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 2006-12-04

00:00:06amiconnThat's just because the linker doesn't throw an error when the beginning of the last section still fits
00:00:23amiconn...even if the end is outside the specified area
00:00:40amiconn(a linker bug that has always been there)
00:00:45amiconnCheck the build log
00:00:50 Quit MarcoPolo (Remote closed the connection)
00:00:59jhMikeSthat's a pretty bad bug imo
00:01:25amiconnOutput is 308 bytes larger than max (233456)
00:01:25amiconnBuild firmware file
00:01:25amiconnremoving UCL file again, making it a fake one
00:01:49 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
00:02:35 Join daurn|laptop [0] (i=daurn@
00:02:38linuxstbSo are any rombox builds still working?
00:03:37amiconnIn the beginning of 2006, all archoses could run rombox. For FM and V2 it broke some time in summer, for the V1 it broke in autumn, and for Ondio FM it broke today :(
00:03:54amiconnThe only archoses still featuring rockbox now are the Ondio SP and the Player
00:04:25*jhMikeS thinks about unneeded strings again - but what a task - and adding the feature selectivity
00:04:58amiconnThe string cleanup should bring back rombox for a while
00:05:29jhMikeSand give it room to grow again, like repotting a plant :)
00:07:12jhMikeSthere's got to be some prioritizing of features with rombox no doubt
00:07:52amiconnIn fact, if rombox would break because of adding cool new features, and in spite of optimising the code, it would be another thing
00:07:57 Quit XavierGr ()
00:08:08 Quit ender` (" "I want an Internet. Can I have one of those?" -- Spice Girl Mel B., aka Scary Spice during an AOL press conference, pointi)
00:08:16amiconnWhat pisses me off is that many of the recent code size bumps are quite unnecessary
00:08:38 Join XavierGr [0] (
00:09:02linuxstbMaybe the code size of each build should be listed somewhere - so people can see the impact of their commits.
00:09:10 Join [Gino] [0] (
00:09:48Soapcode size primarely affects the archos line, correct?
00:10:18jhMikeSI'm wondering where I can trim fat in anything I've done that would affect that or anything else. :\
00:10:32amiconnIt does affect them in a more obvious way because of the hard limits regarding rombox etc,
00:10:42Soaphow does it affect the software codec ports? a reduction in buffer size?
00:10:47amiconnbut the target with the tightest RAM restrictions is the iFP
00:11:37jhMikeSamiconn: what other commits of recent are bloated is what I mean. I know about the latest.
00:11:49amiconnI think all targets benefit from compact & efficient code
00:12:27SoapI don't think anyone can disagree with that, in theory.
00:12:33amiconnThe buffer size is less important on targets with 16MB+ of RAM, but battery runtime also depends on code efficiency
00:13:19jhMikeSI guess for coldfire that means running the stop #2000 as much as possible?
00:14:53Soapso runtime detection of 2048 byte sectors was commited, but runtime detection of 64MB of memory still hasn't. What is holding that back?
00:15:18linuxstbThat means completely reorganising how Rockbox uses memory.
00:15:20amiconnThe detection isn't the problem. The RAM layout is
00:16:22amiconnjhMikeS: Ideally, threads should be sleeping as often & long as possible, and boosting shouldn't happen too often and only when the power is really needed
00:16:44Soapis the RAM layout discussed on the wiki, or do I need to read the code and plot memory usage on a piece of paper?
00:16:44 Join MarcoPolo [0] (
00:17:11amiconnSaving space is often a question of how the function is organised, and what optimisations are applied
00:17:15jhMikeSshouldn't sleep_thread be used in a lot of places instead of yield then?
00:18:00linuxstbSoap: It's currently something like: [Rockbox][ buffer...][codec buffer][plugin buffer] - with Rockbox being at the start of RAM and the codec/plugin buffers being at fixed (at compile-time) addresses at the end of RAM. The audio buffer is what's left in the middle.
00:18:09amiconnRemember that I added support for reading 16bit and 4bit BMPs to the bmp loader (those are the features which affect monochrome targets) with practically constant code size on archos?
00:18:31amiconn(+4 byte in total counting my second optimisation)
00:18:35 Quit BHSPitMonkey (Connection timed out)
00:18:39Soaplinuxstb: what does the 64MB patch in the tracker do? Stretch the audio buffer, or split it in two?
00:18:56amiconnSoap: Afaik there is no runtime detection patch
00:19:03linuxstbSoap: The suggestion is to change that to [codec buffer][plugin buffer][Rockbox][ buffer...] but no-one has stepped up to do it.
00:19:08amiconnThe 64MB patch allows building a special 64MB build
00:19:52SoapI grok that amiconn. I just didn't know if all it needed was runtime detection, or if people were (rightfully) upset with how it placed the memory usage.
00:22:26Soapnevermind, after reading linuxstb's first comment on the tracker it makes sense now.
00:24:55jhMikeSwe should be able to sleep threads only up till their next turn to run but that doesn't appear supported. there's always a timeout.
00:25:21amiconnsleep until their next turn doesn't make sense
00:25:29amiconnThat would be identical to yield()
00:25:34linuxstbSoap: Only 6 months old...
00:26:00Soaplinuxstb: I agree, I asked a silly question in hindsight.
00:26:14amiconnWith sleep(0) you can sleep the thread until the end of the current tick
00:26:19linuxstbSoap: No, I meant that after 6 months, no-one has done any more work on it...
00:26:40 Join TeaSea [0] (n=thunderc@
00:28:35 Quit TeaSea (Read error: 104 (Connection reset by peer))
00:29:01jhMikeSyield doesn't seem to put them on the list of inactive processes and execute a stop (or eq) though
00:29:17amiconnOf course not
00:29:48amiconnThe scheduler only executes (and can only execute) stop if all threads are sleeping
00:30:06jhMikeSright, but why have to wait a min of .01s ??
00:30:18amiconnBut if a thread already wants to run after one scheduling round, that cannot happen
00:30:54 Join TeaSea [0] (n=thunderc@
00:31:17amiconnStop halts the cpu until the next interrupt. The only interupt that is guaranteed to happen in regular intervals is the tick.
00:31:43jhMikeSah so then I guess you have to expect to be out at least 1 tick
00:31:59amiconnNot _at least_, but in the worst case
00:32:12 Join [sellout] [0] (
00:32:22amiconnAs already mentioned, you can do sleep(0); which will sleep th ethread until th eend of the current tick
00:32:43amiconn..which can be anything between nearly immediately and 10ms
00:32:47jhMikeSI mean, you can't count on being running any sooner
00:33:58 Join BHSPitMonkey [0] (
00:34:05amiconnWith th eold scheduler you could do some tricks, dunno whether that's still possible
00:34:36jhMikeS10ms? could be longer if an lcd update takes place but then so can yield
00:35:06jhMikeSthat was a big issue with doing the sound effects
00:35:51amiconnYes, any thread might delay stuff further
00:36:05jhMikeSbut they're boosted and we're cutting it _real_ close for latency
00:36:12amiconnReally time critical stuff requires an isr
00:39:10 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
00:39:16 Join daurn|laptop [0] (i=daurn@
00:39:19 Join TeaSeaLancs [0] (n=thunderc@
00:42:16jhMikeSthat could be a way to go with those. could an interrupt be triggered from the recording handler?
00:43:04amiconnI don't know what you need
00:43:22amiconnIn general it is possible to use interrupts from within plugins
00:43:53amiconnPreferably the plugin shouldn't bang the hardware, but rather use functions provided by the core
00:44:20amiconnThe grayscale library, video.rock (archos) and metronome.rock use the user timer this way
00:44:22jhMikeSto process the sound as soon as another chunk is available from DMA and not have to wait through a screen update
00:44:28 Join MadDog011 [0] (
00:44:33 Quit MadDog011 (Remote closed the connection)
00:44:48amiconnThen you probably need to add the ability to register a callback in the recording isr
00:45:34jhMikeSthere's already the DMA1 registered callback but audio can't be processed there
00:45:44amiconnWhy not?
00:46:19jhMikeSArent there only 6 samples available in the FIFO? I did try it and it just breaks up.
00:46:44amiconnThe DMA interupt is fired at the end of the DMA transfer
00:47:00amiconn...which is way longer than 6 samples (at least I hope it is)
00:47:09jhMikeSyes. but then you can only process a few samples and have to return
00:47:52jhMikeSProcessing audio can take longer than the six samples it would seem
00:47:58amiconnYou need to set up the new dma first in order to prevent fifo overrun. After that you have quite some time for processing...
00:49:04 Part nave7693
00:49:33jhMikeSit makes no provision for that. it's just like the playback. would need a function for that
00:50:06jhMikeSactually would be pretty simple to set up and no plugin hw banging
00:53:20 Quit Rudy4Pez (Read error: 110 (Connection timed out))
00:53:46amiconnSome plugins even bang the hardware, but this is not encouraged
00:54:25amiconnalpine_cdc is one, and wavplay is another. They hijack interupt vectors.
00:55:08jhMikeSI'm thinking about what to do if the next interrupt comes and it's not done. probably spin lock until the previous one is finished
00:55:15amiconnBut then, wavplay is a proof-of-concept preview, and alpine_cdc is a quite specialised plugin
00:55:51amiconnjhMikeS: Shouldn't it be locked automatically, via the interupt priority thing?
00:56:11amiconn(unless the recording interrupt is non-maskable, but why should it?)
00:56:21jhMikeSif it is, then how will the transfer start from the handler?
00:57:57 Quit TeaSea (Read error: 110 (Connection timed out))
00:58:26amiconnThe handler can't be called a second time as long as it's executing
00:58:27jhMikeSIf we've started another transfer, it will start while still in the handler, and the next interrupt from that transfer will be blocked? ok, then no worries
00:59:08amiconnInterrupts (at this and lower level) are reenabled by the rte instruction
01:00:04amiconn...unless it's level 7, then it's non-maskable
01:00:26amiconnBut imho we shouldn't use level 7 for anything without strong reason
01:00:34jhMikeSwith that, latency can be very low indeed
01:01:33amiconnSlightly more than one DMA block I'd think
01:02:18jhMikeSthey're small blocks (128 samples)
01:02:33jhMikeSno real attempt at optimizing those things yet
01:03:08 Join TeaSea [0] (n=thunderc@
01:03:16amiconnLatency might go up to nearly 2* dma_block_size. Depends on how long the calulations take
01:04:37jhMikeSthat's ok, it's 5 now, most of it to keep playback clear of the current effect block
01:05:09***Saving seen data "./dancer.seen"
01:06:37 Quit daurn|laptop (Read error: 131 (Connection reset by peer))
01:06:40 Join dau [0] (i=daurn@
01:10:51*amiconn needs to set up a testing environment for large virtual sectors :/
01:12:04 Join ani-adom [0] (
01:12:40 Quit dau (Read error: 131 (Connection reset by peer))
01:13:02 Join daurn|laptop [0] (i=daurn@
01:13:14ani-adomis it possible to know id3 of the next song?
01:13:16jhMikeSdoing things that way might even be good for playback on a mixing board style pcm buffer without another thread if the final mix point is powered by the interrupt
01:17:43*MarcoPolo is away: ZZZzzz
01:17:45 Nick MarcoPolo is now known as MarcoPolo|AFK (
01:18:39 Quit TeaSeaLancs (Read error: 110 (Connection timed out))
01:20:49 Quit gelatinous ("CGI:IRC")
01:22:24linuxstbani-adom: Yes, a WPS can display tags from the next track.
01:23:54pixelmasee for more info ...
01:25:34 Quit Criamos ("( :: NoNameScript 4.03 :: )")
01:26:27ani-adomi rad this
01:26:52ani-adomfor a few times but still i cant get it.
01:28:26 Quit daurn|laptop (Read error: 131 (Connection reset by peer))
01:28:49ani-adompixelma: i dont see nthing that talks on id3.
01:28:53 Join daurn|laptop [0] (i=daurn@
01:30:58pixelmathat's related to the tags explained before: for example if you use in your wps code "%ia" to display the artist tag (ID3) of the currently playing song you have to write it "%Ia" to display the artist tag of the next song
01:32:18 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
01:34:39pixelmaJdGordon: morning ...
01:35:52pixelmaabout your settings patch (on Ondio) - it seems like nothing gets saved during buffering not even playlist_control
01:36:46JdGordonumm, damn, ok, thanks
01:36:55pixelmahad another shutdown when my batteries lost contact
01:37:54pixelmaand everything was back like before (volume too) so it should be a general problem
01:41:55 Quit spiorf (Remote closed the connection)
01:45:15 Join dau [0] (i=daurn@
01:45:17 Quit daurn|laptop (Read error: 54 (Connection reset by peer))
01:51:14 Quit TeaSea (Read error: 131 (Connection reset by peer))
01:51:50 Join TeaSea [0] (n=thunderc@
01:52:20pixelmaJdGordon: could it be an Ondio problem only?
01:52:51JdGordonIts possible, its wierd the playlist_control isnt getting saved, because thats sperate code
01:53:00JdGordonis playlist_control getting saved with cvs build?
01:57:58pixelmalast one I flashed 26.11. works (I'm not too eager to test the latest...)
01:58:25JdGordonwas that before or after the last ata_idle fix?
01:58:33*JdGordon checks logs
01:58:47pixelmathe one that caused the loop?
01:59:01pixelmathen it was after
01:59:31pixelmathe fix of course
01:59:50JdGordonok, i got soime nivestigating todo toda then
02:02:50 Join daurn|laptop [0] (i=daurn@
02:02:52 Quit dau (Read error: 104 (Connection reset by peer))
02:07:24pixelmaI can rule out that the volume the files are on (MMC or internal memory) has anything to do with it (thought it was worth testing)
02:10:15 Quit ani-adom (":")
02:13:14JdGordonby playlist_control, you mean the /.rockbox/.playlis_control file right? something is tottaly stuffed than if that isnt righting... tha doesnt use ata_idle_nofity
02:13:44pixelmawait a sec... had an idea (and if this is right then it is my fault)
02:15:35 Quit MarcoPolo|AFK ("Bye !")
02:17:34pixelmahmm... no - not my fault
02:18:00JdGordondamn, :p
02:18:53pixelmaI mean the file in which current playlist and position is saved, just to clarify
02:19:06pixelma(used to resume)
02:19:34JdGordonside noe, I want to replace the current bookmark stuff with that file, so bookmarks work much better
02:19:49JdGordondamn this keyboard is annoying me, sorry for all the typos
02:19:54 Quit TeaSea (Read error: 54 (Connection reset by peer))
02:19:59pixelmasounds like a plan :)
02:20:09JdGordonyeah, that file shuoldnt have any problems writing to disk
02:20:26JdGordonits goin to be a bit of work tho, because I'm not exactly sure how it works ;p
02:20:31 Join TeaSea [0] (n=thunderc@
02:22:32pixelmaall the playlists I created with "insert shuffled" could be bookmarked and I use this possibility more often because the Ondio doesn't have a quickscreen where I could set "playback" to shuffle easily
02:23:55 Quit wooo (Read error: 113 (No route to host))
02:24:27pixelma(if that made any sense... :/ )
02:24:35JdGordonanyone know how the vkeyboard works on the player?
02:24:58 Quit [g2] ("Leaving")
02:25:29JdGordonyeah, my breif experimenting with bookmarks found it only lets you bookmark folder playlists.. which i tinhk sux
02:26:09pixelma if it helps any
02:26:28JdGordonta, exaclty what i was loking for
02:26:54JdGordonwould that sort of keyboard work for the other targets?
02:30:00pixelmathere's one big difference to the other targets (afair) the charcell display only lets you have a limited amount of unusual characters - so I guess that's a great difference to all the others if you think about all the accented letters
02:30:17pixelmaand stuff like that
02:31:45JdGordonthats not such a big deal, because the vk lets you load your own keyboard file whih has as many accented characters as you like. It could work
02:31:54JdGordonHopefully will shrink some code also :)
02:33:08pixelmathat's definitely needed (if you read the logs...)
02:34:42pixelmaI cannot imagine how this virtual keyboard would look like on other screens and how usable it would be
02:35:04 Quit mick_home ("brb")
02:35:46JdGordonIm thinking about how easily it would scale compared to the current one, and it wouldnt force the sysfont on the screen
02:38:01pixelmasorry... maybe I'm too tired to see - would all the characters be on one line then?
02:38:04JdGordonwow the current vk drawing code is scary :)
02:39:03JdGordonIm tihnking it would allow for different sets of letters, so presing up/down would go to the next set (lower case, upper case, numeric, etc)
02:39:27pixelmaah.. did you see the proposal in the wiki?
02:39:39JdGordonthe qwerty keyboard one?
02:41:36JdGordonthat looks overly complicated, and scales as badly as the current one
02:42:00pixelmahmm... somehow I thought it was something like what you described but it isn't so...
02:42:04JdGordonhardcoding the character sets would be a big watse of space also
02:42:27JdGordonI dont kow what im thinking just yet :p
02:42:44JdGordoni going to have breaky, (at 12.42pm :D ) ttyl
02:43:09pixelmaI need some sleep though
02:45:26 Quit pixelma (" G'night")
02:46:03 Join r121 [0] (
02:53:50 Quit Strath (Read error: 104 (Connection reset by peer))
02:57:06 Join Strath [0] (
02:57:56 Join |AhIoRoS| [0] (n=ahioros@
03:00:50[Gino]Does Rockbox work on the 30gb iPod's now being sold on
03:03:33JdGordonpartially iirc
03:03:55JdGordonit needs some patches which arnt in the official souce just yet
03:03:59[Gino]Do you know what in specific doesn't work correctly?
03:04:15JdGordonKasperle knows about it i tinhk
03:05:11***Saving seen data "./dancer.seen"
03:05:17 Quit barrywardell ()
03:11:36 Join nave7693 [0] (
03:13:33 Join thomasg [0] (
03:14:45 Join dau [0] (i=daurn@
03:14:47 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
03:15:35thomasghi. I plan to buy a iaudio x5l dap. because of the great features I would like to use rockbox, but I have heard, that the powermanagement is not yet ready. how long could I listen to music with the x5l (e.g. mp3 256kbps)?
03:15:44 Quit TeaSea (Read error: 104 (Connection reset by peer))
03:16:21 Join TeaSea [0] (n=thunderc@
03:18:56 Quit r121 ("Leaving")
03:21:07scorchei think that the x5s have come to the stock firmware's usage now
03:24:34thomasgsounds good
03:24:36 Quit dau (Read error: 104 (Connection reset by peer))
03:24:52 Join daurn|laptop [0] (i=daurn@
03:26:41 Join tychver [0] (
03:45:54[Gino]Can you charge an x5 while it is running Rockbox, I mean while it is turned on... I read you can't, but I don't know if that got fixed...
03:46:27[Gino]Or any significant problems with the x5(l) running Rockbox?
03:49:58[Gino]Oh, does it support video playback?
03:50:32 Quit scorche ("Leaving")
03:51:42 Quit TeaSea (Remote closed the connection)
03:57:27 Join Rudy4Pez [0] (
04:19:40 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
04:19:45 Join dau [0] (i=daurn@
04:31:16 Quit midgey ()
04:33:31 Join mick_linux [0] (n=clamwin@clamwin/admin/mickhome)
04:40:00 Join daurn|laptop [0] (i=daurn@
04:40:07 Quit dau (Read error: 131 (Connection reset by peer))
04:42:46 Quit [sellout] (Read error: 148 (No route to host))
04:43:06 Quit funky ("leaving")
04:44:13 Join Canadian [0] (
04:44:54Canadianok i need some help with my Iriver H10
04:46:51 Quit Canadian (Client Quit)
04:48:45mick_linuxhow can i convert my HFS formatted ipod to Fat32 on debian?
04:48:54mick_linuxi've umounted the nano
04:49:11mick_linuxcfdisk doesn't show it at all
04:49:33mick_linuxi'm wondering what the best way would be to accomplish this
04:49:44mick_linuxand please don't tell me to rtfm :-P
04:53:24 Join Canadian_Pirate [0] (
04:53:26 Nick Canadian_Pirate is now known as Canadian (
04:54:16 Quit Canadian (Client Quit)
04:54:30 Join [ [0] (
04:55:44 Join [sellout] [0] (
05:01:23 Quit JoeBorn ("going to china tomorrow morn, tt you all later.")
05:03:10 Join dau [0] (i=daurn@
05:03:12 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
05:03:31[can anyone help me quickly with an Iriver H10, just put RockBox on and cant find the genaral settings menu...
05:04:26jbahit menu button
05:05:14***Saving seen data "./dancer.seen"
05:05:33[which is which button? (sorry for being stupid with this....)
05:05:52jbahmm not sure, different on each player
05:06:21[ok, ill play around with it...
05:06:22[found it, power button
05:06:35jbamight be power button
05:06:44jbaor try using the sliding selector, that might activate menu
05:07:01 Quit JdGordon (Read error: 104 (Connection reset by peer))
05:07:21 Quit |AhIoRoS| ("Abandonando, see you")
05:10:53 Join daurn|laptop [0] (i=daurn@
05:12:39[thanks mate i got it all working
05:12:45jbawhat was it ?
05:12:52jbaaah power button
05:13:25jbahow is bubbles plugin on the h10? am trying to convince a workmate to change firmware on it
05:14:14[like i said i JUST downloaded it and dont know how to use it, did reseach on it as a whole but not on useage
05:16:42 Quit [ ("Chatzilla 0.9.77 [Firefox 2.0/2006101023]")
05:19:39mick_linuxhow can i convert my HFS formatted ipod to Fat32 on debian?
05:20:04mick_linuxany linux/*bsd info would be helpful
05:30:05 Quit Kingstone ()
05:34:46 Quit dau (Read error: 110 (Connection timed out))
05:39:55 Quit XavierGr (Nick collision from services.)
05:39:59 Join XavierGr [0] (
05:44:14 Join ptw419 [0] (
05:45:43 Quit niskel (Remote closed the connection)
05:48:24 Join _Veseliq_ [0] (n=veseliq@
05:49:28 Nick myzar|away is now known as myzar (n=myzar@
05:53:58 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
05:55:08 Quit daurn|laptop (Read error: 145 (Connection timed out))
06:07:12 Join dantheman [0] (n=danielma@
06:08:48z35mick_linux: i believe that should help... I am not certain if it is the best way, you can also try to mount it using the hfs driver...
06:09:27mick_linuxi can mount w/ hfs+ atm
06:09:34mick_linuxi just want it to be fat32 ;)
06:09:39mick_linuxi'll check it out
06:12:23mick_linux"You can now unplug the iPod and pray ;)"
06:12:41 Join daurn|laptop [0] (i=daurn@
06:12:47mick_linuxno freakin way i'm reading that
06:18:20 Quit dantheman (Remote closed the connection)
06:19:11 Join scorche [0] (i=ScorchE@rockbox/administrator/scorche)
06:25:20 Join dantheman [0] (n=danielma@
06:36:12 Quit perpleXa ("Leaving")
06:36:59 Join shnee [0] (
06:37:01 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
06:37:08 Join daurn|laptop [0] (i=daurn@
06:39:07danthemanhi perplexa
06:56:27 Quit FATTY2 ()
07:00:16 Quit mick_linux (Read error: 104 (Connection reset by peer))
07:05:15***Saving seen data "./dancer.seen"
07:05:22 Quit ptw419 ()
07:18:03 Join mick_linux [0] (n=clamwin@clamwin/admin/mickhome)
07:18:17 Quit daurn|laptop (Read error: 131 (Connection reset by peer))
07:18:26 Join daurn|laptop [0] (i=daurn@
07:22:04 Quit actionshrimp (Read error: 60 (Operation timed out))
07:33:25 Quit dantheman (Read error: 104 (Connection reset by peer))
07:39:18 Part nave7693
07:47:15 Quit jba ("Leaving")
07:48:48 Join EbErT [0] (
07:51:54amiconnSlasheri: Sorry for the ranting last night, in case you read the logs
07:52:38amiconnBut the more I think about it, the more I think the current implementation for large virtual sectors is the wrong approach and should be reverted
07:52:54amiconnSome points:
07:53:17amiconn(1) file.c shouldn't be touched at all. Virtual sectors is a fat only thing and should stay in fat.c
07:53:41SlasheriHmm, but file.c uses directly sectors as unit when measuring things
07:53:52amiconnYes, and one sector is still 512 bytes
07:54:14amiconn(2) None of the sector buffers should need enlarging. The virtual sectors are what their name says - virtual.
07:55:05amiconnThe sector buffers are mainly there because a sector is the smallest addressable unit in the ata protocol
07:55:14Slasheribut then, how fat driver could handle reading and writing if sector size is not enlarged?
07:55:30amiconnLike it does without your commit
07:55:51amiconnThe fat specifies cclusters, and a cluster is usually also multiple sectors
07:55:56Slasheriand all the cache stuff in fat etc..
07:56:17Slasherithat would require quite major change to the driver to support different sector size without changing the units
07:56:18amiconnI tend to see large virtual sectors as an extension to clusters
07:56:41amiconnI think the changes needed are smaller than what you committed
07:57:24amiconnI think the instability you mentioned is just because putting multiple 2KB buffers on the stack
07:57:49amiconnThe main stack is 8KB in total, and the individual thread stacks are similar or smaller
07:58:00*amiconn wishes Zagor were here...
07:58:07amiconnHe's Mr. fat.c
07:58:45Slasheriok, if you think it can be implemented better by not changing sector size, feel free to revert the commit
08:04:34 Join ender` [0] (n=ender@
08:05:19amiconnI'm a bit undecided. I think that implementing another way is easier when working from the status before the commit, but otoh reverting would mean to back out G5.5/30 support
08:09:56amiconnSlasheri: We could even support all possible virtual sector sizes then. Valid sizes are quite limited: 512, 1024, 2048 and 4096
08:10:12 Join spiorf [0] (
08:11:00 Join MarcoPolo [0] (
08:11:13Slasheriamiconn: at the moment we support 512, 1024 and 2048 on fat level
08:11:47amiconnYes, but that's with enlarged buffers, and with file.c knowing about it
08:14:54Slasheriat looking fat.c, i see it would take quite a massive change to do it the other way.. but if you see it's even easier, great :)
08:20:57 Quit hachi ("And I begin to wonder... The dreams I can't remember when I wake in the morning - where in the world did they go?")
08:21:57 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
08:34:19amiconnSlasheri: Why do you think so?
08:36:07Slasherii am not sure, just got "the feeling"
08:37:54*amiconn wonders whether the original implementation of the fat driver even already had large virtual sector support
08:38:39amiconnI remember Zagor speaking about fat.c being cut down from something
08:40:24amiconnSlasheri: Btw, afaics there are 2 buffers in fat.c which can be PHYSICAL_SECTOR_SIZE instead of MAX_SECTOR_SIZE
08:41:59 Join dau [0] (i=daurn@
08:42:07 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
08:42:13 Join Tyronemaskell [0] (n=Dude@
08:42:14amiconn(1) The one in fat_mount(), because the bpb is always 512 bytes (including the signature)
08:43:40Tyronemaskellmorning all
08:43:53Tyronemaskellwhere can i find themes
08:43:59Tyronemaskellif thats what you call them
08:44:02amiconn(2) The one in update_fsinfo()
08:45:04EbErTright there, Tyronemaskell
08:45:05Tyronemaskellthank you
08:45:19EbErTthere called wps (while playing screens)
08:45:40 Quit dau (Read error: 104 (Connection reset by peer))
08:45:50 Join daurn|laptop [0] (i=daurn@
08:46:21Tyronemaskellthe only problem i have these days is that i need to convert my wma to mp3's any ideas pls
08:46:26amiconnEbErT, Tyronemaskell: Rockbox also knows themes, which include wps and some more settings
08:46:42amiconn(font, colours, backdrop)
08:46:53EbErTyou mean already installed
08:47:03EbErToh, right
08:47:10Slasheriamiconn: hmm, true. i just wanted to be sure at first no buffer overflows happens
08:47:13amiconnAnother source is
08:47:13EbErTthose are themes and wps, mixed in
08:48:27Tyronemaskellhow do i convert wma files
08:48:33amiconnSlasheri: Btw, afaics all we need is the secmult thing, then supporting large virtual sectors shouldn't be a problem with 512-byte sector buffers
08:49:09EbErTgoogle it, you just need a program
08:49:10Slasheriamiconn: sounds good if that can be done
08:51:30scorcheTyronemaskell: transcoding is always a bad did you aquire the material?
08:51:58Tyronemaskelli got the songs from a friend
08:52:17scorchewell, i would suggest re-ripping the material if you can
08:52:24Tyronemaskelland ive always gotta boot into the iriver firmware to play the wma tracks
08:52:46Tyronemaskelli can re-rip the stuff
08:53:03Tyronemaskellranscoding is always a bad did you aquire the material? why is it not a good idea
08:53:18Tyronemaskellmeant to say i cant re-rip the stuff
08:53:32Tyronemaskelland transcoding
08:54:16scorchebecause you will always experience a loss in quality
08:54:48Tyronemaskelllike a serious loss
08:55:22 Quit EbErT ()
08:55:29scorchewell, that varies between one's definition of "serious loss"
08:55:31Tyronemaskellive found that i can make an audio cd from the wma files but then i need to re rip from the cd what a mission
08:55:45scorchethat is transcoding
08:56:37Tyronemaskellok i see
08:57:11Tyronemaskellhow do i remove the current installed wps and themes ?
08:57:27scorchejust delete the files
08:57:34scorcheyou cannot remove the default one though
09:02:05Tyronemaskellok so if i cant re-rip and i wanna transcode whats a good program to use
09:05:17***Saving seen data "./dancer.seen"
09:07:35 Join einhirn [0] (
09:07:35 Quit einhirn (Client Quit)
09:07:45 Join einhirn [0] (
09:07:45 Quit einhirn (Client Quit)
09:08:29 Join webguest20 [0] (i=c27f0812@gateway/web/cgi-irc/
09:10:13webguest20I recently noticed that the background in the splash messages on H120 got more dark grey, it's even hard to read. When was it changed? It was very light grey before.
09:10:24 Join pondlife [0] (n=Miranda@rockbox/developer/pondlife)
09:12:53 Join actionshrimp [0] (
09:13:22 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
09:15:04webguest20Hello petur. IIRC you have an H340. Do you also have difficulties reading splash messages? Or is the problem H120 specific? Or is it just me? (see msg above)
09:15:19*petur reads log
09:15:43peturah that will be h1x0 related
09:16:18peturjhMikeS changed lcd gamma which may have caused this... amiconn may know more
09:17:38webguest20amiconn: are you here?
09:18:01*jhMikeS is ready for sleep but will suggest lightening the contrast
09:18:22peturgoodnight jhMikeS
09:18:37*petur just got up and arrived at work ;)
09:18:39jhMikeS'night - will lurk
09:20:36webguest20petur: do you know jhMikeS well? Will he actually DO it when he says "suggest"? :-)
09:21:08peturI haven't met any of the devs if that's what you mean ;)
09:21:41peturthe contrast thing has been talked about before, it'll get fixed
09:21:49linuxstbmick_linux: You should be able to adapt the Mac OS X instructions for converting to HFS+ to other Unixes -
09:22:09mick_linuxthat doesn't exist on linux
09:22:13mick_linuxonly freebsd
09:22:20mick_linux*bsd actually
09:22:21linuxstbWhat doesn't exist?
09:23:00linuxstbmkfs.msdos exists.
09:23:04webguest20petur: ok, I'll wait then
09:23:14mick_linuxand fat32 / vfat formatting on linux is flaky at best, so i may just try it in wine ;)
09:23:17linuxstbOr mkdosfs, or mkfs.vfat (I think they're all the same).
09:23:28rasherNever brought me any trouble
09:23:31linuxstbIt's worked fine for me.
09:24:06linuxstbIt's the fsck which I think is unreliable/non-existent on Linux.
09:24:34jhMikeSwebguest20: one more thing. H120 or H140? Just try move the contrast lighter and see.
09:26:11scorche"Or is the problem H120 specific?"
09:26:21jhMikeSfor my and some others H120 the setting is perfect. maybe the LCDs really vary greatly.
09:26:22mick_linuxcfdisk exists :-P
09:26:25webguest20jhMikeS: H140 actually. I set the contrast so that the text is as black as possible but the screen is not grey
09:26:41rasherLooks too dark on my ihp120 said it was originally a 120...oh well
09:27:15rasherWell, h120 is the modelname rockbox uses for h120/h140, so that's understandable
09:27:33scorcheyeah...just saying =)
09:27:33jhMikeSamiconn has an H140 and says it's too dark. preglow and I have H120s and found it better...hmmm
09:28:09jhMikeSbut the default contrast needed to be one lighter (27 not 28)
09:28:43jhMikeSlooks like a setting is in order in order to make it work for everyone.
09:29:01jhMikeSok nighty night
09:29:31rasherStill looks a bit on the dark side to me
09:29:47 Part hcs
09:29:57webguest20IIRC the contrast is set to 27 or 28 (default setting). The screen (e.g. file browser) looks nice, just the splash msgs are too dark
09:30:33jhMikeSA scale of solid gray bars should look even (0, 85, 170, 255)
09:30:57*petur sends jhMikeS to bed
09:30:59rasherMaybe the splashes just need to be light grey background?
09:31:48jhMikeSIf the bars are even, things should all be readable as they are for me. The grays need to be set for each user for the best possible image.
09:32:56*jhMikeS has been sent to bed and is about to pass out anyway :)
09:38:28webguest20jhMikeS: do I understand right that solid greys are possible at 0, 85, 170, and 255? What value is used now?
09:38:51peturssshhhh let him sleep
09:52:05Tyronemaskelldbpoweramp does not convert wma files ??????????
09:53:45 Join _FireFly_ [0] (
09:54:47scorcheit will, but this discussion is more suited to a PM
09:55:02Tyronemaskelllooks like it might work heheheh
09:55:22Tyronemaskellive installed some codecs
09:56:50scorchesorry if i was being too the above message, i meant that discussion about that particular program does not belong in this channel
09:57:18Tyronemaskellno probs hehehe
10:04:04Tyronemaskellgot it too work thanx again
10:06:02amiconnBagder: ping...
10:06:08*Bagder pongs
10:07:00amiconnAh :)
10:07:03Tyronemaskellanother question
10:07:08*linuxstb fails to find a win32 API call to get the sector size for a disk - only for a drive letter...
10:07:11amiconnBagder: Do you know if and when Zagor will be around?
10:07:32Tyronemaskelldoes running a colour backround and wps use more battery
10:07:41Bagderamiconn: he's not around much these days since on a paternaty leave full-time, so he's kept busy by his son
10:09:40amiconnHmm :/
10:10:09peturUrrrghh... just heard a glitch in my recording - the missing samples probnlem isn't fixed yet. Or maybe I should defrag my disk.... :( :(
10:11:21 Join Hdaackda [0] (n=klj@
10:12:50 Join dau [0] (i=daurn@
10:12:53 Quit daurn|laptop (Read error: 54 (Connection reset by peer))
10:14:24*petur reopened the bug report in the tracker
10:14:49Tyronemaskelldoes running a colour backround and wps use more battery
10:15:13peturnot really
10:15:59petura busy wps (peakmeters) can cause some more cpu activity
10:19:48amiconn...and a bitmap intensive wps will take longer to load
10:25:35webguest20Is a WPS with elapsed/remaining time and the progress bar "busy"?
10:26:20preglowthat's not good
10:27:33webguest20preglow: should it be considered busy? Or what do you mean? Not easy to follow (implicit) threads here :-)
10:27:50preglowwebguest20: talking about the recording glitches, heh
10:28:32webguest20preglow: aha!
10:32:23 Quit Hdaackda ("Bye")
10:34:53peturAAAARGH.. and another glitch (two in a row this time)
10:36:03preglowpetur: what do they look like?
10:36:29peturhaven't looked at the waveform, just listening to the recording
10:36:48peturthe recording jumps a few seconds further
10:38:34peturseeking is borked too... seeked backwards to get the exact timestamp and ended up further in the file (it's one big WAV, 1GB). Tried to seek back 5 minutes as I huessed that is where the glitch was and it just froze
10:38:44*petur gets paperclip
10:38:54 Join God_Eater [0] (
10:41:41 Quit GodEater (Read error: 110 (Connection timed out))
10:42:59peturpreglow: this 'double' glitch is in fact a 2 second part that is present twice. A 2 second part is written over by a previous part.
10:43:26peturlike skipping back 2 seconds, except that after replaying those two seconds it moves on to where it should have been
10:44:00preglowa glitch++, then
10:44:13*petur doesn't think this is funny
10:44:48 Quit scorche ("Leaving")
10:44:50*preglow finds most things funny
10:44:54preglowkeeps me from killing myself
10:44:59peturthe last two shows I recorded are thrashed. I now have 0% trust in recording code
10:46:08zetry transferring and decoding or using an mp3fixer.. i've had recordings (way back in archos days) that seemed screwed up get fixed by that
10:46:13preglowbut the fact that entire buffers got repeated probably does indicate quite nicely what the bug is
10:46:14peturI'll probably check out some CVS code from this summer (before jhMikeS touched it) and use that as production code
10:46:43blue_lizardhi petur
10:46:53blue_lizardthx for your comment for my patch
10:46:58preglowi've pretty much always said that tapers should use a knows safe version of rockbox until something even more known and safe comes along
10:47:08*petur scratches memory
10:47:10blue_lizardunfortunately i have not understood the two last sentences
10:47:13preglowthe code was pretty nice during the summer, i think
10:47:46blue_lizardant therefore i dont know how to react
10:47:48preglowfinding bugs suck when doing something you can't redo two minutes later
10:49:03 Join bluebrother [0] (i=9BEMGdvU@rockbox/staff/bluebrother)
10:49:22peturblue_lizard: what is there you don't understand?
10:50:11blue_lizardI had no choice but to move the file/dir property to a plugin because of the wasted binary size <- what does that mean for me?
10:50:25blue_lizardwhat does that meat at all?
10:50:57peturyour code will never get committed because it will add too much to the binary with little needed functionality
10:51:15[Gino]I'll just ask this again... I asked last night and got no answers... Can you charge an x5 while it is running Rockbox, I mean while it is turned on... I read you can't, but I don't know if that got fixed...Also, does Rockbox support video playback (on the x5) and are there any significant problems with the x5 running rockbox?
10:51:37peturif the whole fileoperations went into a plugin, nobody would bother that it is bigger just to show more info
10:51:49 Join scorche [0] (i=ScorchE@rockbox/administrator/scorche)
10:51:58preglownot a bad idea, the disk access needs to be done anyway
10:52:41peturFile/Dir properties would have added 1500 bytes for a non-critical and not often used thing, same goes for file and dir operations
10:52:53blue_lizardcould this plugin be called from the kernel or does the user explicitely call the plugin instead of the build in function?
10:53:12peturwe need a place to put these 'overlays'
10:53:26peturI put mine in as a viewer but that's just a workaround
10:53:42preglowi think a couple of more plugin category directories would be pretty nice too
10:53:44peturblue_lizard: it can be called from the main code
10:53:47preglowlist is getting big
10:54:07peturblue_lizard: look at how I added the file/dir properties
10:54:21blue_lizardpetur: thx i will look at it
10:54:40bluebrother[Gino]: have a look at the PluginMpegplayer wiki page
10:54:53Bagder[Gino]: and yes it charges while Rockbox runs
10:55:07Bagderand no, there are no "significant" problems
10:55:45bluebrotherisn't the x5 charging also hardware? I don't see a reason why it shouldn't work as long as charging is in hardware
10:56:11[Gino]Thanks guys, one more question. ;) Can you change the wallpaper, or does it depend on the theme you are using?
10:56:22preglowit's independent, just do 'set backdrop', afaik
10:56:38 Quit courtc (Remote closed the connection)
10:56:49[Gino]Not sure, but Iread it doesn't charge correctly while it is turned on and running Rockbox.
10:56:51[Gino]Thanks preglow
10:57:25petur[Gino]: copy the correct bmp into /.rockbox/backdrop, browse to it and from the context menu select 'set as backdrop'
10:57:44peturor something like that
10:58:13[Gino]Thanks petur
11:03:15Tyronemaskellive got a problem
11:03:36Tyronemaskellmy screen went all dark when i loaded a theme ?
11:04:52Tyronemaskellany help ?
11:05:21***Saving seen data "./dancer.seen"
11:05:38bluebrotherTyronemaskell: maybe it has black background and the foreground color is also black?
11:05:54Tyronemaskellheheh you might be right
11:06:33bluebrotheror the theme is broken, your fg color was set to black and the theme didn't set it some something different
11:06:33zeif so i bet menu options light up black when you select them too
11:06:55bluebrotherblack on black might be hard to read ;-)
11:07:32*bluebrother in fact got a book with a cover printed white on white some years ago, but as this was a different surface it was still readable
11:07:37 Join courtc [0] (
11:07:49zeyeah people've done that with black on black too
11:07:52zefor books and cds
11:07:59zeand its like one is matte and the other glossy
11:08:04zeor the like
11:08:38 Join n1s [0] (
11:08:42*bluebrother remebers to have seens something similar on a CD
11:09:15bluebrotherit's a nice idea especially for an art exibition
11:09:15[Gino]Oh, found another question... If a theme has album art, can you turn it off, or does it have to be on if it's in the theme?
11:09:30bluebrotheralbum art is unofficial. No idea
11:09:46rasherMaybe it's a Disaster Area stunt-ship theme
11:10:36[Gino]Oh yeah, I forgot you needed th albumart patch... I wonder if the theme would still work without the patch and just not show the album art...
11:11:04bluebrotherjust try it ... I'm not sure if the tag will be shown or not
11:11:21bluebrotherbut I guess the theme will have some elements somewhere off
11:12:16[Gino]Well, I will, but I don't even have the DAP yet. ;) I'm trying to figure out what I should get.
11:12:28Tyronemaskellany way you can select more than one file for deletion
11:12:42peturpreglow , jhMikeS: petur/glitch.wav"> - glitch is about 13 seconds into that clip
11:12:59n1sTyronemaskell: no
11:13:15markunmultiple selection would be nice for many things
11:13:56Tyronemaskellok some themes have backrounds then when i load another theme the old backround still stays how do i get rid of it ?
11:14:23n1s"clear backdrop" in the lcd settings
11:14:27 Quit sando ("blargh")
11:14:37preglowbluebrother: i've got a cd cover printed white on white, but the one layer with gloss so you can read it at an angle, heh
11:15:35preglowpetur: pretty annoying glitch too
11:15:49peturtell me about it :/
11:16:18*petur wonders if his disk may be to fragmented and this may cause it
11:16:32preglowwell, at least you can see if it is
11:16:36preglowthere's gotta be a tool for that
11:16:46preglowfat isn't exactly known for handling fragmentation graciously
11:16:51 Join |[Gino]| [0] (
11:17:04peturthere's aout 6GB free on it though
11:17:19preglowsure, but it may be fragmented as hell
11:17:24preglowso i'd at least do a check and see
11:17:43preglowsounds a bit unlikely anyway
11:18:15Tyronemaskelldo i have to delete each theme one by one ?
11:18:32peturwell if saving the file takes too much time, the PCM buffer would overflow? Or rather not, as DMA would be halted and recording stopped
11:19:07peturmust be encoder buffer overflow
11:19:16peturor overwriting
11:19:29preglowjhMikeS will probably have a better explanation
11:19:56Tyronemaskelllol another question is there anyway i can view the themes on my pc
11:20:05preglowTyronemaskell: yes, with a rockbox simulator
11:20:40Tyronemaskelllol where do i find that ?
11:21:02n1syou'll have to build it from source
11:21:14bluebrotherthere are unofficial builds on rasher's site
11:21:52 Join webguest26 [0] (i=c2c7fca1@gateway/web/cgi-irc/
11:22:34Tyronemaskellbuild it from source ?
11:22:43Tyronemaskellnot you talking french
11:23:10linuxstbAny Windows using ipod owners want to test a new version of ipodpatcher? I need both normal and 5.5g users...
11:23:39 Quit Rudy4Pez (Read error: 110 (Connection timed out))
11:23:44n1sTyronemaskell: if you're on windows you can get it here (as bluebrother said)">
11:23:55bluebrotherTyronemaskell: from what do you want to build the sim otherwise?
11:24:20scorchelinuxstb: looking for anything in particular?
11:24:21bluebrotherbuilding the sim from fresh air obviously won't work ;-)
11:24:22webguest26on a besond 'un français ?
11:24:23linuxstbHopefully that ipodpatcher should detect the sector size used.
11:24:41linuxstbSo if you've got an old ipod, it should say "512" and work as normal.
11:25:01linuxstbBut it's completely untested, so be careful...
11:25:12linuxstb(although reading from the disk should be harmless)
11:25:39preglowso, now we've the installer will break the filesystem before rockbox itself can? :>
11:25:42scorcheah...just want to gather results from reading and not writing?
11:26:26linuxstbscorche: No, writing needs testing. But obviously, test reading first...
11:27:25linuxstbpreglow: Glad to see you have faith in my blind programming.
11:28:30preglowjust following up on amiconn's recent bout of fs code paranoia, heh
11:28:45 Join Rudy4Pez [0] (
11:31:04rasherSpeaking of (almost).. Anyone want to give this a stamp of approval? - FS #6358 — Check for illegal chars on filename creation
11:31:30preglowso, aren't we supposed to be releasing rockbox 3.1 now?
11:32:05Slasherioh, just got my iriver motherboard back :)
11:32:56linuxstbpreglow: No, a month ago...
11:33:13 Join daurn|laptop [0] (i=daurn@
11:33:30scorchelinuxstb: all is well here
11:33:39 Quit dau (Read error: 104 (Connection reset by peer))
11:33:58preglowrasher: at least i can see nothing wrong with it
11:34:19Tyronemaskellhow does simulator work ?
11:34:21rasherI'm just scared of committing anything to the fat code
11:34:29preglowrasher: understandable
11:36:32Tyronemaskellanyone ?
11:36:37preglowTyronemaskell: try it out
11:36:40preglowit's like using rockbox
11:37:01*petur swears a bit at UPS
11:37:26amiconnrasher: I would prefer holding that back a bit
11:37:40amiconnNot because I don't trust it, but because of the variable sector size thing
11:38:54rasheramiconn: Don't quite get how they relate, but it's your call
11:39:21Tyronemaskellbut how do i get it to work
11:39:38amiconnBoth touch fat.c, and I think the current implementation is the wrong approach
11:39:57rasherTyronemaskell: Did you download the sim from my site?
11:40:10bluebrotherTyronemaskell: just run it. There's also some documentation in the wiki (UiSimulator)
11:40:23linuxstbscorche: Thanks. So I assume it displays the sector size as 512? Did you test writing?
11:40:34scorchei did
11:40:50scorcheit displays sector size on any drive i tell it to read from
11:41:06linuxstbGood. All I need now is a 5.5G owner...
11:42:27|[Gino]|Do you install plugins after installing
11:42:38|[Gino]|Rockbox, or while you're installing?
11:42:47bluebrother|[Gino]|: no. Plugins are part of the installation archive
11:43:24linuxstbSlasheri: Will you be patching ipod_fw.c for the 5.5G?
11:43:31Slasherilinuxstb: hmm, sure :)
11:43:52Slasheriwhat did you do to it?
11:44:05Slasherii was trying to get OF loading but didn't succeed with that yet
11:44:06linuxstbNothing... I haven't touched it. I've just fixed ipodpatcher.exe
11:44:13linuxstb(for Windows users)
11:44:22Slasherioh, but i don't have windows..
11:44:28|[Gino]|So I take it, there are no "extra" plugins and all of them are installed by default?
11:44:34linuxstbSlasheri: I hear some people use it though.
11:44:45 Quit [Gino] (Connection timed out)
11:46:06amiconnI suppose linux can create fat partitions with larger sectors than physical?
11:46:38rasherLinux can create all sorts of strange stuff, so why not
11:46:59rasherBut I don't actually know.
11:47:04bluebrotherexcept from 3rd party plugins that are not part of the official rockbox distribution (but in most cases you'll need to compile those yourself), yes.
11:47:08amiconnThis would be great for testing large sector support...
11:47:23rasher -S logical-sector-size
11:47:23rasher Specify the number of bytes per logical sector. Must be a power of 2 and greater than
11:47:26rasher or equal to 512, i.e. 512, 1024, 2048, 4096, 8192, 16384, or 32768.
11:47:43amiconnGreater than 4096 isn't part of the official standard
11:47:53preglowwhat rasher said, then
11:48:00rasherIt seems Linux is not much interested in the official standard
11:48:34amiconnI'll repartition one of my targets then in order to test
11:48:47amiconnI also have an extra disk I could throw in
11:48:49bluebrotherI'd call that rather "allows the (experienced) user to do stuff outside of the standart ;-)
11:48:51rasherSee, allowable characters in filenames, size of filenames and no doubt many other things
11:49:17bluebrotherfor testing you could also create a filesystem in a file
11:49:23rasherStandards should be followed. Even when they're limiting.
11:49:34*bluebrother agrees
11:49:38amiconnbluebrother: Yes, and there is even fat test code that does that
11:49:43peturI'm sure that if Microsoft would do this another explanation would be given to it :p
11:50:02amiconnBut testing on the real thing is better imho, because it tests limitations not present on the host
11:50:06bluebrothermkfs.vfat <options> /path/to/my/filesystem-file.img
11:50:27|[Gino]|bluebrother: All of the plugins on the PluginIndex wiki page are included right?
11:50:36bluebrother|[Gino]|: yes
11:50:40amiconn(like the inability to do misaligned accesses, stack size limit etc)
11:50:40linuxstb|[Gino]|: Yes, but not all work on all targets.
11:50:59bluebrotherexcept for targets that don't support it (but you can also see that on that page)
11:51:02*amiconn summons Zagor
11:51:13bluebrothere.g. the h100 don't have mpegplayer etc.
11:52:09 Part webguest20 ("'ll be back later")
11:52:47scorcheSummon Failed.
11:53:42|[Gino]|K, Thanks
11:56:03n1sDoes anyone object to this?
11:58:18 Join anathema [0] (
11:59:00 Join wooo [0] (
12:01:19rasheramiconn: How is it not the right approach by the way?
12:05:38amiconnCheck this morning's log...
12:05:53rasheramiconn: Ah, you meant the large sector thing, not my patch?
12:05:59peturor yesterdays rant :)
12:08:40amiconnrasher: yes
12:09:16*linuxstb needs someone with rights to delete attachments from the twiki (Bagder?)
12:09:44linuxstbAlmost everything on the IpodInstallation page - the files you moved to
12:12:05Bagderwhat files should I keep around?
12:12:43 Join thomasg_ [0] (
12:14:42linuxstbThe files not on are bootloader-color-nodefault.bin, loader.bin, ipodloader.conf, ipod_fw_5.5g.exe but they all infringe the GPL.... (apart from ipodloader.conf - a config file for ipodloader2)
12:15:10linuxstbi.e. people have added them to our wiki without source.
12:15:29linuxstb(loader.bin is the IPL bootloader)
12:15:47Bagder and the .conf would makes more sense as a plain wiki text, since it is 6 lines of plain text...
12:17:09linuxstbI'll replace the link to loader.bin in the page to a link to IPL's website so that people can download the files directly from them.
12:17:15linuxstbSo delete loader.bin
12:17:52 Quit anathema (Read error: 113 (No route to host))
12:20:29 Quit petur ("lunch")
12:23:47Slasherilinuxstb: ah, it's possible to eject /dev/sda on linux while ipod disk mode to cause it spin down hd and charge faster
12:25:02Bagderok, only two files left there now
12:25:13Bagderis anyone using the ipod_fw_5.5g ?
12:25:35 Join jba [0] (
12:25:48 Join niskel [0] (
12:28:54 Quit thomasg (Read error: 110 (Connection timed out))
12:31:19 Quit blue_lizard (Read error: 110 (Connection timed out))
12:31:37 Join blue_lizard [0] (
12:37:46linuxstbBagder: It's linked to in the installation guide on the first page of the 5.5G "New Ports" thread, so I guess people are using it.
12:38:24Bagderok, let's leave it there for now then
12:38:31linuxstbBut there's no info about the source, so you could delete it for that reason - it's not our code.
12:42:33 Join funky [0] (n=repulse@unaffiliated/funky)
12:45:43*scorche wonders what divorce has to do with rockbox....or even FOSS ;)
12:47:38_Veseliq_just some fun guys
12:48:03scorchefun?...what is this "fun" you speak of?
12:49:24scorchealso, if you have very overzealous admins, NSFW
12:49:52 Quit Kitt0s (Read error: 131 (Connection reset by peer))
12:50:28 Join Kitt0s [0] (
12:55:43 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
13:02:32tychvermildly amusing, I find it a little bit too crass
13:03:02tychvernot that I'm not fond of some crassness, I just didn't think it was that funny
13:05:23***Saving seen data "./dancer.seen"
13:12:35peturhahaha... 'Toshiba Gigabeat brunch'
13:13:18petur(taken from )
13:16:17 Join dau [0] (i=daurn@
13:16:19 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
13:16:59 Join Gnelik [0] (n=Miranda@
13:19:04rasherpetur: That's a really good article.
13:19:40rasherAs a "what is Rockbox" introduction
13:19:58Bagdergenre9mp3 wrote it
13:20:03Gnelik Hi
13:20:13peturyes, it's not bad
13:20:21Bagder... and I proof-read it before it went public
13:20:49Bagderan "inside job" ;-)
13:21:42rasherAh. That would explain the accuracy...
13:22:23 Join barrywardell [0] (
13:22:58rasherHrm.. I really think rockbox.* should be moved into .rockbox
13:23:25rasherPeople seem to think they have to click it
13:23:35rasher(example, latest post on the ml)
13:23:47GnelikIf i'l teke some memory from sound buffer how can i free it??
13:24:19linuxstbGnelik: You can't... The memory is freed when your plugin exits.
13:25:17 Join bbroke [0] (n=bbroke@promi22.zmms.TU-Berlin.DE)
13:25:20linuxstbAlso, you don't want to use the audio buffer if you can avoid it - it will mean that playback is stopped.
13:25:31 Part bbroke
13:27:05linuxstbGnelik: Do you know approximately how much memory you will be using?
13:27:36markunpetur, rasher, Bagder: and this guy read some things in the article that are just not there:
13:28:21Gneliklinuxstb: not more than 100kb
13:28:24 Quit dau (Read error: 131 (Connection reset by peer))
13:28:30linuxstbmarkun: He must know something (or someone...) we don't.
13:28:47 Join daurn|laptop [0] (i=daurn@
13:29:38linuxstbGnelik: Try to use arrays of a fixed size instead of allocating things using malloc.
13:31:18linuxstbmarkun: I assume there's no wi-fi in the Gigabeat S?
13:31:46Gneliklinuxstb: It'l be not to eaasy for me
13:32:18 Quit [sellout] (Read error: 145 (Connection timed out))
13:32:21*amiconn always found static arrays way easier than malloc
13:32:24linuxstbGnelik: But that's basically what you'll need to do. You then replace pointers with indexes into that array.
13:33:55GnelikCan't i just make malloc that will return a pointer to some array item???
13:34:53markunlinuxstb: correct
13:36:32markunGnelik: there are some simple malloc implementations in the code.
13:37:43markunlinuxstb: I posted a comment to nuance his story a bit, but it doesn't show up on the site. Maybe he first needs to approve it.
13:38:57linuxstbGnelik: I would also suggest writing your code outside of Rockbox first (i.e. a small standalone C program). That will be easier to test and debug.
13:42:23 Join ShacharLiberman [0] (i=590001c1@gateway/web/cgi-irc/
13:42:47 Join [sellout] [0] (
13:43:00 Join yw [0] (n=chatzill@
13:43:02 Join Shachar_Liberman [0] (i=590001c1@gateway/web/cgi-irc/
13:43:02 Quit ShacharLiberman (Client Quit)
13:43:12*Shachar_Liberman ?
13:43:46ywcan the rockbox codec run on sansa e200 now?
13:44:25 Join lamed [0] (i=590001c1@gateway/web/cgi-irc/
13:44:25 Quit Shachar_Liberman (Client Quit)
13:44:33Gnelikyw: We can't play any sound now
13:44:37ywanyone know how to do the translate for the rockbox?
13:44:56lamedyw: how do yoy mean
13:44:56ywI want to add a simpilified chinese lang for rockbox
13:46:00*linuxstb wins!
13:48:04ywif I made the lang file, can I commit it into the cvs directly?
13:48:28lamedonly if you have committers access
13:48:30*lamed win
13:48:54lamed#rockbox @ Freenode stats by Jonas Häggqvist reports I'm asking too much quesions too many times. (or that i'm stupid). I shell stop using question marks from now on.
13:49:47peturit says I'm aggressive but that won't stop me from slapping my forehead again :)
13:50:55lamedI've used "really" 699 times during this year's june!
13:51:17 Join secleinteer [0] (
13:51:26peturread the statistics again
13:51:59lamedpeutr: do you also get "Is Shachar Liberman stupid or just asking too many questions? ... Another old yeller was Shachar Liberman, who shouted 9.4% of the time! ..." as the third result when you google your name (question mark)
13:53:41 Quit yw ("Chatzilla 0.9.77 [Firefox 2.0/2006101023]")
13:54:03lamedbagder: around?
13:56:42lamedbagder : : I've figured if it's gpl'd and you can distribute it freely, then It may use trademarks, as trademarks are for commercial purpose. I'm sorry it took me 2 month to respond...
13:57:21Bagderyou confuse matters
13:57:36amiconn"Poor forehead, nobody likes him/her. He/She was attacked 3 times"
13:57:52Bagderlamed: they can indeed have trademarks that would make it impossible to ship something even if it is GPLed
13:58:28lamedthat's a cruel world we're living in.
13:58:34linuxstbAlso, GPL doesn't mean non-commercial
14:00:55Slasherilinuxstb: hmm, is there something special to do in order to make the COP running? (other than just defining NUM_CORES 2 etc.)
14:01:00amiconnrasher: Is there a reason why your irc stats only list the realnames for some of the nicks?
14:01:12amiconnAll mentioned nickas in the top stats are known...
14:01:52bluebrotherbtw, who is forehead? Haven't seen him in the channel :-P
14:02:05rasheramiconn: I did the configure a while ago, and set realnames for all the nicks I could find
14:02:28 Join forehead [0] (i=3e088e42@gateway/web/cgi-irc/
14:02:54*petur slaps forehead for real this time
14:03:01bluebrotherpetur, beware slapping forehead. He may strike back
14:03:09bluebrother... too late ;-)
14:03:10foreheadOuch! :)
14:03:24 Quit jba ("Leaving")
14:03:42 Join TeaSea [0] (n=thunderc@
14:04:03 Quit petur ("worrrk")
14:04:15 Quit forehead (Client Quit)
14:04:51lameddid any of you see the new zune player by microsoft (question mark)
14:04:52linuxstbSlasheri: The COP support isn't working properly AFAIK.
14:05:13linuxstbSlasheri: Check dan_a's patches on flyspray - I think there's one there which isn't committed.
14:05:14 Quit daurn|laptop (Read error: 131 (Connection reset by peer))
14:05:18 Join daurn|laptop [0] (i=daurn@
14:05:44Slasheriah, ok thanks
14:06:58 Join hsaliak [0] (
14:07:11hsaliakhello , i am trying to install rockbox on my ipod nano from netbsd
14:08:52Bagderand... ?
14:09:32hsaliaki am trying to translate the instructions for linux into something bsd, now it says sdb1 is the firmware partition for linux, however i see 2 partitions only in netbsd, sd0d and sd0e, in netbsd the d partition always represents the full disk, so i have extracted the mbr from that
14:10:06hsaliaknow the firmware.. , has any user done such things from freebsd or any other bsd before? im sure there have been?
14:10:48BagderI know people have used freebsd at least
14:11:18Tyronemaskellone question pls if i load a new theme do i need to load the wps file too ?
14:11:48hsaliakyes if this works surely ill put something in the wiki
14:11:54linuxstbhsaliak: The Mac OS X instructions are probably closer to BSD than the Linux ones.
14:14:02markunin FreeBSD my players show up as da0
14:15:32hsaliaklinuxstb: cool ill take a look at that
14:16:28linuxstbThe question is if bsd gives you a device node for partitions of type 0 (which means "empty"). On Linux you do, but on Mac OS X you don't - so on Mac OS X you need the diskdump tool which accesses the first partition via the "whole disk" device node.
14:17:49hsaliaklinuxstb: it doesnt give me the device node. but i have access to the "whole disk" sd0d .
14:18:18hsaliakso ill use diskdump , i'd have to compile it for netbsd though
14:22:34 Quit lamed ("CGI:IRC (EOF)")
14:23:46 Join szyewei [0] (n=yewei@
14:26:36 Join acespace [0] (n=acespace@
14:27:24 Quit acespace (Remote closed the connection)
14:28:05 Quit szyewei (SendQ exceeded)
14:28:39 Join szyewei [0] (n=yewei@
14:28:39 Quit szyewei (SendQ exceeded)
14:29:20 Join szyewei [0] (n=yewei@
14:30:09 Quit szyewei (SendQ exceeded)
14:30:42 Join szyewei [0] (n=yewei@
14:33:42Tyronemaskellsorry to be a pain but where cani get fonts from
14:34:58rasherThe daily builds page
14:35:25nudel ?
14:35:28Tyronemaskellyou guys know toooooooooooooooo muck
14:35:33Tyronemaskellyou guys know toooooooooooooooo much
14:35:33 Join Criamos [0] (
14:36:23Gneliklinuxstb:if i am using array i need to control free calls..
14:36:34GnelikI can't understand how to implant it
14:40:31bluebrotherTyronemaskell, the manual has answers to about all your questions ;-)
14:41:48linuxstbGnelik: The idea is to rewrite the code so it doesn't need to use free.
14:43:07 Join funky_ [0] (
14:44:49szyeweican the rockbox dualboot on sansa?
14:47:17Gneliklinuxstb: still can;t understand how to implant it, i am to stupid
14:48:23 Join dau [0] (i=daurn@
14:48:25 Quit daurn|laptop (Read error: 131 (Connection reset by peer))
14:48:45barrywardellszyewel; it can, but you need a new version of crt0-pp.S which isn't in CVS yet
14:49:48 Join daurn|laptop [0] (i=daurn@
14:49:49 Quit dau (Read error: 104 (Connection reset by peer))
14:51:00 Quit funky_ ("leaving")
14:51:20barrywardellhere's a working version of crt0-pp.S
14:51:30barrywardellit only builds a working bootloader though
14:51:33barrywardellnot normal build
14:52:20hsaliakhello, is the diskdump mentioned in the ipod nano manual a similar utility to dd? I am trying to install from a BSD. It seems the diskdump is a 3rd party util?
14:53:21linuxstbdiskdump was written (by me) specifically to access ipods from Mac OS to install Rockbox.
14:54:16hsaliakah ok so the 'diskdump' utils i found by googling are some other diskdumps with the same name then
14:54:31hsaliaklinuxstb: i'd be grateful if you could point me to the source
14:54:42linuxstbIt should be on the Mac OS X install page, next to the binary.
14:56:46linuxstbhsaliak: There's a link to it in the table of file attachments at the bottom of that page.
14:57:46hsaliak im following this
14:57:55hsaliaksorry i know im being a bit silly here
14:58:23linuxstbAnd that doesn't mention the source? (tut, tut...)
14:58:30linuxstbThe page I was referring to is here:
15:03:21hsaliaklinuxstb: ok, cool got it and compiled
15:03:46hsaliakthe first partion about 78MB , about right?
15:04:25 Quit Gnelik ("Miranda IM! Smaller, Faster, Easier.")
15:05:27***Saving seen data "./dancer.seen"
15:06:14hsaliakyes it tallies
15:11:28 Quit barrywardell (Read error: 145 (Connection timed out))
15:12:38hsaliaklinuxstb: worked like a charm so far!
15:13:25 Join barrywardell [0] (n=barrywar@
15:15:21 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
15:17:24 Join webguest46 [0] (i=3e9938e6@gateway/web/cgi-irc/
15:18:13linuxstbbluebrother: (and other manual authors) The ipod installation parts of the manual give links to binaries for ipod_fw, ipodpatcher and diskdump, but it doesn't mention that they are GPL'd, or gives directions for obtaining the source. I can understand you not wanting to clutter the manual, but I think that info should be there somewhere.
15:18:23linuxstbAnother questions is if the manual also contain a copy of the GPL and links for obtaining the Rockbox source?
15:18:45 Quit Tyronemaskell ()
15:20:48 Quit webguest46 (Client Quit)
15:23:21n1sWhy would we need to include the gpl in the manual? I do think instructions for getting the source will be nice though?
15:24:11linuxstbIt's the license that lets people use Rockbox...
15:24:28n1sWe do ship COPYING in the zip files...
15:25:55amiconnIsn't the gpl included in the manuals? I remember that this was the case with the old manuals...
15:26:07linuxstbI was just thinking the same...
15:26:33rasherHm, the gfdl is
15:27:38 Quit JdGordon ("Konversation terminated!")
15:28:11bluebrotherin fact I already thought about adding the GPL itself to the manual. We have the GFDL, why not also add the GPL?
15:29:10bluebrotherThat should also make that issue clear, especially if we add a small paragraph that stated the manual to be gfdl and Rockbox itself gpl
15:33:09 Join Genre9mp3 [0] (n=MARVIN_T@rockbox/contributor/Genre9mp3)
15:34:23 Join lee-qid [0] (
15:36:02hsaliakthanks linuxstb i have loaded rockbox onto my ipod nano from netbsd
15:36:11 Join kubiix [0] (
15:36:24 Quit szyewei ("Leaving")
15:38:51linuxstbWhat happened to the old (2.5) manual? The word "manual" on the download page in this sentence "We have a very nice manual written by Christi Scarborough." is a link to the (new) RockboxManual wiki page, with no mention of the old version.
15:40:04 Join CriamosAndy [0] (
15:40:25n1slinuxstb: was there ever a 2.5 manual released?
15:40:38bluebrotherI think the old manual was 2.4 only.
15:40:49bluebrotherbut I suggested putting that to the download page as well
15:40:49linuxstbMaybe not, maybe 2.4 was the last... But in any case, the old manual seems to have vanished. is for 2.1
15:41:16bluebrotheras the current "new" manual is quite different to the old version 2.5
15:42:06linuxstbThere are lots of links to RockboxManual, telling people there are manuals for specific releases there...
15:42:22linuxstbSo maybe just add an "Old manuals" download section on that page.
15:42:30 Quit [sellout] (Read error: 131 (Connection reset by peer))
15:43:38bluebrotherIMO having the old manual on the download page (which covers the latest release) would be good.
15:44:16linuxstbAlthough it seems there wasn't a 2.5 manual...
15:45:31*bluebrother is out for a while
15:47:01Bagderlooks like barrywardell is coming closer to working dual-boot on the h10
15:47:30 Quit Criamos (Read error: 145 (Connection timed out))
15:47:52barrywardellyup, I have it working now :)
15:48:05Bagderthat's very nice news indeed
15:49:51barrywardellso now it works both on Sansa and on H10 :) It would be good to know if the changes to crt0-pp.S work for the iPods too
15:50:18linuxstbWhat did you need to do?
15:53:45barrywardellthat's the version for a bootloader written by dan_a
15:54:25 Quit pondlife ("disconnected has pondlife")
15:54:29barrywardelland there's a version for a normal build
15:54:46 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
15:54:55 Join kubiix [0] (
15:55:22 Join funky_ [0] (
15:55:35 Quit funky_ (Client Quit)
15:56:15barrywardellit just makes sure only the CPU clears the BSS, gives the cpu and cop separate stacks, makes the BL run from IRAM so it's not overwriting in SDRAM and flushing the cache as appropriate
15:56:20barrywardelli think
15:57:17linuxstbThat should all be fine on the ipod. The BL is already copied to IRAM, and I can't see how flushing the cache could hurt.
15:57:24preglowdoesn't sound it differs too much from what we already do
15:57:44linuxstbs/could/will not/ ...
15:58:13linuxstb(ignore that s//....)
15:59:00*linuxstb heads towards the machine generating black liquid
15:59:41 Join perpleXa [0] (n=perpleXa@unaffiliated/perplexa)
16:00:04Bagderyeah, a large stout clears your head ;-)
16:00:42barrywardellit's not all that different
16:01:21*Bagder goes away to cook lasagna
16:01:51*preglow tries forgetting about stouts as quickly as he can
16:02:43 Join half_dead [0] (
16:03:12half_deadany one alive in here
16:03:22Genre9mp3no... we are all dead
16:03:47Genre9mp3you are the only one half_dead ;)
16:03:50 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
16:04:52peturjhMikeS: awake?
16:05:15 Join [sellout] [0] (
16:06:45half_deadcan some one help me finding out th ehitachi default hd password
16:07:38half_deadwhile i search and try out i came onur web site
16:08:44 Join forehead [0] (i=3e088e42@gateway/web/cgi-irc/
16:09:23Genre9mp3petur: forehead seems that came here for a revenge!!! :P
16:09:42preglowwould anyone happen to know if 'inline' also implies 'static' for a function?
16:09:54amiconnpreglow: No it doesn't
16:09:59preglowamiconn: thought so, thanks
16:11:28preglowamiconn: would you also happen to know what is the correct order of keywords? static inline or inline static?
16:11:43preglowdoesn't look like it matters for gcc
16:11:50amiconnShouldn't matter
16:12:19amiconn(but for some reason I would always write static inline)
16:12:41half_deadlet me check there
16:12:50preglowamiconn: yeah, me too, inline static looks weird for some reason i can't quite explain
16:13:07half_deadi tryed that already
16:13:16peturhalf_dead: 'The password is all spaces'
16:13:16preglowprobably just sounds better ;)
16:13:34half_deadall i know is it high protected and i tryed every pass out which i do set none of em works :((
16:13:52half_deadso think maybe default master pass from hitachi could help
16:15:00peturno idea, sorry
16:15:16half_deadIn High security mode, you can unlock the disk with either the user or master password, using the "SECURITY UNLOCK DEVICE" ATA command. There is an attempt limit, normally set to 5, after which you must power cycle or hard-reset the disk before you can attempt again.
16:15:27half_deadhow do i use the ata cmds`
16:15:54half_deadand how do i hard rest it ??
16:16:21*forehead slaps petur
16:16:42*petur didn't see that one coming
16:17:30Genre9mp3forehead's reveng-a!
16:18:13 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
16:18:29 Join miepchen^schlaf [0] (
16:18:31 Join dau [0] (i=daurn@
16:18:48peturGenre9mp3: "Toshiba Gigabeat brunch" ?
16:18:58peturfeeling hungry?
16:19:12amiconnMjamm :)
16:19:22Genre9mp3petur: This paragraph isn't written by me actually! :P
16:19:28barrywardellBagder, linuxstb: what do you think about splitting crt0-pp.S into separate bootloader and non-bootloader files? There are only a few lines not ifdef'ed in the current version (and they're mostly the lines that had to be rearranged.
16:19:44Genre9mp3petur: Also, the sub-headings aren't mine, too
16:21:55 Quit dau (Read error: 104 (Connection reset by peer))
16:22:03 Join daurn|laptop [0] (i=daurn@
16:23:07 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
16:23:10 Join dau [0] (i=daurn@
16:23:29preglowbarrywardell: as long as it leads to very little code duplication, at least it's fine by me
16:23:40half_deadcan any one help ???
16:23:46half_deadfinding out default pass ?
16:23:48 Quit funky (Read error: 104 (Connection reset by peer))
16:25:43 Join funky [0] (n=repulse@unaffiliated/funky)
16:26:40 Quit webguest26 ("CGI:IRC")
16:28:00barrywardelli'll try to combine the two files I have again and see how it looks.
16:28:01linuxstbbarrywardell: I'm in favour as well. I especially #ifdefs with a lot of code inside them (i.e. more than a few lines).
16:28:10linuxstb^especially dislike
16:28:16*linuxstb needs more stout
16:28:46barrywardelli would prefer separate files two. there will be some code duplication, but hopefully not much
16:28:50 Join PaulJam [0] (
16:28:55half_dead A value of 0xFFFE means the Master password is unchanged.
16:29:07half_deadnice so far
16:29:12half_deadmine is there to
16:29:21half_deadi need default pass
16:29:34half_dead3 month search till i came here
16:29:43linuxstbAlso, do we need those STUB tests in crt0-pp ? Isn't that just for the iFP gdb builds?
16:30:40barrywardelli was wondering about that. and also when is the DEBUG stuff used?
16:31:08linuxstbhalf_dead: Locked disks haven't been a Rockbox issue for about 3 or 4 years I think - which is why you're not getting much help...
16:31:44half_deadcan u do ??
16:32:05linuxstbNo, I've no idea... But hang around, maybe someone will appear who can help.
16:33:59linuxstbbarrywardell: Again, probably only on the iFP debug builds.
16:34:52barrywardellthe iFP doesn't use a PP chip, does it?
16:34:59linuxstbNo, a different ARM.
16:40:49markunlinuxstb: I would like to generalise the DAC drivers a bit. I would like to call the functions dac_* because codec_* might be a bit confusing although it is more correct.
16:40:52markunwhat do you think?
16:41:17linuxstbThey are also ADCs...
16:41:25markunyes, I know
16:41:33markunwell, not all of them
16:41:58amiconnadac ;)
16:42:02linuxstbI remember having this conversation before and I think someone suggested audiohw_
16:42:13markunok, I'll go with that
16:42:32linuxstbBut yes, I agree with the idea, and also agree codec_ is confusing.
16:42:34amiconn(German automobile club)
16:42:52markunamiconn: ah yes :)
16:44:11amiconnBut yes, codec is confusing and dac is incorrect, hmm...
16:45:20markunand just audio_ is not very specific
16:46:13barrywardelllinuxstb: DEBUG builds don't even compile for H10 and iPod. and STUB builds are only an option for iFP
16:46:23linuxstbbarrywardell: Yes, that's what I thought.
16:46:37markunlinuxstb: might confuse people with the MAS, don't you think?
16:46:58linuxstbCouldn't we use the same name for MAS functions?
16:47:01amiconnaudiocodec_ , but that's pretty long...
16:47:30linuxstbmarkun: Sorry, you meant confusion with HWCODEC/SWCODEC...
16:47:34half_deadPolice kill N.C. student accused in Playstation 3 robbery
16:48:08 Quit barrywardell ()
16:48:51markunI'll just start. Can always do a search and replace in the patch
16:50:46markunstill no audio on the Gigabeat btw :(
16:52:03 Part thinman
16:53:37 Join Gnelik [0] (n=Miranda@
16:53:47preglowany clues?
16:54:36 Quit thomasg_ (Remote closed the connection)
16:54:50linuxstbmarkun: What's the status apart from audio playback?
16:55:19markunpreglow: not really, but maybe just something stupid I overlooked. someone else will work on it this week.
16:55:50*preglow looks forward to the next gentlemen mail :)
16:56:16*Genre9mp3 wonders about markuns test mp3
16:56:21linuxstbAre you implementing audio in a similar way to the PP targets? i.e. with a FIQ?
16:56:38markunlinuxstb: buttons, backlight, power management, plugins all work very nice
16:56:56linuxstbTried mpegplayer?
16:56:57preglowmarkun: how much is commited?
16:56:59markunwe have some problems with USB but they are almost fixed
16:57:09 Quit dau (Read error: 104 (Connection reset by peer))
16:57:30markunlinuxstb: it gives me a data abort when I play a file > 9MB and just freezes after the 1st frame for smaller files.
16:57:31 Join daurn|laptop [0] (i=daurn@
16:57:39 Quit miepchen^schlaf (Read error: 113 (No route to host))
16:57:45markunWe need to slow down the gameboy emu a bit :)
16:57:59preglowhaha, i forgot about the ridculously fast cpu
16:58:01preglowi want one :>
16:58:11markunperhaps a linear scaling will fix that :)
16:58:45markunpreglow: well, it's not too fast, but there are some things to improve
16:59:46Genre9mp3I suppose rockboy runs realtime on the Gigabeat?
17:00:54linuxstbDoes the gigabeat have IRAM?
17:01:55markunyes, 4KB :)
17:02:02markunpretty useless
17:02:12preglowbut it does have a cache, yes?
17:02:19preglowhave you tested decoding speed yet?
17:02:23markunbut enabling the cache made the gigabeat about 1000x faster
17:02:55markunpreglow: no, trying to play a song just freezes rockbox
17:03:00linuxstbYou could test the mp3/wavpack encoders (if they're still there).
17:03:11markunyes, I will
17:03:11preglowwould be very interesting...
17:04:51 Quit lee-qid ("aufwiederbyebientotsayonara")
17:05:30***Saving seen data "./dancer.seen"
17:08:07markunwav2wv is about 490% realtime. I'll test on my iriver.
17:08:09*forehead slaps petur again
17:08:14 Quit forehead ("CGI:IRC")
17:08:40Gnelikkick and run
17:09:21Gnelik can anyone help me to potr it, i am trying to do this a whole day!!!
17:09:52preglowpetur: gotta start kicking forehead
17:10:11preglowmarkun: well, that's decent...
17:10:19preglowmarkun: what clock do you run at now?
17:10:31 Quit funky (Read error: 110 (Connection timed out))
17:10:49peturpreglow: nah, I think I'll have some fine beer tonight :p
17:11:12markuniriver gives about 195% realtime
17:11:36preglowso will i, and i've got a bottle of very nice chocolate stout waiting for consumption
17:11:53preglowmarkun: definitely decent
17:12:09*petur doesn't like those special fuit and chocolat beers
17:12:22preglowi think arm performs worse at wavpack encoding than coldfire
17:12:30markuna good thing the ipods are so slow. With only gigabeats noone would work on ARM optimisations..
17:13:03linuxstbI hope you're only going to clock it at 75MHz - it's unfair otherwise.
17:13:04peturmarkun: you know of any shop selling F60?
17:13:13preglowpetur: it's around 10% and doesn't really taste much chocolate, heh
17:13:24preglowit does however taste great
17:13:29markunlinuxstb: and make a turbo_boost() function?
17:13:42preglowseveral boost levels!
17:14:40markunalmost every question I see on mygigabeat I want to answer with "just wait for rockbox, it will fix this"
17:14:46linuxstbDoes the Toshiba firmware do anything that needs such a CPU?
17:15:00markunIt has flashing dots in the WPS!
17:15:03zylchesend you details to the fbi?
17:15:31markunlinuxstb: but no, nothing they do needs a fast CPU
17:15:44markunbesides running linux
17:15:57 Join PaulPosition [0] (
17:15:57linuxstbAnd nothing Rockbox does needs it either... (yet)
17:16:11zylcheexcept extreme gaming :P
17:17:19Genre9mp3video playback would be an "easy" task on Gigabeat, though
17:19:40markunDoom froze after "initialising graphics engine" but I didn't debug
17:20:35 Join funky [0] (n=repulse@unaffiliated/funky)
17:25:29 Quit Genre9mp3 ()
17:27:19 Join thomasg [0] (
17:29:31PaulPositionI've seen a few builds using a so-called 'colorised-sudoku' patch/diff.. Of course, it's not on flyspray.. anyone ever seen that somewhere?
17:30:36peturchecked MR?
17:30:51peturor our own forum?
17:31:53PaulPositionYeah,. but no such luck. It's not bad though, will force me to go check in the source and see if it's trivial (or not) to do it..
17:31:57linuxstbI thought amiconn (and/or pixelma) made such a patch?
17:32:27linuxstbI remember testing such a build a while ago.
17:32:40pixelmawell I made coloured graphics for it - I just don't remember how "open" that became
17:33:02PaulPositionIt's said to be Ashen from MR, but seems he hasn't been logging there for a month, and there's no mention of it on their forums.
17:33:21linuxstbPossibly a patch made independently from pixelma's then.
17:33:23 Quit half_dead ("ChatZilla 0.9.61 [Mozilla rv:1.7.12/20050915]")
17:33:42PaulPositionAhh, might well be.
17:34:27pixelmalinuxstb: amiconn asked you about it here - it wasn't put elsewhere
17:36:27 Join barrywardell [0] (
17:37:35PaulPositionpixelma - I'm looking at getting a board where I can see the cells (the grid), as right now I get only one big blank square. I guess it shouldn't be too hard even for a newb.. (well, one could hope!)
17:41:21pixelmasorry.. what did you do and what do you want to do?
17:41:47PaulPositionpixelma - nevermind :)
17:42:00PaulPositionwas more like thinking aloud
17:44:59 Quit _FireFly_ (Read error: 104 (Connection reset by peer))
17:46:52linuxstbPaulPosition: But anyway, whoever distributes those builds should be making the source available...
17:49:29 Nick myzar is now known as myzar|away (n=myzar@
17:49:59*petur resets licking alert
17:54:50 Join pondlife [0] (
17:55:45peturpondlife: are there known seeking bugs?
17:56:19pondlifeAFAIK, only when seeking near the end of a track, and the decoder has moved onto the next track.
17:56:30pondlifeBut I may be out of date
17:56:39peturin that case I found a new one
17:56:58GnelikDo ARM support'sfloating point ariphmetics?
17:58:15linuxstbThe ARMs in the currently support Rockbox players don't.
17:58:55linuxstbI mean there is no hardware support, but gcc will generate software floating-point code for you.
18:00:18 Join Pikcc [0] (i=underwat@
18:00:32Pikccrockbox won't let me shut down my 4g ipod while it's charging, it stays booted up the entire time. any way around it?
18:01:04peturpondlife: in a large WAV (1GB but shouldn't be related) somewhere in the middle playing, I just held 'left' to seek back 10 seconds in the file. Instead of resuming playback it continued quite a bit further in the file. Trying to seek back again seemed to do nothing at all. At one point it even froze.
18:01:31peturI had it several times so it should be fairly reproducable
18:01:56linuxstbPikcc: That's a hardware feature. It needs to be on to charge.
18:02:01 Join Lear [0] (i=chatzill@rockbox/developer/lear)
18:02:21Slasherihehe, that's not even funny, LinusN flashed back my iriver with just the original iriver fw without rockbox bootloader at all :D
18:03:15PikccWill it shut down when fully charged?
18:04:22pondlifepetur: Sounds like an overflow? Maybe try an old version...? Pop it on Flyspray either way...
18:05:01pondlifeI just booted H340 RetailOS and all my LCD settings were odd. Weird colours etc, nothing you can set up intentonally.
18:05:08 Quit Gnelik (Read error: 104 (Connection reset by peer))
18:05:11peturknown issue
18:05:19pondlifeAh, how come?
18:05:31peturrockbox corrupts the eeprom on h3x0
18:05:42pondlifeI'm testing JDGordon's patches, not likely to be that then.
18:05:53peturI even have a good copy on my webspace to hand out to people who get stuck
18:06:18pondlifeI just reset defaults and it seemed ok. Luckily I could read the screen I guess.
18:06:18petursometimes it's so bad that iriver fw refuses to boot
18:06:39LearWhat is the eeprom used for (by Rockbox)? I mean, Rockbox can't just do that by accident, right?
18:06:47peturnothing atm
18:06:49pondlifeDo we have EEPROM writing code at all? Or is it a sideeffect.
18:07:07peturwe do have now but the issue was there before
18:07:19pondlifeOK, so very likely a side effect of some other i/o.
18:07:28LearThere's code for the H1x0 series at least, don't know if it is enabled by default...
18:07:38peturthe only reason h3x0 has eeprom writing code was to help out somebody who got stuck
18:07:54amiconnSlasheri: around?
18:08:21peturthe h3x0 eeprom write code is in the debug menu at the bottom and just writes back a whole file
18:08:22amiconnzylche: asquare? ;P
18:08:47peturbut the eeprom and pcf share the i2c lines
18:09:51amiconnpetur: Thinking about the mysterious eeprom corruption bug?
18:10:10*amiconn has no idea what his original H300 fw does.
18:10:20amiconnDidn't use it for months...
18:10:31peturamiconn: [18:05] <pondlife> I just booted H340 RetailOS and all my LCD settings were odd. Weird colours etc, nothing you can set up intentonally.
18:11:04 Join My_Sic [0] (
18:11:15*petur enters it by accident if he forgets to power on before inserting charger
18:11:49peturand (don't dare to mention it) usbotg
18:12:20peturgtg - back later
18:12:27 Quit petur ("bang")
18:13:32pixelmapondlife: I am testing JdGordon's menu settings patch also on my Ondio and it seems like something prevents writing playlist (and resume position, settings changes) during rebuffer so everything is lost when the batteries went flat before I return to file browser - is there any possibility that you can test whether it is only a problem on Ondio?
18:17:27pondlifeDo you mean if you edit a playlist?
18:18:35Slasheriamiconn: yes
18:19:00Slasheriamiconn: btw, do you think i should commit the rom support for h1xx (just got the unit back, should be working soon)?
18:19:08pondlifeI also went into my OF by pluggin in the charger... The charging screen was in very odd colours!!
18:19:11amiconnSlasheri: I was thinking a bit more about the virtual sector stuff
18:19:38amiconnNow I think that implementing it without increased sector buffers is not only possible, but in fact nearly trivial...
18:19:45pixelmapondlife: no... that happens during normal playback (so resume position won't be right and settings a changed while music was playing)
18:20:02Slasheriamiconn: hmm, that sounds great
18:20:19amiconnWe just need to calculate secmult in fat_mount, and then multiply it into the cluster size and a few other size/position values
18:20:25amiconnAfter that we can forget about it...
18:20:30pondlifepixelma: I haven't noticed that, but may not have tried. Will check.
18:20:41Slasherihmm.. might be working
18:21:33amiconnI will set up a test disk with large virtual sectors (with linux) and try my idea
18:21:43amiconnThen I'll provide a patch for you to test :)
18:21:47Slasherinice :)
18:22:16amiconnWell, you're the only "full dev" (to say so) with a G5.5 ...
18:22:49*linuxstb repeats his call for a 5.5g owner with Windows to test the latest ipodpatcher
18:23:11Slasherilinuxstb: i might be able to test that soon
18:23:21amiconnDoes anyone know how to fetch a certain state from cvs, with multiple files but not all?
18:23:23Slasherinow i am home and have one windows machine too
18:23:51amiconnHmm, probably just cvs up to a sticky date in one working copy and then move the files over...
18:23:58Slasheriamiconn: just fetch the individual single files?
18:24:02Slasherifrom a certain date
18:26:44 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
18:27:07 Join daurn|laptop [0] (i=daurn@
18:28:23 Join nudelin [0] (
18:29:12 Join dau [0] (i=daurn@
18:29:22 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
18:30:39Slasherihmm.. iriver fw says it's H115 version but unit is H140..
18:30:56Slasherimaybe linus accidentally flashed the H110 firmware..
18:31:10Slasheriat least firmware upgrade doesn't seem to work
18:32:06n1sisn't the hex called something different for h110/115?
18:32:41Slasheriyes.. just wondering if it's enough to rename the hex file from ihp_120.hex to ihp_110.hex
18:33:06n1stry it ;-)
18:33:18Slasherihmm, probably not.. need to patch the h120 bootloader with h110 hex file
18:36:34 Quit nudel (Read error: 145 (Connection timed out))
18:37:55barrywardellSlasheri: the usb click of death fix you did for the iPod would be useful on H10 too. I just added ata_sleepnow() to usb_enable(). is that right?
18:37:57amiconnCan't you flash an unpatched ihp_120.hex before?
18:38:28 Join lowlight [0] (i=c730180b@gateway/web/cgi-irc/
18:38:29Slasheribarrywardell: ah, sounds correct. Just add sleep(HZ*2); before the poweroff too
18:38:35Slasheriamiconn: how?
18:38:41*n1s wonders why we build one manual for h110/115 and one for 120/140 when they're the same...
18:38:57amiconnSlasheri: By renaming it to ihp_110.hex ?
18:39:02Slasherii don't think i can flash h120 firmware when the fw thinks it's h110..
18:39:13Slasheribut there is the model flag?
18:39:16Slasheriwell, i will try
18:39:25amiconnIs there? I don't know
18:39:39Slasherii think i saw that in the firmware patcher utility..
18:39:42amiconnBest would be to ask LinusN probably...
18:40:38SlasheriLinus just wanted some challenge so i could brick it again ;D
18:45:27barrywardellok, thanks Slasheri
18:45:28pondlifepixelma: Just tried a skip, skip, shutdown during playback with the settings patch and it remembered my place perfectly. Does that cover your case?
18:45:32Slasheriamiconn: yep, didn't work. There seems to be a model flag
18:45:57preglowso he's given your h140 a new personality, how nice!
18:46:03bluebrothern1s: those manuals are the same.
18:46:16 Join charkins [0] (n=casey@pdpc/supporter/active/charkins)
18:46:22Slasheripreglow: or he wanted to be sure i won't brick it again -> no chances to install rockbox ;)
18:46:37n1sbluebrother: yes that's what i meant but why do we build two of them? (on the manual page)
18:46:38Slasheriwell.. i will now try patching the firmware "incorrectly" and flashing that
18:46:41bluebrotherjust look at the links
18:46:45amiconnThe h140 thinking to be a h115 shouldn't cause major problems, but the optical in/out won't work properly
18:46:58Slasheriamiconn: optical out seems to be always on
18:47:01bluebrotherthere are only two links (to the same file) as it's the same table as for the daily builds
18:47:02Slasheriat least the led is
18:47:10n1soh so we have two links nex to each other pointing at the same thing :-)
18:47:12amiconnIt should be on when you set it to off and vice versa
18:47:19Slasheriah, hehe
18:47:54bluebrotherexactly ;-) Easier than dealing with user complaints because the manuals are the same but the builds different ...
18:48:04pixelmapondlife: not exactly I guess - with stopping playback and shutdown there's no problem - but when the unit shuts itself off because of low battery
18:48:07amiconn(and because of that it probablywon't work at all. The led will be on when you set it to off, but the firmware thinks it's off and doesn't send data)
18:48:12lowlightpondlife: ever get a chance to look at The bug was identified, but I don't know the best solution.
18:49:04 Join TeaSeaLancs [0] (n=thunderc@
18:49:18pixelmapondlife: although it had chances to write it during rebuffering
18:49:30 Quit TeaSeaLancs (Remote closed the connection)
18:49:49 Join TeaSeaLancs [0] (n=thunderc@
18:50:31Slasheriamiconn: ok, now trying to flash h110 hex with h120 bootloader
18:50:59Slasheriat least it does something now :)
18:51:35Slasheriamiconn: worked, i have rockbox again o/
18:52:04Slasherinow iriver feels so much better than ipod..
18:53:26 Join NickDe [0] (n=nicholas@
18:54:29lowlightI've been thinking that maybe playback be easier of the there wasn't a separate codec thread, but rather, the playback engine just calls a function like codec_decode(buf) whenever it needs more data?
18:55:06pondlifelowlight: If we can simplify it then yes, that would be very good
18:55:24pondlifeSorry, haven't actually looked at playback.c in about a month now :(
18:56:01Slasherilowlight: that sounds rather complex.. how codec could handle all events etc.
18:56:13*pondlife thinks there must be a good reason for a seperate codec thread.
18:56:39pondlifeI assume the events would be handled by the audio thread only..
18:57:02Slasherilowlight: hmm, ah.. now i see what you mean
18:57:07lowlightSlasheri: how more complex...let the playback engine handle all events
18:57:28Slasheristill, buffering needs to be done simultaneously with decoding
18:57:42Slasheriand that is the main thing for audio thread to do..
18:58:10pondlifeMaybe the split should be buffering/playback...?
18:58:12lowlightwhere's your new buffering api?
18:58:33Slasherilowlight: haven't had time for that..
18:58:54Slasheriand playback engine seems to have stabilized a bit
18:58:54pixelmapondlife: I don't know but maybe the low battery handling that got commited a while ago prevents such a sudden shutdown on H300 now
18:59:00 Join petur [0] (n=petur@rockbox/developer/petur)
18:59:57lowlightI was thinking that video could also use the playback engine then too...playback would just push some data to the video codec and direct it to the lcd instead of the dsp
19:00:21pondlifepixelma: I've never risked a low battery situation really. I don't know if I'd expect it to save in such a situation, hard disk writing should be avoided if you're about to die.
19:01:35pondlifelowlight: I like the idea, but speak to lostlogic before doing much on it! I'm sure there must be a reason we didn't do this before.
19:02:20pixelmawell I meant that it didn't save during all the rebuffer periods before - but maybe there's no way to test for you because it will always save before a shutdown
19:03:25lowlightpondlife: I'm not doing much more than thinking :)
19:03:28amiconnpondlife: There's an easy way to test sudden shutdown on iriver: just use the reset button
19:04:32 Quit dau (Read error: 104 (Connection reset by peer))
19:04:42 Join daurn|laptop [0] (i=daurn@
19:05:14lowlightamiconn: how does playback work for archos...does the playback engine just "push" data to the decoder?
19:05:32***Saving seen data "./dancer.seen"
19:05:36pondlifeOne reset later - I confirm it does not save track or position!#]
19:05:47amiconnThe MAS signals that it needs data via an interrupt
19:06:02pondlifepixelma: Do you want me to update Flyspray, or have you?
19:06:02amiconnThe isr enabled or disables the dma accordingly
19:06:12thomasgI plan to buy a iAudio X5L DAP. If I install rockbox on it, will I miss the BBE features? And how's the sound of rockbox in comparison to the original fw?
19:06:20amiconnThe end-of-dma interrupt fetches the next chunk
19:06:57Slasheriamiconn: hmm, can i "jmp %d0" or does it need to be "jmp %a0" register?
19:06:57pixelmapondlife: haven't so far - I guess it would be easier for you to explain it (I'm not very good at it today)
19:07:24pondlifeI will do now
19:07:44pixelmathanks :)
19:08:21pondlifeHmm, I may not have had any rebuffering... probably didn't. actually..
19:08:26pondlifeI'll just retest
19:08:30lowlightmarkun: is the gigabeat f10 supported in your port?
19:09:03 Join Gnelik [0] (n=Miranda@
19:09:22*amiconn dissects his recorder to put his test HDD in
19:09:23pondlifeI'm also getting 100% boost..
19:10:10 Quit idnar (Nick collision from services.)
19:10:13 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
19:11:22XavierGrw00t for Slasheri that finally has his H140 working! :D
19:11:38XavierGrSlasheri: You must write a book on your H140 flashing adventures :P
19:11:42SlasheriXavierGr: yep, now i try to be more careful not to brick it again :D
19:11:48Slasherihehe :D
19:12:49 Quit TeaSea (Connection timed out)
19:13:45pondlifeAnyone else testing the settings patch on a target with CPU boost? I find that just starting up leaves CPU boosted,... now to try an unpatched CVS build.
19:14:09 Quit PaulPosition ()
19:14:17 Quit CriamosAndy ("( :: NoNameScript 4.03 :: )")
19:14:29preglowSlasheri: btw, now that you've got an ipod, do you have any ideas on how to improve ui handling?
19:14:47Slasheripreglow: yes, i have two ideas in mind
19:15:00Slasheripreglow: at first need to get the COP working
19:15:33 Join heyJulius [0] (
19:15:45Slasheriand after that UI would have the highest priority on its own core, and when there are multiple wheel events in queue, cpu would get boosted
19:17:06Slasheripreglow: and i seem to get a lot of data aborts and other kind of problems
19:17:16Slasheriproblems i never had with iriver
19:18:03preglowSlasheri: it's no secret that the ipod still lags somewhat behind in stability, no
19:18:13preglowSlasheri: even the cpu freq switching is giving us headaches like that
19:18:21Slasherihehe, true
19:18:33Slasheribtw, is there a key combination to switch on the morse mode on ipod?
19:18:36 Quit ender` (" Top reason why compilers are like women: Miss a period and they go crazy")
19:18:41Slasheriusing that click wheel for typing is horrible :D
19:18:52preglowSlasheri: i have absolutely no idea, i don't use the keyboard often enough too bother learning morse
19:18:59preglowas a matter of fact, i don't use it at all :>
19:19:24Slasherii use it for searching with the database
19:19:44 Join Criamos [0] (
19:20:04preglowah, i don't use the database
19:20:09Slasheriand in general, clickwheel needs serious improvements
19:20:16preglowindeed, no secrets
19:20:25Slasheriit should react to speed of motion, not the amount
19:20:29preglowi'd like some acceleration
19:20:51amiconnSlasheri: Afaik, morse mode is only enabled for irivers
19:20:52preglowi think it should work like it does in retailos
19:20:56preglowit's the one thing apple does right
19:21:00Slasheriamiconn: ah, ok
19:21:17Slasherihehe, i didn't use the retailos at all..
19:21:19amiconnImho it should be there for all targets
19:21:34Slasherijust booted it and patched rockbox.. :)
19:21:43 Quit TeaSeaLancs (Read error: 104 (Connection reset by peer))
19:21:43 Join TeaSea [0] (n=thunderc@
19:21:47*amiconn also didn't use apple retailos on his mini G2 :)
19:21:50preglowit should be there for all targets
19:21:59preglowif i were to use the keyboard, i'd learn morse code for sure
19:22:17amiconnYes, even though it will probably also break rombox
19:22:39amiconnThis is a real feature
19:22:40 Join miepchen^schlaf [0] (
19:22:42preglowi used retailos a tiny bit before rockbox ran on ipods
19:22:52 Join ender` [0] (n=ender@
19:23:08preglowamiconn: do you have a list of specific code that can be optimised for size?
19:23:40amiconnI think there are many places which could benefit from optimisation
19:23:42Slasheriand now i got the whitenoise and all kind of cripplecracks again
19:23:50Slasherineed to reboot ipod to get rid of it
19:23:51 Quit TeaSea (Read error: 104 (Connection reset by peer))
19:23:56preglowwell, sure, but a list would be cool
19:23:59preglowSlasheri: never got that
19:24:05 Join TeaSea [0] (n=thunderc@
19:24:09amiconnBookmarking is one of the first that come to mind
19:24:12Slasheripreglow: i have got that everyday :/
19:24:16preglowSlasheri: how/when?
19:24:26 Quit pondlife ("disconnected has pondlife")
19:24:34Slasherihmm, in fact might be related to crossfade
19:24:40preglowahh, that might explain it
19:24:40preglowi never use that
19:24:47Slasherii got it almost always after skipping tracks
19:24:55*preglow uses perhaps 1% of rockbox, heh
19:25:04amiconnTagcache (ahem, database) might be another, at least Slasheri said so iirc
19:25:17preglowmost of my own code that i run is probably codec based
19:25:46preglowthe code i like to deal with isn't even included in rombox
19:26:04amiconnThe resampler is, on H1x0
19:26:29amiconnBut that one won't break any time soon
19:26:53 Join mirak [0] (
19:26:54preglowah, i don't think of the h1x0 one as rombox
19:26:55*amiconn suddenly wants 512KB ROMs ...
19:27:11Gnelikdouble i;
19:27:32Gnelikbut it rounds
19:27:49amiconnTry 1.0 instead of 1
19:28:16petureuh? we don't do float, don't we?
19:28:24preglowthen plot.x_mod is an integer
19:28:33preglowpetur: i think it can be excused in a function plotter
19:28:43amiconnpetur: Not in the core. Plugins can (but preferably shouldn't)
19:28:46peturah well
19:28:46preglowdoing that in fixed point is a bloody nightmare
19:29:35Gnelikamiconn: Tnx. It works.
19:30:13GnelikIs float do much speed dec.
19:30:23preglowGnelik: oh boy, yes
19:30:33preglowGnelik: but unless you're insane, you don't have much of a choice
19:30:46GnelikIn %?
19:30:53preglowmany thousand, probably
19:31:07Gnelikho ho
19:31:09preglowfloating point is emulated in software
19:31:18Gneliki know
19:31:24preglowit's _very_ slow
19:31:38preglowbut like i said, for a function plotter, just use it
19:31:43Gnelikbut if i need it
19:31:53Gnelikmust i implant it myself
19:32:05Gnelikusing int?
19:32:19Gnelikbut it won't be to fast alsow
19:32:29preglowbut keep in mind that pow/sqrt/etc does not exist
19:32:38preglowyou pretty much only get +, -, / and *
19:32:58Gneliki have downloaded math lib
19:32:59amiconnDoesn't gcc provide a math lib?
19:33:04Gnelikin C
19:33:08 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
19:33:20preglowamiconn: probably not impossible, but i don't know if it's built by default
19:33:24 Join _FireFly_ [0] (
19:33:37amiconnPerhaps we need a linker option
19:33:39Gnelikit implant all this but i dont know if it will work
19:33:40preglowamiconn: libm is what's supposed to be used anyway, and that's not bundled with gcc
19:33:51amiconnOh, it's not?
19:34:06 Join daurn|laptop [0] (i=daurn@
19:34:26*amiconn didn't dela with such stuff so far
19:38:44preglowwell, i don't think it is
19:38:49preglowi'm not exactly an expert in the field myself
19:38:59preglowthe only time i've used libm has been on computers
19:40:13amiconnSlasheri: Using my simple idea shaves off more than 1KB of code on SH1. Now I hope it will work...
19:43:44 Quit barrywardell (Read error: 110 (Connection timed out))
19:49:23Slasheriamiconn: hmm, interesting
19:50:03*linuxstb spots libm.a in /usr/local/arm-elf/lib/ - so it must be part of either binutils or gcc...
19:50:29 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
19:51:01*linuxstb was looking at the wrong ARM compiler and should be ignored
19:51:24 Quit Criamos ("( :: NoNameScript 4.03 :: )")
19:52:12amiconnWow, is less than half the size of
19:52:19amiconn(for recorder that is)
20:00:09*amiconn installs dosfstools
20:01:06*Slasheri verifies bootloader now many times before flashing on iriver :)
20:01:15 Quit TeaSea (Read error: 110 (Connection timed out))
20:01:21 Quit My_Sic (Read error: 110 (Connection timed out))
20:02:22Slasheribtw, rockbox logo looks great on black background on ipod
20:03:08Slasheriblack ipod, dark screen and blue texts :)
20:04:59 Join TeaSea [0] (n=thunderc@
20:05:30 Join My_Sic [0] (
20:10:19 Quit zylche (Read error: 104 (Connection reset by peer))
20:10:26amiconnmkfs.vfat -S 2048 /dev/sdc1
20:10:40amiconncopying contents...
20:12:03jhMikeSman, wtf is with Cowon. they give a number for obtaining service and then just refer you to the website with an automated message and hang up. bastards.
20:14:25 Quit secleinteer (Excess Flood)
20:14:28 Quit shnee (Excess Flood)
20:14:41 Join secleinteer [0] (
20:14:51 Join zylche [0] (
20:15:15amiconnW00t! Seems to work at the first try :D
20:15:30 Join shnee [0] (
20:17:48bluebrotheramiconn: shouldn't you have used -F 32 also for mkfs.vfat? At least for me this is still required to get FAT32
20:18:05amiconnNot if the partition is large enough
20:18:13amiconnI use a 10GB disk for testing
20:18:22amiconn(single partition)
20:18:52n1samiconn: does the larger sectors affect performance?
20:19:10amiconnThe way I implemented them now: not at all
20:20:19bluebrotherhmm. I once tried that on a large disk and it created FAT16 for me, so I was only able accessing ~2GiB
20:20:39amiconnOnce the partition is mounted, everything works with 512 byte physical sectors
20:20:54 Join xamox [0] (
20:20:55amiconnbluebrother: Maybe that was an old mkfs.vfat
20:21:22KasperleSlasheri: :) i'm using rockbox and the "Black" wps on my black ipod
20:21:42xamoxIs there a way to reset my rockbox. For some reason I can copy files to/from my rockbox on my iPod nano, but for some reason when I try to play a song it locks up and reboots. It didn't do this at first and now for some reason it's doing it
20:21:56 Quit |[Gino]| (Read error: 110 (Connection timed out))
20:24:07 Quit Gnelik (Read error: 110 (Connection timed out))
20:24:28 Join |[Gino]| [0] (
20:24:48bluebrothermaybe. I was still under the impression it's still needed.
20:25:13bluebrotherxamox: you can reset the settings by enabling the hold switch upon boot
20:25:40xamoxbluebrother: alright, thx I'll give it a try.
20:25:40rasher Specifies the type of file allocation tables used (12, 16 or 32 bit). If nothing is specified, mkdosfs will
20:25:43rasher automatically select between 12 and 16 bit, whatever fits better for the filesystem size. 32 bit FAT (FAT32
20:25:47rasher format) must (still) be selected explicitly if you want it.
20:26:00rasherSorry about the extra spaces
20:26:33xamoxbluebrother: awesome, that fixed it thx!
20:27:01rasherCould be the manpage is inaccurate. Wouldn't be the first time that happened.
20:27:32 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
20:27:40amiconn_rasher: My mkfs.vfat auto-selected fat32...
20:27:50 Quit amiconn (Nick collision from services.)
20:27:51 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
20:28:03bluebrotherrasher: that was why I was still under the impression it's needed
20:28:22rasheramiconn: Must be smarter than the manpage thinks
20:28:44amiconnI am running debian testing/unstable, up to date
20:29:08amiconnStill not really new:
20:29:11rashertesting or unstable?
20:29:18amiconnA mix
20:29:29rasherI see.
20:29:36amiconnmkfs.vfat -v −−> mkfs.vfat 2.11 (12 Mar 2005)
20:30:29amiconnMy man page says the same as yours.
20:30:30|[Gino]|<New question>Does the microphone on the X5 work with Rockbox and can you still record from the line in on the subpack?
20:30:47rasherNewest there is.
20:31:05rasherThe Debian packager is also upstream maintainer it seems.
20:33:01 Join dan_a [0] (n=dan-mirc@
20:33:52 Quit dan_a (Client Quit)
20:34:29 Join PaulPosition [0] (
20:38:13 Quit _FireFly_ ("Leaving")
20:40:23 Join TeaSeaLancs [0] (n=thunderc@
20:40:39 Quit zylche (Remote closed the connection)
20:41:00 Join zylche [0] (
20:42:22 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
20:42:43 Join kubiix [0] (
20:46:12 Quit TeaSea (Read error: 60 (Operation timed out))
20:47:12bluebrotherstrage. I'm just playing a ogg vorbis file and my wps tells it's flac :o
20:48:10peturtalk about compression ;)
20:48:29rasherOr a wps that hasn't been updated to the new codec order.
20:48:38rasher(there's a new codec order, right?)
20:49:05 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
20:49:10 Join dau [0] (i=daurn@
20:53:22 Join webguest88 [0] (i=55101677@gateway/web/cgi-irc/
21:01:51amiconnHehe, windows doesn't like virtual sectors different from physical sectors
21:02:57amiconn"The disk in drive G isn't formatted"
21:03:15amiconnIt is formatted, and linux doesn't have problems reading / writing it
21:04:04 Quit TeaSeaLancs (Read error: 104 (Connection reset by peer))
21:04:07 Join TeaSeaLancs [0] (n=thunderc@
21:04:12bluebrotherrasher: that might be the cause. Haven't noticed before ... will check later
21:05:01 Part xamox ("")
21:05:01 Quit God_Eater (Read error: 104 (Connection reset by peer))
21:05:36***Saving seen data "./dancer.seen"
21:10:27jhMikeSpetur: the glitch is exactly 41 chunks long so I can see it's related to encoder output OR DMA input. Did you have prerecord on? I'd like to know the exact sample indexes these occurred at.
21:10:52peturpre-rec = 30s
21:11:26peturI'll open the original and hunt for the exact position
21:11:35 Join Cassandra [0] (
21:11:41jhMikeSThat might tell me if it's related to buffer wrapping or what not
21:12:01*linuxstb spots a stranger returning
21:12:02 Join unleet [0] (i=185c8710@gateway/web/cgi-irc/
21:12:11peturyou saw my last comment in the tracker?
21:12:13Cassandrabagder - you around - I carelessly forgot my CVS password
21:12:17unleethey Paul.
21:12:19peturheya Cassandra
21:12:23*bluebrother decides to finally leave his working desk at uni
21:12:36Cassandralinuxstb, heya. ;) I was never really gone in spirit.:)
21:12:38rasherCassandra: Did you get the mail I forwarded to you
21:13:18unleetanybody have a clue why I cant access rockbox settings on my H10 20gb?
21:13:22Cassandrarasher - yeah.
21:13:34unleeti click "rockbox" and it does nothing
21:13:42Cassandraas it happened, I've been playing about with the installer again.
21:13:43amiconnHi Cassandra :)
21:13:48bluebrotherunleet: you never click "rockbox"
21:13:53amiconnAny news from AT&T?
21:13:55CassandraIt's still not what I'd call finished, but it works.
21:14:03bluebrotherthat's the firmware itself and it will reload rocbox
21:14:05unleetSo, that does nothing?
21:14:36bluebrotherit reboots (but I don'
21:14:36Cassandraamiconn - nah. They totally ignored the letter. After I spend 5 quid sending it inernational recorded too. :(
21:14:46bluebrothert know if that's still working on h10)
21:14:53bluebrotherthis is intended behaviour
21:15:33bluebrotherunleet: please have a look at the manual. There is a quick start section which should answer most usual problems / questions
21:15:43 Join daurn|laptop [0] (i=daurn@
21:15:44 Quit dau (Read error: 104 (Connection reset by peer))
21:15:49jhMikeSmemcpy can take 0 for a size right?
21:16:03unleetOhh. Very sorry.
21:16:38*bluebrother is in the process of leaving
21:17:11 Quit bluebrother ("back when arrived at home")
21:17:21amiconnSlasheri: Ready to test something? :)
21:17:40peturjhMikeS: do you have enough with one glitch location or what do you need?
21:18:39jhMikeSI know both glitch locations in the sample. Would like to know them from the start of the recording (0 based)
21:18:53peturok, I get those two
21:18:59linuxstbjhMikeS: You could check the source in firmware/target/coldfire/memcpy-coldfire.S
21:19:17 Quit kubiix ("Miranda IM! Smaller, Faster, Easier.")
21:19:24 Join kubiix [0] (
21:19:24 Join lee-qid [0] (
21:19:37jhMikeSlinuxstb: thanks. I'll have to make sure of that since 0 can be passed during record
21:19:47Slasheriamiconn: hehe, testing soon :) ipod just charging (battery totally empty)
21:20:10amiconnI moved the definition of MAX_SECTOR_SIZE to config-ipodvideo.h
21:20:32amiconndisk.c is changed a bit against cvs to try all powers of 2 from 512 to MAX_SECTOR_SIZE
21:21:21amiconnfile.c and fat.h are back to before the commit, as is fat.c, with the added changes
21:21:31 Quit lowlight ("CGI:IRC (EOF)")
21:21:40amiconnWorks like a charm with fat32 on recorder, 2048 byte virtual sector size
21:21:51amiconnDidn't test fat16 yet, but will do
21:21:51jhMikeSpetur: and as a reference what's the total length of the recording in samples?
21:22:26amiconnSlasheri: Btw, did you check where the config sector ends up? Shouldn't be a dangerous place, but better check...
21:23:03amiconnHopefully the config sector will be gone soon (as soon as the remaining problems in JdGordon's patch are fixed)
21:23:23peturjust a sec, I'm still figuring out were the glitches are. I can hear them but finding the sample boundary isn't that easy ;)
21:23:40jhMikeSpetur: in your clip, the first is at 617672 and the second at 701640
21:23:56unleetLol my rockbox is stuck in a plugin
21:24:19peturthat won't help, I don't know at what offset I took the clip (was at work in a hurry)
21:24:39 Quit TeaSeaLancs (Read error: 110 (Connection timed out))
21:24:43peturunleet: correction: you are stuck in a plugin
21:24:57peturunless it crashed
21:25:13*linuxstb is often stuck in the puzzle plugins
21:25:22unleetNone of the buttons are exiting the plugin. I shall reset.
21:25:23amiconnpetur: Hmm, is there a tool that can compare 2 files and find the place where the smaller one was cut from the bigger one?
21:25:44amiconnsomething like strstr() as a tool
21:25:51peturnever heard of...
21:26:30 Join lowlight [0] (i=c730190b@gateway/web/cgi-irc/
21:27:08 Join Arathis [0] (
21:27:56 Join God_Eater [0] (
21:29:17Slasheriamiconn: hmm, looks interesting. but i will test it tomorrow, need to go sleeping soon :/
21:29:28amiconnHmm :/
21:29:32linuxstbBagder: Can you put in and also extract ipodpatcher.exe from the zip file and put that in the same place (replacing the previous version).
21:29:36*amiconn wants rombox on Ondio FM back
21:29:49Slasheriamiconn: well, i can give a quick try to it
21:29:53Slasherijust to see if it works
21:30:09Slasheribut don't have time to do real testing now :/
21:30:34Slasheripatch applied fine, compiling now
21:31:04mick_linuxrasher: no email yet :(
21:31:17rashermick_linux: It's Cassandra who just joined
21:31:29pixelmaunleet: what plugin? (maybe you can find the button mappings in the manual, too - though I have to admit that the H10 manual is a bit incomplete yet)
21:31:44 Join Lars_G [0] (n=lars@unaffiliated/lars-g/x-000001)
21:31:50amiconnSlasheri: I'll do some testing with fat16 as well, both standard and non-standard sector size...
21:32:07Lars_GWhat woudl you dudes think if I said I'd like to find a coherent way to mash the three views together? (playlist, browser, tag database)
21:32:23 Join Rob2222 [0] (
21:32:47linuxstbSounds desirable to me.
21:32:49*petur thinks virtual directories again
21:32:58markunI have sound!!
21:33:14linuxstbReal, proper, continuous sound?
21:33:31*linuxstb goes to fridge to celebrate
21:33:31markunwhite noise, but I used a white noise generator :)
21:34:25*petur checks e-bay again for an F60
21:34:27*amiconn is listening to music from his 2048-byte-virtual-sector-recorder for quite a while now
21:34:34linuxstbI think you need music playback for a gentlemen (and ladies) email though...
21:34:37Lars_GBtw little question, what is different in a m4b over a m4a file? afaik DRMed files are .m4s right?
21:34:50markunlinuxstb: yes, but I was too excited to keep it to myself :)
21:35:04courtcchar *strstr(const char *haystack, const char *needle) { const char *h, *n; do { while (*haystack && *haystack != *needle) ++haystack; if (*haystack) { h = haystack; n = needle; while (*(++n) && *(++h) && *h == *n); if (!*n) return haystack; } } while (*haystack); return 0; }
21:35:12*courtc shrugs
21:35:21lowlightmarkun: is the F10 the same hw? can the hd be swapped?
21:35:35courtccoded in irc, no warranty.
21:36:04mick_linuxrasher: where? nick?
21:36:09rashermick_linux: Cassandra
21:37:16linuxstbLars_G: I think m4p is protected, m4b is an audiobook and m4a is normal. I don't know what Apple do differently with m4b files though.
21:37:28markunlowlight: yes, but because it's single platter that max size is limited
21:37:58Lars_Glinuxstb: Nod, I'll see if ffmpeg has any docs on the differences, I wonder if we/I can make m3b work on the rockbox
21:38:09lowlightsize of the case?
21:38:15mick_linuxrasher: oh, duh :)
21:38:27*mick_linux waves to Cassandra
21:38:53LearIsn't m4b "enchanced" AAC, with bookmarks and such? And m4p is protected?
21:38:53 Quit lowlight ("CGI:IRC (EOF)")
21:38:56Slasheriamiconn: works fine :)
21:38:59Lars_Goh Cassandra, I wonder wether it's the singer or the witch
21:39:01amiconnNice :)
21:39:01Slasheriat least audio plays and so on
21:39:11Slasherii think you can commit it :)
21:39:21Lars_GLear: You must be correct, since this m4b comes from a "podcast" with marks in it.
21:39:23amiconnSaving .plylist_control etc should also work
21:39:32amiconn* .playlist_control
21:39:36Slasheriamiconn: yep, works
21:39:44 Join lowlight [0] (i=c730190b@gateway/web/cgi-irc/
21:39:53LearLars_g: Then it should play in Rockbox, as long as it isn't too long... :)
21:40:00 Quit Rob222241 (Read error: 145 (Connection timed out))
21:40:12amiconnI added a note for struct bpb (locally), as it no longer holds raw values if bytspersec != 512
21:40:12Slasheriamiconn: nice that you spot it was trivial to implement ;)
21:40:13Lars_Gif i don't get track numbers in the title when I browse by artist but I do when I browse by album, I guess I could fix it on my naviconf.cfg file right?
21:40:20Slasherii wouldn't have thought that
21:40:35Lars_GLear: it doesn't it bails with a "codec" error, and I even had to rename it on player for it to be recognized.
21:40:44amiconnI think it's so little code now that all targets can keep it
21:41:08amiconnThis way someone could make his rockbox target appear unformatted to windows ;)
21:41:09LearLars_g: Likely too long then. Like 75 minutes or more?
21:41:27*n1s testbuilds manuals
21:41:27amiconnI'll still perform the fat16 tests
21:42:30Lars_GLear: 46 minutes or so
21:42:33amiconnSlasheri: Sometimes I need to be beaten to implement something. If it's worth the effort, I might come up with a clever solution ;)
21:42:49peturjhMikeS: what tool did you use to get the samples position? audacity won't tell me now that I'm on the spot it's 1:34:17.470 for point 1
21:42:56 Join bluebrother [0] (i=O9hAFjMq@rockbox/staff/bluebrother)
21:43:17lowlightany velvet underground fans?...
21:43:28LearLars_g: hm... 46 minutes should be fine. You do need a recent build for it though.
21:43:37linuxstblowlight: Yes, I heard about that....
21:43:51Lars_GLear: I thought I had it... well I have to build a patched build anyhow so I'll take the opportunity to do it
21:44:03Slasheriamiconn: hehe :)
21:44:16jhMikeSpetur: cool edit pro. I can work in sample indexes and zoom to sample level to see the glitches
21:44:34amiconnI'll look into coldfire ata dma next. Might have a nice effect (aligned access only, unfortunately)
21:44:38lowlightlinuxstb: maybe a copy will show up on dime ;)
21:45:00LearLars_g: well, if you haven't updated in the last 24 hours, it isn't new enough...
21:45:03 Quit MarcoPolo ("Bye !")
21:45:16CassandraI think I missed something again.
21:45:18Lars_GLear: Well wait for me I don't have an ARM cross yet
21:45:26CassandraThis channel is too fast moving for my old brain.
21:45:40CassandraDid someone just call me a witch, or what? ;)
21:45:42markunlowlight: yes, because of the size of the case
21:45:42Lars_GCassandra: old brains are better!
21:46:00markunlowlight: but the F40 is relatively cheap
21:46:03CassandraLars_G, well, it's the only one I've got.
21:46:05Lars_GCassandra: that was me, I wondered whether it was Cassandra Willson the singer, or Cassandra the famous witch
21:46:21peturjhMikeS: position 2 = 1:34:19.120 don't know if that helps
21:46:22CassandraProphetess, really.
21:46:24linuxstbmarkun: Is the F series still being manufactured?
21:46:30CassandraAnd yeah, it's after her.
21:46:53Lars_GCassandra: Nice. And yes you're right prophetess.
21:47:04Lars_GCassandra: She was the one doomed never to be taken seriously, wasn't she?
21:47:16markunlinuxstb: I don't think so
21:47:17CassandraI prophesy a new version of the much delayed Rockbox installer.
21:47:26jhMikeSpetur: So, prerecording was full when you started it?
21:47:39peturthat I cannot tell
21:47:41CassandraLars_G, yes. Don't f*** with Apollo. He's a right bastard.
21:47:48Lars_GCassandra: meh what for? it's still easier to install than Ipod Linux
21:48:02jhMikeSand what does "29 fragments" mean if there's nothing audible?
21:48:16Lars_GCassandra: Most of them where bastards... I remember the grief triton gave to poor odysseus while going home
21:48:16CassandraCome to think of it, not f***ing with Apollo was what got her into trouble in the first place, so maybe that's not a great idea.
21:48:22 Quit lowlight ("CGI:IRC (EOF)")
21:48:28markunlinuxstb: always the same story with rockbox.. finished by the time the player is out of production.
21:48:31rasherCassandra: I have added a personal wishlist on DesktopTools, which I made ages ago.
21:48:48peturjhMikeS: fragments as in file fragments on disk (relating to disk fragmentation
21:48:54Cassandrarasger: I
21:48:55*amiconn builds rockbox for recorder with fat16 support
21:49:00XavierGrmarkun: are we going to see the usuall "Gentlemen we have sound" mail anytime soon? :D
21:49:13Cassandrarasher, I'll check it out, but I'm done for today. Spent most of today working on it.
21:49:14peturjhMikeS: pre-rec was probably full
21:49:29amiconnmarkun: Not entirely true... the X5 is still being manufactured afaik
21:49:39markunamiconn: yes, also the ipods, right?
21:49:43Lars_GAnyhow thanks for rockbox
21:49:46CassandraI hate uploading stuff to the wiki. You never know if it's working/
21:49:56linuxstbAlthough we now only support one of the current ipods...
21:50:02markunXavierGr: probably
21:50:11amiconnBut only the G5.5 and Nano G2
21:50:13Lars_GBtw I also bought a Griffin itrip for the nano, I wonder how hard it'd be to reverse eng and implement the serial driver for the CPU and the protocol for remote control once and for all....
21:50:23amiconn...and we only support the G5.5/30 so far
21:50:24Lars_GMaybe if I can put a sniffer in between the devices... hmmmm
21:50:30jhMikeSpetur: I should be able to round those number to chunk indexes
21:50:54*petur doesn't slap forehead this time, but audacity
21:50:59linuxstbLars_G: The protocol is more or less known (see the IPL wiki page). The problem is the serial driver.
21:51:28Lars_Glinuxstb: Yes, the odd part is that if this is a common arm arch the serial should be known, unless they're using an off-chip driver... hmmm
21:52:10*amiconn now needs to install rockbox via linux, as windows doesn't like the partition
21:52:30Lars_Gamiconn: ??? using mac type partition?
21:52:55Lars_GI personally preffer FAT32 with DOS type partition, since it allows me to use the ipod in linux, mac and windows
21:53:19 Quit God_Eater (Read error: 110 (Connection timed out))
21:53:26amiconnFAT32 using 2048 bytes per virtual sector
21:53:44amiconnTesting code intended for G5.5, and I don't have one myself
21:53:56 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
21:53:58*petur formats his h340 to defrag it
21:54:09 Join God_Eater [0] (
21:54:13 Join daurn|laptop [0] (i=daurn@
21:54:19amiconnpetur: Why not defrag it for real?
21:54:38peturtoo slow
21:55:22jhMikeShmmm those times round to a 35-37 chunk span, not 41.
21:55:52 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
21:56:06preglowamiconn: also, defrag seldom completes with a completely defragmented drive
21:56:06 Join daurn|laptop [0] (i=daurn@
21:56:16 Join midkay [0] (n=midkay@rockbox/developer/midkay)
21:56:34peturformat + copy again goes faster and with better success, imho
21:56:35 Quit NickDe ("Leaving")
21:56:40 Join MarcoPolo [0] (
21:57:02jhMikeSpetur: don't suppose I could get a copy of the recording in total could I?
21:57:46peturjhMikeS: 1.02 GB :(
21:58:14peturand first glitch is only at 42 minutes
21:58:58jhMikeSok everything from the start to the last glitch would be enough. maybe flac encode or even ogg and it should hurt it.
22:00:10peturwhat tool shall I get to encode to flac (windows)? <−− /me lazy
22:00:33 Quit mirak (Remote closed the connection)
22:01:08jhMikeSI just have the command line tool downloaded from the site
22:01:25peturmono is ok?
22:01:29petursaves 50%
22:01:58jhMikeSlemme check
22:02:26amiconnHey, what's a gig? ;)
22:02:48jhMikeSyeah, actually makes the glitches more visible
22:03:55 Join dau [0] (i=daurn@
22:03:56 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
22:05:33 Quit Lear ("Chatzilla 0.9.75 [Firefox 2.0/2006101023]")
22:05:34jhMikeSI have to run to the post office, be back in an hour or so.
22:06:23PaulPositionAnyone knows where does the api 'functions' ->lcd_hline() and lcd_vline() take their colour from? Can't understand why sudoku would draw me white lines on a white board making the grid invisible...
22:07:03amiconnThey use the foreground colour the same way as all other graphics primitives
22:07:30 Join merbanan [0] (
22:07:42 Join lowlight [0] (i=c730180b@gateway/web/cgi-irc/
22:07:53PaulPositionstill, the theme's config has them set as Fore-FFFFFF and back-000000...
22:07:53 Quit webguest88 ("CGI:IRC (EOF)")
22:08:53n1sset them your self in the plugin
22:09:50lowlightjhMikeS: wanna fix a playback.c error?
22:09:53PaulPositionn1s - Hmmm.. I'm not very familiar with variable's scope in C...
22:10:06PaulPositionBut I'll have a look, thanks.
22:10:32lowlightjhMikeS: line 3562, the "else" is #if'ed out
22:11:27jhMikeSlowlight: ok, will have a look when I get back, gotta run
22:11:53 Quit dau (Read error: 104 (Connection reset by peer))
22:11:59 Join daurn|laptop [0] (i=daurn@
22:12:55n1sPaulPosition: rb->lcd_set_foreground(COLOR_BLACK);
22:13:19n1sor maybe LCD_BLACK
22:14:10PaulPositionn1s - Thanks, I'll try that... Is there any way to get back to theme's default on exit? Or is it automagic?
22:16:57lowlightjhMikeS: ok...actually it would be clearer if just the inner "if ( ( == SYS_TIMEOUT)..." statement was #if'ed
22:17:13amiconnPaulPosition: It's automagic
22:17:29PaulPositionamiconn - Thanks for the tip. :)
22:17:58*amiconn now listens to music from a 2048-byte-vsector FAT16 partition
22:19:08 Join Cassandra- [0] (
22:19:27*Slasheri runs advanced rom loader bootloader on h140 without bricking it :)
22:19:32Slasherinow a good time to go sleeping
22:19:37amiconnHehe :)
22:19:47Slasheritomorrow it's hard to choose which one player to pick up with :)
22:19:52amiconnExpect a commit while you're sleeping :)
22:19:58Slasherihehe :)
22:22:30 Join bluey- [0] (
22:23:16 Quit |[Gino]| (Read error: 110 (Connection timed out))
22:24:54 Quit midkay_ (Read error: 110 (Connection timed out))
22:28:51Cassandra-Install till your little hearts burst.
22:28:57Cassandra-I expect it's full of bug.
22:29:06unleetIve searched and cannot find backdrops for download. Any suggestions?
22:29:07Cassandra-Bugs even.
22:29:32Cassandra-Get better at searching? Stop looking?
22:29:42XavierGrunleet: make some on your own
22:29:47Cassandra-Or did you want *helpful* suggestions?
22:29:56XavierGrunleet: it is quite easy
22:30:11XavierGrunleet: just find a picture you like and crop it on your player's resolution
22:30:19unleetI tried but the I cannot access my music. Ill read some documentation.
22:30:29 Join webguest89 [0] (i=513efce3@gateway/web/cgi-irc/
22:30:29*petur wonders if it's wise to put exe's in the wiki
22:30:51dionoeaunleet: are you sure that you're setting the bmp as backdrop and not editing it with some plugin ?
22:30:59Cassandra-petur: Maybe not.
22:31:25lowlightadd with an md5sum
22:31:37unleetIt shows the picture but I cannot see anything else.
22:31:41 Join TommyTott [0] (
22:32:04dionoeaunleet: if you press up once and down once after setting the picture do you see a toolbar ?
22:32:08dionoea(just to make sure)
22:33:06rasherCassandra-: I see I can still select Rockbox Stable version by navigating with the arrow-buttons
22:33:32pixelmaunleet: did you choose "set as backdrop" or did you only "click" on the bmp
22:34:26unleetset as backdrop
22:34:46TommyTottmy ipod is dead.. 2 days ago, i couldnt start it, it just showed the " :( " picture, but now it wont start at all.. help..
22:34:57rasherCharge it.
22:35:24dionoeaTommyTott: did you try plugin it to some charger / usb ?
22:35:30unleetDownload the ipod firmware back to it.
22:35:32 Quit Cassandra (Read error: 110 (Connection timed out))
22:35:33 Nick Cassandra- is now known as Cassandra (
22:35:42Cassandrarasher - hmm - not so good.
22:35:49TommyTottbut, if i have the charger in, shouldt it start?? cause i have it in charger now, but nothing happens
22:36:19TommyTottdo it have to be in the charger for long time if the battery is flat before something will happen??
22:36:23 Join dau [0] (i=daurn@
22:36:29 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
22:36:31unleetCan you send the old ipod firmware back to it?
22:36:44TommyTottnow it wont start at all
22:36:47dionoeaTommyTott: usually 5 to 10 minutes i'd guess (well you still see something on the screen though)
22:36:52dionoeawhen charging using usb
22:37:02TommyTotti now charge inn wall
22:37:46unleetWhen charging with usb, do you see a plug icon?
22:38:02TommyTottno. not now.. 2 days ago, i did
22:38:26 Join Cassandra- [0] (
22:38:31unleetYour battery is dead then.
22:39:22TommyTottits little used.. had it for some time, but not used so much..
22:39:45TommyTottbut it has been a little while since last time
22:39:50dionoeatry keeping it plugged in for like 1 full day
22:40:00dionoeathen you'll know for sure
22:40:35TommyTottif nothing happens, i should know in one hour??
22:40:41unleetpixelma my backdrop works now.
22:41:01unleetThe bg color was black, as in text black so i saw nothing
22:42:29 Quit webguest89 ("CGI:IRC (EOF)")
22:43:24unleetProbably an hour is best
22:43:47unleetIf nothing happens, the battery is most likely dead.
22:43:53 Join mirak [0] (
22:45:00TommyTottthe last post here, does this include ipod also??
22:45:14unleetLet me have a look.
22:47:16unleetThat is a very good idea. If you could charge it seperate it might do something.
22:48:03TommyTottoki.. but i dont know how to do that, so.. dont even know how to open the ipod..
22:48:25unleetLook in the forums under the section "Hardware"
22:48:30TommyTotti know theres problably an tutorial on the net
22:48:31unleetMight help.
22:48:37 Quit Cassandra- (" HydraIRC -> <- IRC with a difference")
22:48:54unleetShould find something
22:49:47 Quit unleet ("CGI:IRC (EOF)")
22:50:02 Join unleet [0] (i=185c8710@gateway/web/cgi-irc/
22:50:03preglowanything else left for proper 5.5g support now?
22:50:13unleetI may have found you something.
22:50:17amiconnYes, figuring the 80GB ata issues
22:50:24amiconn*figuring out
22:51:10amiconnBut for the 30GB model I would simply say it's supported at the same level as the old G5
22:51:11unleetThis has pictures on how to replace the ipod battery
22:51:34bluebrotherrasher, what do you think about (a) hiding the rockbox.{iriver,ipod,...} file in the "supported" view or (b) making the "music" view the default?
22:52:00amiconnpreglow: That's all that was needed:
22:52:13amiconn(plus the multiple mount checks in disk.c)
22:52:36dionoeabluebrother: you should only hide the default one (in case people want to load another file named rockbox.ipod stored in another directory)
22:52:44unleetNot guarunteeing success. Thats just to replace the battery
22:52:46dionoeaor even better, hide only the loaded one (if that's possible)
22:52:57bluebrotherhmm. That's an idea.
22:53:18bluebrotherit's mainly the problem users tend to "click" on that "rockbox folder"
22:53:29TommyTottunleet: i see.. dont have any powersource to charge it like that, so..
22:53:47bluebrotherbut as the database view also only shows music files I thought about simply making the music files view the default
22:53:56unleetI bet you could either buy one or........
22:54:18unleetDepending on the battery connection
22:54:24amiconnbluebrother: 'music' was the default view long ago. People comlained...
22:54:35 Quit Cassandra (Read error: 110 (Connection timed out))
22:54:45bluebrotherit was? Oh.
22:54:55bluebrotherwhy did they complain? Seems more logical to me.
22:55:38TommyTottthe problem is that I discovered this problem 2 days ago.. and im gooing to thailand on thursday.. so i dont have so good time.. thats the problem..
22:55:52preglowwhat the hell
22:55:56preglowi switched dircache on
22:55:58preglowthen shut down
22:56:01preglowthen started again
22:56:04preglowand it had not taken effect
22:56:15unleetI see well umm... Short notice makes a bit of a problem.
22:56:27preglowit was forgotten, i see...
22:56:30unleetUnless you can wait till after you come back.
22:56:43preglowi can reproduce it easily
22:57:01amiconnbluebrother: Umm, this was changed in June 2005. There should be some hints on the ml
22:57:56preglowalso for other options, what the hell
22:58:37TommyTottunleet: well, seems like i have noe other choise.. But now i dont have any music on the plane.. :(
22:58:56Kasperleoh. forum is down?
22:59:28 Join Criamos [0] (
22:59:39preglowcould someone please try to reproduce this? clear settings. reboot rockbox, enable dircache, then hold stop until it shuts down
22:59:43preglowthe settings are not saved
23:00:02unleetLaptop? Cd player? Other Mp3?
23:00:08preglowalso does the same for any other option, apparently
23:00:18rasherpreglow: You need to exit the menu
23:00:35Soappetur wonders if it's wise to put exe's in the wiki
23:00:43unleetI just received 2 free 125 mb mp3 players + speakers from Office depot for purchasing large amounts for work.
23:00:48preglowrasher: what menu? the menu the option is in?
23:00:54Soapcan the uploading of .exe files be limited to only a select few wiki members?
23:00:55TommyTottdont have either of those.. :P
23:01:02rasherpreglow: The main menu. Settings are only saved when exiting the menu, it seems.
23:01:04SoapThere are a lot of unvetted people with wiki access.
23:01:08preglowrasher: well, that's...
23:01:21unleetNo CD player?
23:01:31TommyTottCD, whats that?? :P
23:01:34rasherpreglow: amiconn says it's always been like that
23:01:35unleetyou can get one for 10-20$ at walmart.
23:01:46unleetI have tons of Cd's for my car.
23:01:50preglowrasher: i still say it's pretty weird behaviour
23:01:59TommyTotti have a computer in my car.. :P
23:02:03unleetToo lazy to remember the iriver transmitor thing
23:02:04rasherpreglow: Or I may have read him wrong. I've been known to do that
23:02:19amiconnrasher: No, you're right
23:02:23unleetInstall rockbox in your car :D
23:02:30rasherpreglow: I'd expect settings to be saved when I exit the single submenu as well
23:02:38TommyTottComputer, 7" touch screen, gps, centrafuse frontend..
23:02:41amiconnIt's always been lik ethat, and it's a good idea to do it like that in general
23:02:49preglowamiconn: i'd love to hear why
23:02:54preglowit's annoying and illogical
23:03:14unleetWish my accord had something that nice.
23:03:19 Quit PaulPosition ()
23:03:44TommyTottinstall it.. my car didnt have it..
23:03:48*bluebrother failed to find something on the ml
23:04:23unleetI'll take a look.
23:04:26unleetPrice Range?
23:05:25TommyTotthmm.. depends of what specs you want.. mine, about 800usd.. but i live in norway, and here its expensive
23:05:38***Saving seen data "./dancer.seen"
23:05:41TommyTottso in the us, less i would think
23:05:58preglowpetur: what kind of mic do you use, btw?
23:06:04TommyTottyou use a normal computer.. only the psu is special.. for 12V
23:06:34goffai'm liking aux in better
23:06:40goffajust hook up one of the players
23:06:50unleetLook up "PhatBox" I think it is. I want that.
23:06:57 Quit unleet ("CGI:IRC (EOF)")
23:07:12TommyTottthen you dont have all you music, gps navigation, radio, movies ect
23:07:19peturpreglow: a pair of cardoids and a mic-amp inbetween
23:07:27goffayeah.. watch the road ;)
23:07:30 Join unleet [0] (i=185c8710@gateway/web/cgi-irc/
23:07:47TommyTottgoffa: thats borning.. :P
23:07:54goffahe he he
23:08:08Soapdrifting quite a bit offtopic.
23:08:41peturpreglow: but the recording quality is not good, I couldn't get good mic placement and got too much bass and almost no vocals/guitar
23:08:45goffaso.. when's rockbox going to get gps?
23:08:48preglowpetur: it's not bad either
23:09:02unleetNeed a chip in your mp3 player for gps.
23:09:12TommyTottunleet: the PhatBox seemed hard to navigate.. all that music, and no other indication that a number
23:09:12unleetIm sure it can be done.
23:09:13goffai know.. i was just kidding
23:09:13amiconngoffa: The idea isn't even new...
23:09:35unleetI only want music though. I dont need internet and crap too.
23:09:38amiconn[IDC]Dragon once wanted to connect a gps module to his archos recorder
23:09:39goffamaybe i'm stuck inside the box.. but HOW?
23:09:44 Quit MarcoPolo (Read error: 113 (No route to host))
23:09:44amiconn(via serial i/f)
23:09:54Soapserial NEMA output from a good GPS.
23:10:01 Join MarcoPolo [0] (
23:10:11Soapthough most of the recent consumer ones only do USB.
23:10:22goffausbotgps ;)
23:10:27peturpreglow: preamp =
23:10:27unleetSolder a gps chip onto your mp3 player and pretend like "hey my mp3 has a gps!"
23:10:34bluebrothersure they don't simply use a serial to usb converter?
23:10:42goffaprobably do
23:11:02jhMikeSpetur: back. have a link?
23:11:09Soapbluebrother: the garmin etrex series, I believe, has no serial at all.
23:11:10TommyTottmy ipod still dosent show anything.. :(
23:11:18peturjhMikeS: got my PM ?
23:11:23unleetWait exactly 1 day.
23:11:25bluebrotherI mean inside ;-)
23:11:29unleetKeep it plgged in
23:11:36jhMikeSsorry didn't see it there covered up
23:11:46n1sTommyTott: tried select+menu for about 30 sec?
23:12:06goffawould be fun to take a player geocaching
23:12:07unleetHis battery is dead.
23:12:21peturpreglow: mic = (or maybe the previous version of it)
23:12:37goffadon't know that it would be any less cumbersome than having a gps and a player though
23:12:45TommyTotthahahahahhahahahha.. it started..
23:12:49n1sunleet: yes but some of those ipods with dead batteries behave strangely when they are charged again
23:13:02unleetWoah thats messed up.
23:13:03peturpreglow: prices are a bit better for members ;)
23:13:39TommyTottit works.. now i just have to charge, and then try to get the orginal firmware in..
23:13:44peturjhMikeS: speed will be lowish, thank my IPS
23:13:49amiconngoffa: The advantage of a dap is still the huge disk space. You could store tons of detailed maps on them
23:13:52unleetThats goo to hear.
23:14:03peturshould get 30KB/s peak
23:14:13TommyTottn1s: Thanks a lot.. :D
23:14:26n1snp :-) glad it worked out
23:14:46 Part lowlight
23:15:00TommyTottbut, if i dont have the orginal backup files, how do i get the orginal firmware back??
23:15:04 Join daurn|laptop [0] (i=daurn@
23:15:06goffaamiconn: good point
23:15:14SoapTommyTott: the apple restore tool.
23:15:27Soapor, I can send you the bootloader.bin for a 5th gen.
23:15:27n1sbut it will wipe the ipod clean
23:16:37TommyTottoki.. then i will see if i have to transfer the music first
23:16:45TommyTotti think i have the 4th gen..
23:17:13TommyTottyes, i have
23:17:24jhMikeSplayback.c 3562 is #ifdefed out? Not for MEM > 8
23:17:40 Quit daurn|laptop (Read error: 104 (Connection reset by peer))
23:17:45TommyTottarent there any way of doing it without wipe it clean??
23:18:10Soapfind someone to send you the 4th gen bootloader.bin file.
23:18:20jhMikeSmy checkout is a couple days behind though. was sort of hesitant to update with any ata problems present.
23:18:26 Join daurn|laptop [0] (i=daurn@
23:18:40TommyTottAnyone who has it??
23:19:12TommyTotti think rockbox has a potensial, but my version has a lot of bugs i think
23:19:45 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
23:21:36SoapTommyTott: you know there is an unofficial build for the 4th generation ipods which prevents the random crashes.
23:21:39jhMikeSpetur: says about 7hrs !! :P
23:21:45SoapASSuming that is your chief complaint.
23:21:56peturI'll leave the pc on this night :)
23:23:00jhMikeShope my ISP doesn't drop the connection. seems to happen every evening and have to reboot (AT&T after SBC takeover)
23:23:16peturjhMikeS: I hope you use a client that can resume
23:23:27jhMikeSit should
23:23:35 Quit bluey- ("Leaving")
23:23:40TommyTottsoap: i havent checked for while, but the biggest problem i have, is that when i scroll the music, it scrolles to fast.. suddenly im at the bottom.. if i try to scroll up, it goes to the top..
23:23:55 Join bluey- [0] (
23:24:29 Quit dau (Read error: 60 (Operation timed out))
23:24:49peturjhMikeS: should get a bit faster now, cpu and disk are idle again :)
23:25:12TommyTottcan i fint the bootloader file on the rockbox site??
23:25:29TommyTottor isnt it worth searching??
23:25:30*jhMikeS wonder's if Cowon is gonna stall and try and screw him on his warranty
23:25:37Soapno, it's not exactly kosher to distribute I believe, as it is Apple'
23:25:42Soaps bootloader.
23:25:48 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/
23:25:51 Quit lee-qid ("aufwiederbyebientotsayonara")
23:25:52jhMikeSpetur: Cranking at 18.2kB/s
23:26:15TommyTotthmm, how can i find it?!?
23:27:00peturjhMikeS: strange, it's going out here at around 30KB/s with a small dip from time to time
23:27:45linuxstb_TommyTott: If you didn't keep a backup, then you'll need to use itunes to restore. But your scrolling problem sounds like you're using an unofficial/experimental version of Rockbox
23:28:39peturjhMikeS: anyways... off to bed, gotta get up early tomorrow :/
23:28:57 Quit petur ("Zzzzz")
23:29:20TommyTottlinuxstb: i did, but my laptop crashed, so i dont have it anymore.. but maybe i should try to install a new rockbox version.. but can you guys help me find the best for my ipod?? im not shure about all the different versions
23:30:05 Quit merbanan (Remote closed the connection)
23:31:16linuxstb_Probably this one: - the only patch it includes is the one that disables CPU frequency scaling, which still seems unstable on the 4g.
23:31:18TommyTotti have a ipod 20gb photo
23:31:42TommyTottthat may not be 4th gen?? or??
23:31:45Lars_GI have a nano 2Gb 1G
23:32:08Lars_Ghmmm if I remember correctly, iPod photo was 4G but I'm not %100 sure
23:32:40TommyTotton, the grayscale says 4th gen..
23:33:05TommyTotti dont know if the color and grayscale is different else than the screen
23:33:39 Part n1s
23:33:41Soapphoto is the 4th gen
23:33:53Soapthere is a 4th gen grey, and a 4th gen colour/photo
23:34:21Soapand I hope both of your players have GB capacity, Gb would be a rip off.
23:34:47Lars_Ghehehe true
23:34:55Lars_GYes, afaik 5G is iPod video
23:36:06TommyTottthen it was how to install the new build.. long time since i did this
23:36:54Soapunzip it onto your player, overwriting the existing files.
23:37:35TommyTottoki.. then i try
23:37:43*Lars_G falls face first on the floor frozen as a ice cube and shatters
23:38:51bluebrotherthe ripoff is that it's GB and not GiB (and windows pretending GiB to be GB)
23:39:18TommyTotthmm, where is the bootloader located on my ipod??
23:39:26mirakdoes anyone had crash on wav playing with H300 recently ?
23:39:59mirakI failed a wav recording of a live show also
23:39:59Lars_Gbluebrother: Gb vs. GB is a bigger much bigger lose than GiB over GB
23:40:11Lars_GTommyTott: In a private partition
23:40:22Lars_GTommyTott: And there's also the initial loader which is in rom/flash
23:40:29 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
23:40:37bluebrotherindeed, but with bigger drives that GB vs. GiB gets quite some amount
23:40:48Lars_Gbluebrother: But I understand you well, people handling GiB as GB pissess me off to no end
23:41:12bluebrothereven Rockbox doesn't do it correctly :(
23:41:14rasherRockbox does this.
23:42:18rasherbluebrother: Patch!
23:42:23rasherConfig option!
23:42:38rasherIn fact, this is probably set in the language file
23:42:46rasherI should quietly change it in dansk.lang
23:43:27rasherI guess the voice file whould say "gibibyte" as well.
23:43:37bluebrotherenglish.lang should start doing it
23:43:40rasherWhich is.. annoyingly stupid.
23:44:01bluebrothergibibyte is stupid?
23:44:14rasherWell, saying it aloud is
23:44:24rasherThe word is just.. not good
23:44:39preglowbeats confusion
23:44:49rasherYeah, that's true.
23:45:02 Quit bluey- ("Leaving")
23:45:08bluebrotherwell, at least the word itself makes sense. And there are words that are way more worse as words
23:45:13Soapthere are some truths the general public is not ready for.
23:45:28bluebrotherin germany you say "handy" for mobile phone. *That* is really crap
23:45:29SoapI believe the mere existance of gibibyte is one of those truths.
23:45:46preglowbluebrother: sweet lord, that's cheesy
23:46:20bluebrotherthe funny thing is the germans going to some english speaking country and not knowing that difference ...
23:46:34jhMikeSpreglow: don't know if you ever saw this: . Maybe want to give a look?
23:47:14Bagderbluebrother: I can acknowledge that a german I met looked very surprised when I told him it isn't called like that in english ;-)
23:47:35preglowjhMikeS: i think i've seen it, and i think it does phase vocoding
23:47:40 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/
23:47:44bluebrotherI simply call mine "phone" and quite a few looked really strange at me when they realize I'm referring to my mobile
23:47:44preglowyes, indeed
23:47:49preglowi don't want to code a fast fft :/
23:48:27jhMikeSno, why? just the code or it's just too cpu intensive?
23:48:35preglowjhMikeS: both, really
23:48:39bluebrotherI'm pretty sure that word was intended by one of those marketing guys that have no idea about the technical stuff they are advertising
23:48:43preglowjhMikeS: it'll probably suck down a very decent amount of cpu anyway
23:50:06jhMikeSI might find it fun to try it out myself and learn something important in the process anyway.
23:50:10 Quit ender` (" Don't worry about the world coming to an end today. It's already tomorrow in Australia.")
23:50:15preglowjhMikeS: might not take too long to get it going, though, if you've already got a fixed point fft/ifft
23:52:49 Quit TommyTott ()
23:53:24jhMikeSI'll probably have to make it up just from the math itself. I don't have anything readily available but that will be beneficial in learning.
23:55:24preglowjhMikeS: don't expect too much from it, though. while it doesn't have as obvious transient artifacts as the time domain one, it does smear them quite badly
23:58:11 Join TommyTott [0] (
23:58:51TommyTottunleet: here??

Previous day | Next day