen__8gb internal and 16gb sd card sweet!
21:09:35ThomasAHStephen__: and no longer waiting for "Refreshing your media" :)
21:10:08Stephen__will still need the of to do file transers :o( and my work only allow mtp so will need to use that. not as much tho
21:10:27notlisteninghas there been any progress on USB support on these devices?
21:10:34Zagorsandisk showed 32gb cards on a trade show recently
21:10:53notlisteningyeah just looking at prices for them 60
21:11:07 Quit m3dlg (Ping timeout: 248 seconds)
21:13:56Zagorwho can whip up an svg image of the clip+?
21:14:28Stephen__isn't there a patch on flyspry of one ?
21:14:51wodzfunman: ping
21:14:53Zagoris it?
21:14:53 Join mt [0] (~mtee@rockbox/developer/mt)
21:14:59Stephen__one sec
21:15:14Zagorfs #11048
21:15:33Stephen__Knew I'd seen one
21:16:30notlisteningStephen what battery performance you getting?
21:16:36notlisteningor times
21:16:42topik32GB uSD sets you back 199 USD.
21:16:55 Quit pixelma (Disconnected by services)
21:16:56 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
21:17:00 Quit amiconn (Disconnected by services)
21:17:02 Quit kugel (Ping timeout: 245 seconds)
21:17:03 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
21:17:09Stephen__no idea yet just rockboxed a while ago should have better figures next week after some extensive use
21:17:15 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
21:17:25 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
21:17:40CIA-5New commit by zagor (r25320): Added link to twitter commit feed.
21:17:51***Saving seen data "./dancer.seen"
21:22:34 Join m3dlg [0] (~m3dlg@
21:22:58CIA-5New commit by zagor (r25321): Added Sansa Clip+ image by Marko Pahlke (FS #11048)
21:27:38 Join _deV_nuLL_ [0] (
21:28:12CIA-5New commit by zagor (r25322): Added Sansa Clip+ and moved Clip v2 to status 2 (unstable).
21:30:55 Join funman [0] (~fun@rockbox/developer/funman)
21:31:05funmanwodz: pong
21:31:48wodzfunman: I am the author of the #11137 and You are the only reviewer
21:31:59wodzI would ask few questions than
21:32:48funmansure but i know nothing of coldfire so others might know just as much as i for your questions
21:33:38wodzusing HAVE_WM8751 instead of HAVE_WM8750 is no particular problem but how should I extend definitions than?
21:33:44funmannotlistening, topik: the clipv2 has an hold switch, unlike the clip+
21:34:23wodzfunman: add target specyfic ifdefs?
21:34:54funmanwodz: hm i suggested reusing wm8750.* files but i have no idea how much they will differ when 8751 support is added
21:35:36funmanerr, revert 8750 and 8751 in my last sentence
21:35:43 Quit m3dlg (Read error: Connection reset by peer)
21:36:19topikfair enough funman :). i'll pick the uSD over the hold switch though
21:37:09 Join m3dlg [0] (~m3dlg@
21:37:13wodz8751 do not have input ADC and so it does not support recording
21:38:13funmanif the bits do not differ you could add #ifdef HAVE_WM8750 ... input adc ... recording ... #endif inside wm8751.c ?
21:38:47wodzfunman: aaa that way
21:39:13funmanwe can't rename wm8751.c to wm875x.c though because there is also a wm8758.c
21:39:28wodzI have to check register addresses side by side
21:39:56funmanwodz: do you think your patch should be committed now, or wait for playback to work?
21:40:27wodzfunman: At least wait for finished bootloader
21:42:16 Quit m3dlg (Quit: RAGE QUIT)
21:42:20wodzam I correct that there is no official datasheet for wm8758?
21:43:15pixelmafunman, topik: with software hold, a missing hold switch shouldn't be a big thing, no?
21:43:43 Quit mt (Ping timeout: 258 seconds)
21:43:50topikit's pressing the home button longer, isn't it?
21:43:52funmanpixelma: software hold only works in the wps, so when playlist is finished buttons are unlocked
21:44:02funmantopik: in rockbox it's home+center in wps
21:44:28topiktwo buttons are at the limit of my skill level!
21:44:33FlynDicenotlistening: re clip+ battery performance, I listened to mine for ~ 7 hours on Sunday and still had some power left. I find that rather amazing as we still have some clocking issues to work out and understand!
21:44:38funmanlong home is already used to exit various screens, so i picked up one at random
21:44:58funmanbtw clip+ doesn't seem to charge very fast
21:45:03pixelmaI know, as the Archoses have software hold too and I have an Ondio. I like it that way btw.
21:45:17topikjust not having it turn on accidentally would be enough
21:46:29 Join RoronoaZoro [0] (~vijayss@
21:46:36pixelmathe on/off button is very unlikely to press on the Ondio, not a big problem there
21:47:23topiksame with the 2-button combo funman picked
21:47:33 Quit komputes (Ping timeout: 276 seconds)
21:48:23 Join blairb [0] (
21:48:40Stephen__funman, is charging via rockbox enabled yet. same way as the 200v2 via a button hold on usb insert ?
21:48:57funmanStephen__: nope just plug USB, rockbox doesn't reboot
21:49:13Stephen__it'll charge then ?
21:49:21funmanyes, although slowly it seems
21:49:25Stephen__ok cool.
21:50:31*Stephen__ looks at the nice shiney new clip in the current build page
21:51:23 Join m3dlg [0] (~m3dlg@
21:51:44pixelmatopik: I meant accidentally turning on (which would not be protected with software hold. The button combo to enable hold when it's turned on and in the WPS, is a two button combo on the Ondio as well
21:52:06 Quit JohannesSM64 (Quit: WeeChat 0.3.2-dev)
21:52:39topikpixelma: it doesn't take a 2-key combo to unlock?
21:53:26pixelmait does (the same), so I should have said "to toggle hold"
21:53:36notlisteningcan you post the link to the picture plz?
21:54:11topikthat makes a lot of sense
21:54:46funmannotlistening: (the pic is without the arrows and smaller)
21:56:07 Join komputes [0] (~komputes@ubuntu/member/komputes)
21:57:15topiki'll pick me up a 4GB clip+ tomorrow. they're on sale this week it seems
21:59:14topikthey use a picture with a hold button in the advert though, so i better make sure it's the +
21:59:25 Quit Farthen (Quit: ChatZilla 0.9.86 [Firefox 3.6.2/20100316074819])
21:59:27 Quit m3dlg (Max SendQ exceeded)
21:59:58CIA-5New commit by gevaerts (r25323): Recalibrate build scores
22:00:12 Join Farthen_ [0] (
22:01:21 Nick fidencio[AWAY] is now known as fidencio (
22:02:24 Quit hebz0rl (Quit: Ex-Chat)
22:02:50 Join m3dlg [0] (~m3dlg@
22:03:02 Quit robin0800 (Remote host closed the connection)
22:03:30 Join mt [0] (~mtee@rockbox/developer/mt)
22:03:31 Quit Farthen_ (Client Quit)
22:06:09 Nick Ypsy is now known as YPSY (
22:07:12 Quit anewuser (Ping timeout: 246 seconds)
22:07:44 Quit kunal (Quit: Ex-Chat)
22:08:08domonokydoes somebody know the usb-id of the clip+ ?
22:08:39funmandomonoky: the one in config/sansaclipplus.h should be correct for msc
22:10:58domonokyoki, does somebody know the mtp usb-id ? (working on rbutil support)
22:11:38domonokyis the clipv2 also unstable ? i dont see a bootloader for that on
22:12:33funmanyes but the bootloader hasn't been tested yet, and clip+ bootloader hasn't been uploaded to server yet : I have written the links on SansaAMS based on the links for other models
22:13:07funmanthe bootloader "i want to release as 1.0-RC"
22:15:07domonokythe clipplus bootloader is already on haxx, just not on all mirrors :-)
22:15:30 Quit n17ikh (Read error: Connection reset by peer)
22:15:52 Join Ozzo [0] (
22:16:43Ozzoheyy ppl i m new here.. using the clipv2 and trying to update your bootloader however the clip is only deleting the firmware file
22:17:12Ozzono feedback from it whatsoever
22:17:40funmanOzzo: did you build the clipv2 bootloader?
22:17:59Ozzoyep and renamed it to m30pa.bin
22:18:08Ozzo..and put it in the root
22:18:21funmanthen you didn't read the instructions, you need to patch the original firmware with this bootloader
22:18:35Ozzoyes yes I used mkamsboot
22:18:51OzzoI already did this once successfully
22:19:00Ozzoit just doesnt want to happen again
22:19:32funmandid you follow the steps from ?
22:19:34CIA-5New commit by Domonoky (r25324): add clipv2 and clip+ to rbutil.
22:19:57Ozzofirst was r25316 that worked
22:20:13Ozzonow i tried updating the bootloader to the latest build
22:20:23Ozzoand the clip only deletes the file
22:20:34funmanyou don't need to update the bootloader once it works
22:20:43Ozzooh lol
22:20:54Ozzoic then hehe
22:20:56funmandoes it turn off / display the "update in progress" message ?
22:21:05funmanand do you eject the clipv2 properly?
22:21:09Ozzono thats what i was wondering
22:21:18Ozzoyes yes
22:21:41 Quit S_a_i_n_t (Read error: No route to host)
22:21:55funmantry updating with a non patched OF and if it doesn't work i'm clueless
22:22:19Ozzoi maybe thought it has sthing to do with the new sansa firmware
22:22:29 Join S_a_i_n_t [0] (S_a_i_n_t@
22:22:37Ozzomaybe the clip is able to notice same version or sthin
22:23:14Ozzotrying with original firmware give me sec
22:23:28 Join n17ikh [0] (
22:23:37 Join RadicalR [0] (
22:24:26Ozzonope same thing boots directly to rockbox
22:25:00b0hoondomonoky: could you look at it please?
22:25:18funmanOzzo: download the OF again, i bet you downloaded the clipv1 OF
22:25:24 Join panni_ [0] (
22:25:30domonokyb0hoon: looks good
22:25:48Ozzono i kept using the same original OF file all throughout
22:26:08Ozzobut i ll give it a try
22:26:34b0hoondomonoky: great, should i commit it?
22:26:54domonokyb0hoon: yes.
22:27:15 Quit S_a_i_n_t (Ping timeout: 264 seconds)
22:27:31Ozzofunman: should I boot to the OF after transferring the firmware file ?
22:27:54funmanOzzo: doesn't matter
22:28:45OzzoO.o worked this time
22:28:54Ozzolet me try and patch this one
22:30:03 Quit _deV_nuLL_ (Quit: YoP !)
22:30:22funmanOzzo: would you try a bootloader i give you ? i want to release a binary bootloader for clipv2 but have no device to test it
22:30:27CIA-5New commit by b0hoon (r25325): add Packard Bell Vibe 500 to RBUtil.
22:30:49Ozzofrom 1 to 10 what ś the risk of briking it ? xD
22:31:08 Quit planetbeing_ (Quit: planetbeing_)
22:31:23Ozzoif it s just like a new release or sthin i hv no trouble
22:31:25 Join S_a_i_n_t [0] (S_a_i_n_t@
22:31:32funmanas usual
22:31:39*gevaerts tries to parse that
22:31:43Ozzowell ok
22:32:03Ozzosome release notes maybe ? ;P
22:32:11funmani would give one finger to cut on this, but not the whole hand
22:32:38Ozzono i meant like ..what should be better ?
22:33:02 Quit jd (Quit: Ω)
22:33:08funmannothing better, it just changes the version to "1.0-RC"
22:33:18Ozzooh ok
22:34:18 Join Farthen_ [0] (
22:34:36Zagorfunman: should I put that on download?
22:34:42 Join planetbeing_ [0] (~planetbei@
22:35:03funmanZagor: yes please
22:35:17funmanOzzo: thanks
22:35:36ender`clip v2 is becoming usable?
22:35:45*domonoky added the new clips to the themepage.. :-)
22:35:58funmanZagor: where did you put it ? is 404
22:36:00*Stephen__ starts themeing
22:36:14gevaertsfunman: mirrors aren't updated instantly
22:36:31 Join Kuitsi- [0] (
22:36:37Zagoralso, I put it in the wrong dir... fixing
22:36:40funmanhm ok, i get 404 on both servers though
22:36:52funmanah clipplus is there as well
22:37:02Zagoryes, but also in the wrong dir
22:37:11Ozzodamn.. this thing is doing the same thing again
22:37:19Ozzotry 2 on the way
22:37:25Zagormoved now
22:38:12Stephen__domonoky, clipline from the themes seems to be missing for the clipv2 and +
22:38:20Stephen__but its there for the v1 ?
22:38:27Stephen__just for consistency ;)
22:38:49 Join jd [0] (
22:38:50 Quit jd (Changing host)
22:38:50 Join jd [0] (~jd@Wikipedia/HellDragon)
22:39:00Ozzofunman tryin your bootloader
22:39:09Ozzoit works
22:39:12CIA-5New commit by Domonoky (r25326): adjust the bootloader paths to the changed locations.
22:40:25Ozzostill can t understand my firmware file problems but i can put up with it
22:40:48domonokyStephen__: its there for all of them. (probably checkwps was still running when you looked).
22:40:58b0hoondomonoky: i suppose that's no all, now i must equipt the with the compiled bootloader and other files...?
22:41:01 Quit planetbeing_ (Quit: Poof.)
22:41:06Ozzoso when do I really have to update the bootloader ?
22:41:32Stephen__domonoky, ah yes. i'm a bit too quick off the mark it would seem ;)
22:41:35 Part Kuitsi-
22:41:35domonokyb0hoon: yes, let Bagder, or Zagor move a bootloader file from you to the download server :-)
22:41:38funmanOzzo: never
22:42:24Ozzoic.. so how are your improvements ,like this one you sent me, applied ?
22:42:37 Join planetbeing_ [0] (~planetbei@
22:43:01gevaertsActually, the correct answer is 'almost never'
22:43:03funmanupdate the build (.rockbox folder), bootloader rarely if ever needs updating
22:43:22Ozzooh I get it
22:43:24 Quit planetbeing_ (Client Quit)
22:43:51Ozzocool hehe that just about solved all my questions regarding rockbox
22:44:27Ozzo..i m lovin it even though it s only my 1st day
22:46:03b0hoondomonoky: ok, thanks :-)
22:46:33Stephen__Ozzo, wait till it's even more complete you'll be hooked!
22:46:48Ozzoye that s what i m thinkin
22:47:00Ozzoeh btw i forgot to ask bout this
22:47:03Stephen__any mp3 player you'll be you'll want supported
22:47:06Ozzodatabase isn t working
22:47:20Ozzois it cause of the read only problem ?
22:47:20 Nick fidencio is now known as fidencio[AWAY] (
22:47:23Stephen__that has to write files
22:47:27 Join planetbeing_ [0] (~planetbei@
22:47:37Ozzooh k guessed so
22:47:41ender`heh, i've been waiting for this day :)
22:48:05*Bagder stands up straight
22:48:08Zagorb0hoon: since I've just uploaded for funman, I can take yours too
22:48:08Ozzoso Load to Ram in the settings does not apply to it s creation right ?
22:48:11 Join CGL [0] (~CGL@
22:48:20*Bagder hides again
22:49:01Ozzolol Create to Ram would be cool at this point
22:49:15AlexPand pointless
22:49:41Ozzoyes but still it s a feasable workaround
22:49:46Ozzokind of
22:50:00Ozzo..don t really know what it involves
22:50:12b0hoonZagor:ok, thanks. so must i give it to you or you have a compiled version already for example from automatic build?
22:50:19AlexPOzzo: Isn;t going to happen
22:50:30gevaertsIt's probably possible to build the database using the host-based tool and to enable it by hand in the config file
22:50:32Ozzolol kk
22:50:34domonokyOzzo: that work would be better spent on making write work :-)
22:50:35Zagorb0hoon: I'd like a tested binary from you
22:51:04Ozzoyes I understand
22:51:04AlexPgevaerts: Yeah, good thinking
22:51:30Zagorthe build farm doesn't upload the bootloaders
22:51:45Ozzohost-based tool?
22:51:46 Nick Farthen_ is now known as Farthen (
22:51:52Ozzois that in the sources ?
22:51:58AlexPAssuming this for a release it needs the release version set too
22:53:50b0hoonZagor: will it be ok to send it on your mail?
22:54:00Zagorb0hoon: sure
22:54:42Ozzocould somebody pleeaasse kindly explain (just an overview) how one can do this database creation (gevaerts)
22:54:58gevaertsI've never used the tool
22:55:21Ozzohmm well time to experiment then
22:55:28Stephen__hmmm the theme site is showing the wrong image for the sansa clip+
22:57:15 Quit pamaury (Ping timeout: 264 seconds)
22:57:48b0hoonZagor: sent :-)
22:58:50 Nick Farthen is now known as Farthen_ (
22:59:06domonokyStephen__: it does not, look again :-) (thank you for noticing)
22:59:34Stephen__good work domonoky ;o)
23:00:21 Nick Farthen_ is now known as Farthen (
23:03:30 Join bieber [0] (
23:03:49 Quit planetbeing_ (Ping timeout: 258 seconds)
23:05:20 Join flyback [0] (
23:10:21 Quit bieber (Ping timeout: 245 seconds)
23:12:39 Join kugel [0] (~kugel@rockbox/developer/kugel)
23:14:30Zagorshould the vibe500 bootloader be called jukebox.mi4 or something else?
23:15:19b0hoonZagor: yes, jukebox.mi4
23:15:57domonokyZagor: rbutil expect it now as: /pbell/jukebox.mi4
23:16:08Zagorpbell? the target name is vibe500
23:16:39domonokybut we can change rbutil of course :-)
23:16:52 Join planetbeing_ [0] (~planetbei@
23:17:53***Saving seen data "./dancer.seen"
23:17:57b0hoonZagor: the folder should be pbell on the server if i understand correctly how it works :S
23:18:51kugelisn't it pbellvibe500?
23:18:58Zagorkugel: no
23:19:04Zagorlook in tools/configure
23:19:18Zagoror tools/
23:19:31kugelright, nevermind
23:20:07 Quit funman (Quit: free(random());)
23:23:39 Join anewuser [0] (anewuser@unaffiliated/anewuser)
23:24:33 Quit Ozzo (Quit: CGI:IRC (EOF))
23:24:39 Quit CGL (Quit: Saliendo)
23:33:17 Join Luca_S [0] (
23:34:49 Quit killan (Quit: ( :: NoNameScript 4.22 :: ))
23:35:27 Quit jd (Quit: Ω)
23:36:31 Join matsl [0] (
23:38:00 Join killan [0] (
23:38:07 Join arbingordon [0] (
23:38:56 Join jd [0] (
23:38:56 Quit jd (Changing host)
23:38:56 Join jd [0] (~jd@Wikipedia/HellDragon)
23:44:09 Quit matsl (Remote host closed the connection)
23:47:09 Quit blairb (Quit: Leaving)
23:47:31 Join blairb [0] (
23:48:48 Quit planetbeing_ (Quit: Poof.)
23:49:39 Part blairb
23:52:51 Join blairb [0] (
23:53:31linuxstbNo-one tagged the released bootloaders for the clipv2 and clip+? And they were called 1.0-RC ?
23:53:50 Join ozzo [0] (
23:54:47ozzohmm using r25326 Clipv2 is playing nothing
23:55:24 Quit adnyxo (Ping timeout: 258 seconds)
23:56:09ozzoseems to be due to thefact that I generated a database on my pc
23:56:42ozzoafter using it everything gives out an Error Accessing Playlist Control File
23:56:45linuxstbHmm, and there's no manual for the clip+/clipv2 either?
23:57:12ozzohowever if I reboot and not use the Database everything is fine
23:57:22Zagordoes clipv2 need its' own manual?
23:57:38linuxstbYes, the installation instructions point to the wrong files.
23:58:11ozzono man I just did the whole thing today and the instructions were perfect
23:58:38Zagorozzo: did you read the wiki or the manual?
23:58:46kugeldo we need different manuals? or could we put the installation so that it works for all?
23:58:57Zagorozzo: right. we're discussing the manual.

Previous day | Next day