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 2005-09-05

00:00:14zeuntill we get vector digital audio, digital has drawbacks
00:00:25zeOnkelJonas: really both cost a good bit
00:00:37zefor high quality
00:00:56zemy soundcard was $175 and that was a really good deal on a pretty low-end one
00:01:02OnkelJonassure... but outfitting a studio with for digital vs. analog I believe will show quite a large saving
00:01:08zeas far as 24bit/96khz multichannel interfaces go
00:01:28]RowaN[hmm sacd players have high resolution analog outputs, they wont be encrypted
00:01:35OnkelJonasalthough the lifetime of analog equipment might be better... not sure about that
00:01:56zeand then my cpu's not really powerful enough to do very many plugins or a whole lot of tracks (especially with plugins on many tracks)
00:02:16OnkelJonastheir digital outputs probably have the s/pdif bits for no copying set
00:02:19zeand here i was thinking 1ghz was still fairly decent
00:02:40zeOnkelJonas: yeah you have a point about equipment lifetime too
00:02:50]RowaN[ah heres the dvd-audio ripping stuff
00:02:54zethere's lots of 30+ year old analog equipment thats raised in value
00:03:04linuxstbOnkelJonas: The S/PDIF output is limited to (I think) 16-bit/48KHz for DVD-Audio players. I'm not sure if SACD players have S/PDIF.
00:03:04zewhile anything digital's obsolete and lost value by the time you get it home
00:03:23 Quit muesli- (Read error: 110 (Connection timed out))
00:03:33zelinuxstb: 16-bit? but dvd-audio is 24bit isn't it?
00:04:08***Saving seen data "./dancer.seen"
00:04:15OnkelJonasanyway... i have talked about digital vs. analog with my uncle a few times (hes a composer) ... his reasoning in going digital was *very* large timesavings... which in that business can translate into quite a lot of money (he makes musicals etc.)
00:04:33linuxstbze: DVD-Audio is either 16/20/24-bit - the S/PDIF limitation is all part of the copy prevention nightmare.
00:04:46 Quit TCK- (Read error: 110 (Connection timed out))
00:04:51]RowaN["One interesting aspect of CPPM is that it allows for a hacked playback device to be revoked using a Media Key Block (MKB). Each licensed decoder – in this case WinDVD – has assigned to it a set of unique keys that allow it to obtain the Media Key (used to encrypt the audio content) from the MKB and decrypt the audio content. As WinDVD is now compromised in its current form, future DVD-Audio releases could be prevented from
00:05:08zelinuxstb: but i'd imagine most people using s/pdif would want to do so to put it into some high-end DAC's for quality's sake
00:05:24linuxstbze: Exactly. Now do you know why DVD-Audio hasn't caught on?
00:05:41zelinuxstb: seems like they'd be fucking over the exact market that such a feature would appeal to
00:05:46zelinuxstb: heh
00:05:59zethen they might as well leave it off
00:06:27linuxstbThere is an alternative - a proprietory and encrypted digital link between some DVD-Audio players and some multi-channel DACs.
00:07:30zeOnkelJonas: makes sense
00:07:45zeOnkelJonas: digital is easy to deal with in general
00:08:01linuxstbI don't know if it's the DVD-Audio equivalent of Macrovision - i.e. if it's possible for a DVD-Audio disc to be authored to allow full-resolution output.
00:08:08zethe way you can manipulate tracks in a DAW is unparalleled in the analog realm
00:09:30OnkelJonasif the music industry keeps using these draconian anti piracy methods, we are going to see bootlegs with higher quality audio than studio recordings
00:10:27zestill, i know someone who recently bought a tape deck to put drums through before going into their multi-channel digital soundcard, specifically to that "tape warmth" that a lot of recording drummers are nuts over
00:11:34zeand we still rely on a good bit of analog equipment for digital recording
00:11:36 Quit paugh (Read error: 60 (Operation timed out))
00:12:01zemic's, preamps/amps, etc
00:12:11zea lot of instruments
00:12:18zeelectric or mic'd acoustic
00:12:58zeand if somebody comes out with an audio transduction method that doesn't have an analog stage, i.e. a digital mic, i'm sure lots of people won't like it
00:13:12OnkelJonasheh... somehow reminds me of these guys i saw at roskilde festival a few years ago... they played on everything made from metal, and captured the vibrations somehow (electrodes or some stuff like that) - very weird, but quite fascinating
00:13:33OnkelJonasof course they were the opposite - everything digital
00:13:33Bagderheavy metal? ;-)
00:14:03OnkelJonasbtw - only thing i really miss about rockbox atm is a good wps
00:14:22OnkelJonasI tried making one myself, but I suck quite a lot at it
00:15:58OnkelJonasffs... its gotten late here - I better get to bed - i have to sit through 8 hours of talking tomorrow :(
00:16:01]RowaN[wps is pretty flexible, great wps is possible with rockbox =]
00:16:15OnkelJonasi know... just havent been able to make one yet
00:16:44]RowaN[best thing to do it first of all sit down and write down the info YOU want displayed on a wps
00:16:59OnkelJonasand noone else seems to share my idea of the perfect wps (simple, big letters AND a "next" display)
00:17:00]RowaN[well planning is the key to any project i guess
00:17:14]RowaN[my wps has a next =p
00:17:21]RowaN[big letters? just choose a big font
00:17:23OnkelJonasif the speakers tomorrow are boring, i might just do that ;)
00:17:28 Quit tvelocity (Read error: 113 (No route to host))
00:20:23linuxstbHas anyone ever tried to write a WPS designer as a plugin?
00:21:46 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
00:25:06 Quit Mark_ (Read error: 104 (Connection reset by peer))
00:26:49bagawklinuxstb, I started one
00:27:02bagawkabout a year ago
00:27:54bagawkNot sure if I have the source around, if I do it would be on a school server (I got bored during some off-time)
00:33:01 Quit bagawk ("Leaving")
00:34:19 Quit paugh (Excess Flood)
00:35:31 Join Sucka [0] (
00:39:16 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
00:43:46 Quit OnkelJonas (Read error: 110 (Connection timed out))
00:45:36preglowlinuxstb: not anything more sofisticated than measuring the number of ticks generating a frame of audio takes, no, and just dividing that by total time, or something
00:45:47preglowlinuxstb: got people hanging around now, so i'm a bit slow :]
00:48:45 Quit pike (Read error: 110 (Connection timed out))
00:49:56linuxstbI was thinking of that approach - but is 100 ticks/second a high enough resolution for timings?
00:52:13preglowi'm not certain
00:52:19preglowthe alternative is using the new timer module
00:52:39linuxstbWhat new timer module?
00:53:03preglowamiconn added it not too long ago, it's able to reach higher resolutions that the tick counter, afaik
00:54:12linuxstbJust looking at it now...
00:54:48preglowanywho, it'd be a welcome addition to the audio thread debug screen
00:55:04preglowthe boost counter isn't reliable enough for me to use it for performance measurements
00:55:16 Quit paugh (Read error: 60 (Operation timed out))
00:55:27 Join paugh [0] (
00:55:29preglowlinuxstb: you should be aware that using the new timer disables the backlight dimming, so not something that should be enabled for ordinary builds
00:58:16linuxstbI don't know enough (anything) about the threading system, but would it be possible for the task switcher to count how much CPU time a thread has been consuming?
00:59:11preglowi don't think the task switcher has any idea what thread is running
00:59:43preglowhmm, threads do seem to have names
01:02:18linuxstbMaybe a simpler option would be to resurrect the idea of a xxx2wav plugin - but can plugins use IRAM in the same way as a codec?
01:07:59preglowyes, they can
01:08:25preglowa simple wav writer context option would be good
01:08:49preglowand shouldn't be _that_ hard to implement
01:13:56linuxstbI'm not sure what you mean. Do you mean as a core part of rockbox, instead of a plugin? Or a special kind of plugin?
01:14:05preglowjust a new way of using the codecs
01:14:18preglowprocess audio as usual, just feed the result to a file instead of the dac
01:16:22linuxstbHow would that help us with timings? We would still need to use the codec buffers (or at least, the same API).
01:17:33linuxstbI think I get it now. We would not need to measure the "idle time".
01:17:44linuxstbWhich is basically the probably now.
01:17:54preglowwe'd pretty much be back to where we were with the xxx2waz plugins
01:18:12preglowwith no need to modify any plugins to do so
01:18:20preglowjust add a rockbox option to use the codecs in a different way
01:18:26preglowsomeone is going to ask for it sooner or later anywya
01:18:52preglownot to mention it'll be needed to verify the codecs produce correct output
01:19:20 Join Naked [0] (
01:19:23linuxstbIs there any such thing as "correct output" with lossy codecs?
01:19:41linuxstbI mean will two MPEG decoders give bit-identical output?
01:20:11preglowthere is more and less correct, yes
01:20:45preglowthe mpeg spec even specifies a measurement method
01:20:59preglowand there are certain bounds you have to stay within to be able to call your decoder a valid decoder
01:21:00 Nick Naked is now known as Hadaka (
01:21:39linuxstbI still think it's more a developer's need, than a user's need though - so maybe we shouldn't be thinking about adding such a feature to Rockbox.
01:21:46preglowstill, comparing our libmad with x86 libmad is still a valid way to find out if i've done anything wrong
01:21:58preglowwell, yeah, but i need it
01:22:13linuxstbSo do I. But only because we are developing the codecs.
01:22:15preglowif i can't wav write, i don't know if the codecs provide good sound
01:22:58linuxstbI assume you also want to write the output of your DSP code?
01:23:13pregloweverything that would normally be done
01:23:16preglowjust to wav instead of dac
01:23:51linuxstbSo we just need an option to write to WAV instead of DAC, and also to make Rockbox run as fast as possible (so we can time things) ?
01:24:05preglowwell, yes, that's how all other wav writers work
01:24:38preglowbut seriously, i don't think a wav writer option would be unwelcome
01:24:43preglowsome people would probably use it
01:25:12 Join ciccia-88 [0] (
01:25:31linuxstbYou're probably right.
01:25:53ciccia-88u speak with me?
01:26:14linuxstbciccia-88: No - the person above you.
01:26:21linuxstbBut you may be right as well.
01:27:04preglow'cause that's more or less what i'm talking about, a wav writer with some performance figures
01:27:16ciccia-88i don't understand lol
01:27:21ciccia-88it's funny °_°
01:27:44ciccia-88where do u live?
01:28:16linuxstbpreglow: I'm not sure how it would fit into Rockbox though. A "viewer plugin" would seem simpler - if we could give the plugin access to the codecs.
01:28:44preglowlinuxstb: yes, that'd also work, but why not a simple context menu option?
01:28:52preglowlinuxstb: open with, delete, write to wav, etc
01:29:36 Part ciccia-88
01:30:10linuxstbI'm thinking from an implementation point of view.
01:30:27preglowloading a plugin from a plugin might be tricky, heh
01:31:31 Join solex [0] (
01:32:32 Quit gromit` (Remote closed the connection)
01:32:37linuxstbI'm liking the idea. A "plugin plugin" should be able to tell Rockbox to load a "codec plugin".
01:33:56linuxstbBut are you thinking of making it a part of the core rockbox - a heavily cut-down version of playback.c?
01:34:56preglowwell, yeah, that was how i thought it would have to be done
01:35:44preglowcalling it a cut down version of playback.c is a bit misleading
01:36:00preglowsince all it would do is basically just call the plugin, fetch the audio, and write it to disk
01:36:20preglowit would have to do basic buffering, of course
01:37:04linuxstbIt would have to do all the buffering that playback.c does. Unless we assume the codecs don't need all the features, e.g. seeking.
01:37:49preglowsome codecs do seeking even if you don't ask for it :/
01:37:55preglowbut that's probably neglectable
01:38:05linuxstbBut of course, it only needs to buffer exactly one track.
01:38:11preglowbut i can't see a need for seeking when wavwriting
01:38:19preglowyeah, sure, but a track can be pretty big
01:38:26preglownot fit in the buffer, etc
01:39:06linuxstbSome codecs may seek around the metadata - or at least seek forwards to skip it. But I think codecs should do that by calling advance_buffer().
01:40:09linuxstbI think it's a little more than trivial though - we need to keep the codecs happy, and the codecs expect to be talking to playback.c
01:40:34linuxstbBut maybe it will be a good opportunity to clean up the API between playback.c and the codecs - if it needs cleaning.
01:40:49preglowthey don't do much talking
01:41:11preglowbut yeah, it does depend on a lot of code i have no idea of how works
01:41:17preglowso might easily be less than trivial
01:42:40preglowbut i need a good way of measuring performance soonish anyway
01:42:53preglow'good' changes have resulted in worse performance before :/
01:42:56linuxstbI think it's worthwhile - not sure what the other devs will think though.
01:43:47 Quit ender` (Read error: 113 (No route to host))
01:43:54preglowi don't think they'll be too ardently opposed
01:43:57preglowit's a decent feature
01:44:26linuxstbSo why were the ???2wav plugins removed?
01:45:30preglowbecause of having to maintain two seperate versions of the same thing, i believe
01:46:03preglowit's better to make rockbox wav write, than have two versions of the same things outputting the same thing a bit differently
01:48:47 Quit solex_ (Read error: 110 (Connection timed out))
01:50:13linuxstbOn a different subject, ALAC decodes the audio to two arrays of 32-bit integers. Should I (can I?) pass those two arrays directly to pcm_insert_split?
01:56:56preglowi believe so
01:57:03preglowbut not sure
01:59:33 Join tvelocity [0] (
02:01:53preglowi gotta go to bed
02:02:00preglowsee you later
02:02:18 Quit preglow ("off")
02:04:09***Saving seen data "./dancer.seen"
02:24:58 Quit ]RowaN[ ()
02:29:33 Join pike [0] (
02:58:02 Join Maxime [0] (
03:15:32 Quit Mxm`Pas`Bien (Read error: 110 (Connection timed out))
03:51:16 Join cheriff [0] (
03:55:03 Nick paugh is now known as AliasCoffee (
04:04:11***Saving seen data "./dancer.seen"
04:05:32 Join QT_ [0] (i=as@madwifi/users/area51)
04:17:48 Quit QT (Read error: 110 (Connection timed out))
05:12:02 Part cheriff
06:04:12***Saving seen data "./dancer.seen"
06:07:58 Quit AliasCoffee ("Leaving")
06:10:55 Join Aramil [0] (
06:31:14 Quit tvelocity (Read error: 110 (Connection timed out))
07:13:07 Join LinusN [0] (
07:17:36t0masg morning
07:18:14 Part LinusN
07:18:44amiconnt0mas: You scared him ;)
07:18:50amiconnGood morning
07:19:07t0mashe has a problem with people? ;)
07:23:06 Join matsl [0] (
07:38:59 Quit Slasheri (
07:39:23 Quit DMJC (Read error: 110 (Connection timed out))
07:51:56 Join B4gder [0] (
08:04:16***Saving seen data "./dancer.seen"
08:17:37*amiconn just noticed an opportunity to make the firmware file smaller
08:18:35NJoinSlasheri [0] (
08:20:06amiconnI wondered why yesterday's rather small changes increased h120 binary size by ~8KB
08:20:31amiconnTurned out it was caused by Slasheri's move of the voice codec stack to IRAM
08:20:51amiconnObviously we don't have an .ibss section, only .idata
08:21:16amiconnSo all IRAM variables get included in the binary, even if they're uninitialised
08:22:18amiconnChanging this probably won't save much on archos, but quite a bit on iriver
08:22:33amiconnImho it's still worth it
08:23:43amiconnWe should also have an .iconst section, merely for allowing to declare an IRAM variable const
08:32:47 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
08:35:30 Join ender` [0] (
08:37:36 Join kurzhaarrocker [0] (
08:39:46 Join gromit` [0] (n=gromit`
08:43:26linuxstbMorning all. I'm curoius why the SWAB16/SWAB32 functions have those names - POSIX defines the "swab" function which AFAICS is completely different to those macros.
08:44:18zelinuxstb: they don't SWAp Bytes?
08:44:54linuxstbSWAB32 doesn't swap *adjacent* bytes.
08:45:11linuxstbIt's also not defined on a little-endian target. i.e. it's a convert little-endian to host-endian macro.
08:45:30linuxstbI mean it's defined as a null operation.
08:46:59 Quit Slasheri (
08:49:41linuxstbIt also seems to be used strangely in the fat.c driver - fat.c is also defining its own SWAP_WORDS define, and it only calls SWAB if SWAP_WORDS is defined.
08:50:23linuxstbShouldn't we be using the ROCKBOX_BIG_ENDIAN there?
08:50:47B4gderit certainly feels like we should
08:53:01linuxstbThe reason I ask is that I would like to add a set of LE/BE to Host-Endian conversion macros. All the codecs are using their own macros.
08:54:10amiconnI've already thought about this too
08:54:14B4gderthat define is just so much newer than the ata code so I guess it just hasn't found its way in there
08:54:51amiconnThe problem is that we probably need multiple sets
08:55:01linuxstbBut it's also the same test done twice - the SWAB macros do nothing on a little-endian machine already.
08:55:18linuxstbWhy multiple sets?
08:55:22amiconn(1) swap a variable (in register)
08:55:44amiconn(2) read from memory / write to memory
08:56:00amiconnThe latter can be unaligned, we must take care of this
08:56:33amiconnIt's not a big problem on coldfire (and neither on x86 sim), but it is on SH1
08:56:35 Join webguest43 [0] (
08:56:42webguest43Hi !
08:56:57webguest43it seems the charger for my iriver 140 has gone to hell
08:57:02amiconnColdfire can read/write words and longs unaligned, this only affects performance
08:57:13webguest43didn't like the rain I think :)
08:57:17amiconnSH1 throws an address error in this case
08:57:43webguest43i am about to buy a replacement, anything I have to be aware of except 5v and polarity on the plug ?
08:59:24linuxstbamiconn: I'm curious - how do we know if a variable is in a register?
08:59:45linuxstbOr is that only for use in assember functions?
09:00:54amiconnThe compiler will ensure this, the asm() block specifies the constraints
09:02:38amiconnThe point why I think separate macros for reading from / writing to memory are useful is that it would be not very efficient to read/swap resp. swap/write separately
09:07:02 Join einhirn [0] (
09:14:43webguest43hmm, no thoughts on the charger ?
09:15:49 Part Aramil ("Leaving")
09:20:06 Join Wett [0] (
09:23:16linuxstbI'm guessing from the ReleaseTodo page that there is still lots of work to do before the release?
09:41:14 Quit linuxstb ("Leaving")
09:56:08solexam i missing something or is there no %-sequence for whether the acis plugged in?
09:56:17*solex is talking about the wps
10:02:42B4gderwrite a patch!
10:04:20***Saving seen data "./dancer.seen"
10:04:56kurzhaarrockerCan LinusN be expected to appear today?
10:05:05B4gderI think so
10:05:32B4gderI met him yday and he didn't mention anything special
10:06:05kurzhaarrockerDid he drink a lot of beer?
10:06:55B4gderwe played "floorball" or whatever you call it in english
10:06:56*kurzhaarrocker shivers
10:07:10B4gderkind of like hockey but indoors with a ball
10:08:15kurzhaarrockerAnd you won over him?
10:08:35zehockey usually is indoors
10:08:36B4gderof course!
10:09:08B4gderze: hm, true, but not "real" indoors ;-)
10:09:27zewell its a refridgerated room or something i guess
10:09:47zeso... but uh yeah i've no idea what you're talking about
10:10:14B4gderits one of the most commonly practised sports in Sweden
10:10:37zeit sounds like what us americans call soccer and what everyone else calls football... except, indoors
10:11:09zeoh except i guess it uses sticks still eh
10:11:31zei see
10:11:38 Join nc [0] (
10:11:51zeso its like you said.. hockey with a ball instead of a puck, and a floor instead of ice
10:12:01B4gderbasically, yes
10:12:13zei think i've played that before :p
10:12:15B4gderand its really tiresome
10:12:36ncguys do the "voice menu" files work with the daily builds? if yes, do i just unzip the file in root?
10:12:42zei dunno what we called it though, i'd intuitively referr to it as ball-hockey or just hockey
10:14:32nc(using iriver btw)
10:14:47B4gdernc: yes I think you should
10:14:58B4gderI admit I've never used voice more than some initial testing
10:15:00B4gderyears ago
10:15:21*B4gder doesn't use many rockbox features at all
10:15:29nci am curious to try them :) i only loaded rockbox last night and i am excited!
10:15:43B4gderwelcome to a whole new world to explore! ;-)
10:16:08ncthank you, i am starting to feel prouder about my iriver already
10:16:25nccrossfading is here and i am more than happy!
10:16:38solexdon't forget to create your own wps and share it on the wiki!
10:16:50solexthere's more and more crazy stuff there
10:16:59solexthe latest one features a little animation
10:17:24ncerrr i don't know what wps is, but i have a feeling i will learn soon, and hopefully i will build one to share.
10:17:35zewhile playing screen!
10:17:48solexthe screen you are seeing when your player is, erm, playing
10:18:08nccool! checkin it out now
10:20:27 Join random_man [0] (
10:21:59solexand I guess you already found out how to change the font?
10:23:40 Quit kurzhaarrocker (Remote closed the connection)
10:23:46random_mani think i may have found a bug
10:23:56 Quit webguest28 ("CGI:IRC (Ping timeout)")
10:23:57solextell us!
10:24:31random_manwhen i go to load any .cfg file the player seems like it was to load it for about 1 second then stops playing music and freezes have to reset it
10:24:50B4gderyeps, that's a known issue
10:25:01B4gderon iriver
10:25:04nci'm off. see u guys around ;)
10:25:14 Quit nc ()
10:25:18random_manoh ok sorry about that just found out playing around wtih tit myself
10:25:26B4gderno worries
10:26:36 Quit random_man ("CGI:IRC")
10:27:49 Join ashridah [0] (
10:30:19 Join Slasheri [0] (
10:30:19 Quit Slasheri (
10:30:23NJoinSlasheri [0] (
10:32:37 Join preglow [0] (
10:32:58preglowso, is 2.5 coming out today? :P
10:33:52B4gderI don't think so
10:44:37 Join linuxstb [0] (
10:58:57amiconnpreglow: Still too much red & orange in the todo list :-/
10:59:35preglowyeah, kind of figured
11:00:28preglowany of you guys firmly opposed against a wav writer in rockbox, btw?
11:01:19*Slasheri is waiting when the feature freeze ends so i could commit the new features waiting on my computer ;)
11:02:19 Join Zagor [0] (i=foobar@pdpc/supporter/sustaining/Zagor)
11:02:22amiconnSlasheri: You could commit some bug fixes in the meantime ;)
11:02:54Slasheriamiconn: hehe, true :) hmm, in fact i could do a fix to the audio flushing bug
11:03:29bluebrother^maybe someone should add a delay notification to the main website? I guess a lot will start asking about this.
11:03:30preglowthat would be lovely
11:03:54Mode"#RockBox +o B4gder " by ChanServ (ChanServ@services.)
11:04:03Topic"The 2.5 feature freeze is here! No release just yet" by B4gder (
11:04:59amiconnSlasheri: You could also fix the playback engine behaviour for nonexisting tracks...
11:05:26Slasheriamiconn: Hmm, but that requires some modifications to the playlist engine too.. But if that's ok, i could fix it
11:05:44amiconnA bug fix usually requires changes...
11:06:12amiconnI did quite some changes to the archos recording engine, and there will be more
11:06:29amiconn...probably also to the playback engine
11:06:45Slasheriah.. probably i should then do that :)
11:17:04CoCoLUSdoes rockbox follow the odd/even versioning scheme?
11:17:52ZagorCoCoLUS: no
11:19:18 Join [-AIR-] [0] (n=air@
11:19:37[-AIR-]hey, is 2.5 released today?
11:19:47[-AIR-]or is it postponed ?
11:20:41 Join darkskiez [0] (n=darkskie@
11:21:25darkskiezmy ihp140 just broke :( What would folks recommend for a new player... Is it worth getting the Ihp3xx series ?
11:21:27[-AIR-]i know, i saw that obv. but "just yet" could refer to it being released in a few hours, or a few weeks
11:21:49[-AIR-]darkskiez i'd go to the misticriver support channel
11:21:50Slasheridarkskiez: how it broke?
11:22:07Slasherii think you should fix it, really
11:22:14 Quit webguest43 ("CGI:IRC")
11:22:23darkskiezjust didnt power up one day, its out of warranty. Used 3rd party charger, it was fine for one charge, but doesnt turn on now.
11:22:50[-AIR-]o dear.
11:23:05Slasheriwhat are the power ratings on that charger? (voltage, current, polarity etc.)
11:23:14darkskiezmatched perfectly
11:23:25Slasherihmm.. and the charger is regulated?
11:23:55darkskiezI believe so
11:24:23Slasheriif it uses switching power regulation technology (very lightweight etc), that might also cause some problems
11:25:21Slasherihmm.. is the charge led coming up if you connect the original charger? can you reset the player?
11:25:28preglowthe original charger is switching, isn't it?
11:25:33Slasheripreglow: i think so
11:25:42preglowsure as hell doesn't seem to be heavy enough to have a transformer
11:26:00Slasheribut some switching chargers might give high voltage output peaks if there are not enough current consumption
11:26:20bluebrother^does it turn on with the charger attached?
11:26:21Slasheripreglow: of course it has a transformer but only a very tiny one
11:26:52darkskiezthe charge led does come up, reset button does nowt, doesnt turn on with charger attached
11:27:08darkskiezmeasured battery with multimeter and it has charged up to 4v
11:27:37Slasheripreglow: that transformer is used mainly for galvanic mains separation for safety reasons (it will not lower the voltage level much)
11:27:43bluebrother^have you tried replacing the battery with another? Or a regulated power supply?
11:28:21darkskiezattached a regulated power supply in place
11:28:26darkskiezno luck
11:30:56bluebrother^have you tried measuring the supply voltage on the chips? Maybe it's only the power part of the unit that is broken.
11:31:03Slasheridarkskiez: have you tried measuring the voltage output from the charger you had charged it?
11:31:23bluebrother^really sad there is no complete schematics of the player avaliable.
11:32:02darkskiezits fine
11:32:07Slasherii think you should try measuring at least cpu (and possible other power control chips) supply voltage
11:32:13darkskiezit was 5v, took about 1/2 A
11:32:28bluebrother^hmm. What type is the reset circuit in the unit?
11:32:33darkskiezconnected 3.3v to the VDD 3v3 line on the pcb and the red led lit.
11:32:36Slasheridarkskiez: what it was without any current?
11:33:01Slasheri0v? that is not good..
11:33:10darkskiezwithout current you have no voltage
11:33:19darkskiezotherway around.
11:33:28Slasherithat is what i was worried about..
11:33:47Slasheritry measuring with a very low current or using an oscilloscope
11:33:52darkskiezyes, the laws of physics are different in irivers
11:34:06darkskiezwhat do u mean with measuring with low current ?
11:34:10darkskieza current limiting power supply ?
11:34:38Slasherii think when the battery was going to trickle charging mode, eventually current drops and when there was not enough current, the charger started outputing over voltage peaks
11:34:48Slasherino, with some resistor
11:35:44Slasheriand without any current, the charger outputs possible very high voltage peaks (but those are so fast that multimeter cannot measure them)
11:36:15wubblaare there any new successes concerning the h300 port?
11:36:45preglowSlasheri: ahh
11:36:51Slasheriand then chargers internal protection mechanism activates for a short period and power is cut. That's why you see only 0v coming out which means you have used a bad charger
11:37:01preglowSlasheri: haven't really studied them that much, but it makes sense
11:37:24darkskiezi made up 0v because i didnt understand the question
11:37:41 Join hicks [0] (
11:37:44darkskiezthe VDD trace 3v3, should that have power all the time, or only when its on?
11:39:46Slasheridarkskiez: so you have tried measuring the charger output directly with the multimeter without any load?
11:43:48preglowarghh! apropos switching regulators, how i hate this high pitched whine that comes from my computer
11:44:41darkskiezwill that help repair it ?
11:45:18Slasheridarkskiez: that should have been the first thing to do even before connecting the charger first time
11:45:34darkskiezyes, it was fine
11:45:48Slasheriit was 5V?
11:45:55darkskiezjust over
11:46:01Slasherihow much?
11:46:02*amiconn summons [IDC]Dragon
11:47:42Slasheriok, weird..
11:48:04darkskiezit charged and worked after charging it with that
11:48:43preglowSlasheri: do you think it'd be hard to a wav writer to the playback engine somehow?
11:48:52preglowadd a wav writer, that is
11:49:10Slasheripreglow: no, that should be easy thing to add
11:49:18Maximeand, about FM recording, will it be possible?
11:49:42Slasheriin fact we should add it to the pcmbuf level so it would include all the dsp processing
11:49:53preglowSlasheri: i'm pretty much talking about a context option that just dumps one track to wav while doing some performance measuring
11:49:56SlasheriMaxime: it is possible, already
11:50:17preglowSlasheri: i've pretty much lost all easy means to measure codec performance
11:50:28Maximeokai lol
11:50:49preglowSlasheri: and btw, the voice and main codec runs in different threads, yes?
11:51:03Slasheripreglow: Hmm, would that mean when using wav writer, there would be no real pcm audio out?
11:51:10preglowSlasheri: indeed
11:51:47Slasheriah, then we could create another "lowest level pcm-driver" for the wav writer
11:51:48preglowSlasheri: wouldn't make sense to have audio out, what makes sense (to me at least) is rockbox pushing to the codec as fast as it can to render the file as quickly as possible
11:51:54Slasherithat should be simple
11:53:59preglowwhat i want is more or less just xxx2wav functionality back
11:54:24preglowso i can measure codec performance and correctness
12:02:04amiconnWhy not just re-add the xxx2wav plugins (for developer builds)?
12:02:20amiconnImho it makes no sense to have this code in the core playback engine
12:02:21preglowbecause that would mean maintaining two versions of the same thing
12:02:39preglowand making the codec plugins back to a xxx2wav plugin would require some work
12:02:44preglowi bet adding a wav writer would be less work
12:03:35preglowamiconn: if it's a useless feature for normal users (which i don't think it is), can't it just be a debug feature, then`?
12:04:24***Saving seen data "./dancer.seen"
12:07:13preglowi'm still pretty certain it would be less work than recoding the codec plugins to be xxx2wav based again, and besides, a wav writing option would also allow me to time and measure quality improvements in the dsp part
12:09:16darkskiezwhat type of hdd adapter do i need to use the iriver ihp140 hdd
12:14:50 Join Mr_C [0] (
12:23:44preglowhaving workmen blast 120dbs in through my windows must certainly be called an enhancement of my environment
12:28:49 Join Moos [0] (
12:29:04MoosHello! :)
12:31:58 Quit [-AIR-] (Read error: 110 (Connection timed out))
12:45:46 Quit wubbla (Nick collision from services.)
12:46:50 Quit ender` (Read error: 104 (Connection reset by peer))
12:54:03 Join ender` [0] (
13:24:12 Quit preglow ("leaving")
13:24:24 Join preglow [0] (
13:25:33 Join Febs [0] (
13:27:40 Join pilot000 [0] (
13:29:36 Join edx [0] (
13:35:09 Quit darkskiez ()
13:46:15 Join noC|andY`fRa [0] (
14:01:38linuxstbThinking about preglow's need (which I share) for a wav writer facility, as well as the "need" for the iriver port of the video player plugin to access the codecs, I'm thinking we should give plugins access to the codecs.
14:02:03linuxstbThis will also be needed if we want to make the splitedit plugin work with software codecs.
14:03:29linuxstbThe WAV writer could be implemented as a normal plugin/viewer - a single "wavwrite" plugin which would work with any supported codec.
14:04:26***Saving seen data "./dancer.seen"
14:10:04preglowthe problem is that a pretty significant portion of the playback code would be needed
14:10:44preglowi don't know how suited that code would be for being called from plugins
14:13:09linuxstbNot really - the plugins using the codecs would need to implement the necessary functions - depending on their needs.
14:13:34preglowmight also need to call dsp code
14:14:09linuxstbCouldn't that be included in the "codec API for plugins" ?
14:14:49preglowstill, i've got a feeling this'll be quite a bit harder
14:14:52preglowand i want this now :/
14:15:00linuxstbI think the problem is that plugins would want to tell the codecs to decode a small memory buffer of compressed data (e.g. one MPEG frame). I don't think there is a way for the codec API to do that at the moment.
14:17:32linuxstbThis just seems to be something Rockbox will need in the future, and it will meet the need of a WAV writer at the same time. But I also want a quick solution as well.
14:17:49 Join [1]Moos [0] (
14:18:21 Quit Moos (Read error: 104 (Connection reset by peer))
14:18:21 Nick [1]Moos is now known as Moos (
14:24:42 Join darkskiez [0] (n=darkskie@
14:39:41 Quit pilot000 ("CGI:IRC (EOF)")
14:45:44darkskiezis the MK4004GAH similar to a D322* model ?
14:47:50darkskiezi cant find anything on the D332,want to know if i could use this
14:48:00darkskiezwith my dead iriver hdd
14:52:06*amiconn summons [IDC]Dragon .-.-.-.-.-.
14:54:32darkskiezmy dead iriver's hdd.
15:11:04 Join webguest15 [0] (
15:11:41webguest15hello, i have a problem
15:12:17ashridahwebguest15: what's up?
15:12:22webguest15the bootloader says checksum error withe the newest cvs version and starts the original firmware
15:13:04webguest15is this a known issue or just my problem?
15:13:27ashridahi can check out latest cvs and see, gimme a few
15:13:54webguest15k, thx
15:14:00B4gderyou probably didn't unount your player correctly
15:14:25ashridahsounds plausible to me
15:14:45webguest15ok ill try that
15:14:51B4gderseems very unlikely that the CRC would end up wrong in this particular build
15:14:56B4gder(not that I have checked)
15:17:18ashridahlatest cvs is definently working here
15:19:27webguest15ok thx, ill try around a bit
15:20:15 Quit webguest15 ("CGI:IRC (EOF)")
15:30:34 Join webguest35 [0] (
15:31:22 Join XavierGr [0] (
15:31:44webguest35for the sake of completeness: i dled the wrong build, h100 instead of h120.
15:31:58ashridahthat'd help
15:32:47 Quit Hadaka ("memory install")
15:32:47 Quit webguest35 (Client Quit)
15:33:00 Quit XavierGr (Client Quit)
15:36:07preglowi still don't get why things don't work after my macsr changes :/
15:36:34preglowwhat's the point in declaring an inline function static, btw?
15:37:05preglowi don't believe gcc has the habit of including a non-inline version of a function unless you enable an option
15:39:32 Nick QT_ is now known as QT (i=as@madwifi/users/area51)
15:43:16Zagordeclaring functions static only affects the naming scope, not the code per se
15:44:21 Join Naked [0] (
15:44:47 Nick Naked is now known as Hadaka (
15:45:31 Join Sagemaster2000 [0] (
15:45:37Sagemaster2000Hi all!
15:45:52Sagemaster2000Anyone on early on this eastern mornin?
15:46:18Sagemaster2000guess not :(
15:46:41amiconnpreglow: If you don't declare it static, the compiler includes a non-inline version in the object module
15:46:57amiconn...because then other modules might reference that function
15:47:10Sagemaster2000Oh there are people on...just not polite enough to say hello..........
15:47:28*Zagor learned something today too :-)
15:48:08ZagorSagemaster2000: this is a development channel. we're not here for idle chat.
15:48:13 Quit edx ()
15:48:52Sagemaster2000Im just simply being respectful and that is no excuse to be rude...developers have morals...and I was just leading into a question...
15:48:59 Quit matsl (Read error: 110 (Connection timed out))
15:49:29ZagorSagemaster2000: we are not rude. we answer all questions. we just don't respond to each and every statement.
15:50:12B4gderit would only become an awful lot of hellos
15:50:24Sagemaster2000Anyway, I DID notice the topic and I just wanted to know if any officials hinted on when the release will possibly occur because I have been following this for quite a while.
15:50:33 Join webguest48 [0] (
15:50:53B4gderbtw, linus has been locked out from internet all day, that's why he hasn't been around today
15:51:25preglowamiconn: that's what i was talking about, i thought that only happened if you explicitely told gcc to do so
15:52:37 Quit webguest48 (Client Quit)
15:53:36Sagemaster2000B4gder: Since you set the topic this morning does that mean that it won't be today?
15:54:04amiconnB4gder: Can't be "all day", he was in this morning for a few minutes
15:54:32amiconn[07:13:00] *** LinusN ( joined
15:54:32amiconn[07:17:28] <t0mas> g morning
15:54:32amiconn[07:18:06] *** LinusN ( left ()
15:56:26ZagorSagemaster2000: I'd say it's a safe bet it's not going to happen today.
15:57:08Zagorhowever I'm a bit out of the loop at the moment even though I have the big RELEASE button in my hand :-)
15:59:11Sagemaster2000Release button?
15:59:23ashridahnot Release, RELEASE
15:59:43Sagemaster2000Let me rephrase that then..
15:59:44Zagorrelease script, to be precise
15:59:54Sagemaster2000RELEASE BUTTON? lol
16:00:51Sagemaster2000Umm, I dont know if anyone is allowed to answer this but does anyone know if the 2,5 release will support the LCD On the Remote of the H-120?
16:01:43preglowit's not for iriver at all
16:01:46Zagor2.5 concerns the archos only. iriver is still in development.
16:01:49preglow2.5 is an archos relase
16:02:18 Join kurzhaarrocker [0] (
16:02:44Sagemaster2000I see, so i should come back in like 4 months and maybe something would have happened for the iriver port.
16:03:00*ashridah suspects that's going to be a very FAQ when 2.5 rolls out
16:03:03Zagorbut the dailies support the remote lcd, don't they? (/me reveals his ignorance)
16:03:13B4gderZagor: not really
16:03:18preglowSagemaster2000: things are happening all the time
16:03:24ashridahZagor: not really. just a placeholder image
16:03:27preglowSagemaster2000: it's just not formally released, feel free to use the daily builds
16:03:37Sagemaster2000I use the dailies all the time
16:03:44ashridahsomeone had a dodgy build with some remote support, but it wasn't complete, and it never got submitted to cvs
16:03:46kurzhaarrockerwps.c, in function long wps_show(), line 650 - shouldn't that return SYS_USB_CONNECTED instead of true? (which isn't really of type long)
16:04:28Sagemaster2000I absolutely love Rock Box the lack of remote is killing me, and to be honest the remote makes me feel safe since it would allow me to hide the player and have potential thieves think the remote is the player.
16:04:29***Saving seen data "./dancer.seen"
16:04:55B4gderfeel free to join in and make the remote happen
16:05:18 Quit Sucka (Read error: 104 (Connection reset by peer))
16:05:20Sagemaster2000I believe there was an unofficial release from a user that allowed the remote
16:05:33B4gderxaviergr has a patch for that, yes
16:05:36Sagemaster2000but I am not to sure about that
16:05:42B4gderbut we will not use that
16:05:55Sagemaster2000Any particular reason why?
16:06:02B4gderyes, its not the right approach
16:06:07Sagemaster2000I see.
16:06:14B4gderits a fast way to get a decent lcd shown
16:06:19B4gderbut not "properly"
16:06:41Sagemaster2000But its something as we wait for a more sophisticated approach no?
16:07:05B4gderif you think so then use his patch
16:07:18Sagemaster2000Im just thinking about my safety
16:07:33B4gderI don't see how that could harm you
16:07:36bluebrother^try the patched version if you like to.
16:07:46Sagemaster2000If I use the patch, I will be giving up my right to use the daily builds correct?
16:08:04Sagemaster2000Not right, but the ability to keep the patch and the newest build.
16:08:15 Join Sucka [0] (
16:08:36preglowdepends how good you are at patching things up when stuff collides :]
16:08:36bluebrother^you will probably have to wait for xavier to update his build. Or build it yourself.
16:08:46 Quit ashridah ("Leaving")
16:09:16bluebrother^depends on you. But you always could switch back to the daily builds :)
16:09:54Sagemaster2000B4gder: What I meant by safety is that I hate taking the player out from the belt case to change a folder or a setting. With the remote I can do all that without ever revealing the player, making me more safe in general...given the area I commute in.
16:10:05kurzhaarrockerapropos "build it yourself": Yesterday I tried to build gcc 3.3.6 using the knoppix dvd (I tried newlib, too) but failed. Has anybody tried that sucessfully?
16:10:39B4gderSagemaster2000: then the patch should make safer, doesn't it?
16:10:41B4gdermake you
16:10:57B4gderbut doesn't the remote keys work already?
16:11:06bluebrother^they do
16:11:20Sagemaster2000They do but there is no telling where the cursors are.
16:11:23B4gderenable the voice UI! ;-)
16:11:37bluebrother^it's only the remote screen.
16:11:43Sagemaster2000The voice thing is horrid
16:11:54amiconnkurzhaarrocker: Worked for me with the newlib workaround (debian)
16:11:58Sagemaster2000I pass through options and it says eveything.
16:12:03bluebrother^try a different voice?
16:12:08amiconn...and also worked for on cygwin without newlib
16:12:42kurzhaarrockerI read that in wiki, amiconn. That's what encouraged me to try it with knoppix. Without success.
16:13:02 Quit Sagemaster2000 ("CGI:IRC (EOF)")
16:14:18amiconnI never tried knoppix yet. What problem did you run in?
16:15:18kurzhaarrockerI can't tell right now as I'm @ work here. The build aborted with some errors when running make.
16:16:35amiconnPerhaps some required tools are missing. I think Knoppix must be rather limited as it is running from a single CD.
16:17:11kurzhaarrockerI tried the dvd version that even has eclipse with it. But still you're probably right.
16:17:40 Join DMJC [0] (
16:17:50 Quit Speedforneed (Read error: 104 (Connection reset by peer))
16:18:04DMJCcan the ihp-140 record from fm radio?
16:21:51kurzhaarrockerThe mic in jack of the ihp-1x0 devices is probably mono only?
16:22:10crwlisn't it stereo? the line in (in the same jack) definitely is stereo
16:22:28preglowit's the same plug
16:23:40*kurzhaarrocker is positivly surprised
16:31:33 Quit B4gder ("Lämnar")
16:35:11 Join matsl [0] (
16:50:41 Join OctalENPH [0] (
16:50:56 Join ender1 [0] (
16:51:28 Quit ender` (Nick collision from services.)
16:51:31 Nick ender1 is now known as ender` (
16:52:09OctalENPHDrew, we have these nifty things these days called "URL"s.
16:52:15OctalENPHEh heh, wrong channel.
16:55:33preglowdamn, you guys got urls
16:55:42OctalENPHWe're cool like that.
16:55:44OctalENPHCutting edge.
16:56:01preglowsomeday i too will use urls
17:01:58DMJCwow I finally mastered playlists
17:02:01DMJClike damn
17:03:42OctalENPHNow you just need to work on baiting. <_<
17:03:56OctalENPH..Yeah, I'm sorry.
17:05:23DMJCis there a way to update the bootloader once rockbox is on?
17:05:28DMJCfrom within rockbox?
17:07:53DMJCno biggie
17:07:57DMJCit works I guess
17:08:05DMJCspeed is something I'll live without
17:08:33bluebrother^if you want to update the bootloader you need to use the iriver fw to flash the firmware.
17:09:30DMJCmeh maybe later
17:09:36DMJCnot fussed atm
17:25:03 Join Mark_ [0] (
17:25:18Mark_could someone add a wps tag to check if hold is on?
17:25:41Mark_so those of us using conditional images to make their own status bar can have a full set?
17:25:45Mark_i think its the only one missing
17:35:51 Join Leperkawn [0] (
17:36:08 Quit Leperkawn (Client Quit)
17:44:08 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
17:45:51 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
17:48:13 Part kurzhaarrocker
17:53:33 Join _Wett [0] (
17:55:38 Nick Mark_ is now known as Mark_`afk (
17:58:00 Nick Mark_`afk is now known as Mark_ (
18:01:34 Join zezayer [0] (n=jake@
18:04:30***Saving seen data "./dancer.seen"
18:11:29 Join amiconn_ [0] (
18:12:13 Quit Wett (Read error: 110 (Connection timed out))
18:14:34 Quit linuxstb ("CGI:IRC")
18:17:32 Quit amiconn (Nick collision from services.)
18:17:32 Nick amiconn_ is now known as amiconn (
18:25:43 Nick _Wett is now known as Wett (
18:26:57 Join Lear [0] (
18:30:49Mark_could someone add a wps tag to check if hold is on?
18:30:50Mark_so those of us using conditional images to make their own status bar can have a full set?
18:30:54Mark_i think its the only one missing
18:32:01bluebrother^read this lines before ...
18:34:15Mark_a few dev-ish people came in, i though they might care
18:35:07bluebrother^we're in feature freeze, so I don't think this will come in the official cvs these days.
18:35:34Mark_a lot of wps shit's been added during the freeze
18:35:50Mark_a hold tag is so small its practically a bugfix :P
18:36:33bluebrother^really? Haven't followed the wps stuff the last time.
18:36:39 Nick Mark_ is now known as Mark_`away (
18:54:03 Join webguest59 [0] (
18:55:52webguest59i own a igp100 which seems to be practically the same hardware inside as the h1x0 so if i donated a igp100 to rockbox would it be much trouble to port the h1x0 version after its finished to the igp?
18:58:17preglowigp100 ?
18:58:49preglowi never heard of that
18:59:00bluebrother^preglow: found it on amazon germany
18:59:19webguest59erm its the 1.5 gb 'round ihp1x0'
18:59:26webguest59with the 1" hd
18:59:46webguest59but missing line in and therefore record
19:00:09webguest59hell the screen and even the config menus etc are all off the h1x0
19:00:44webguest59plays oggs wma mp3 etc like h1x0 too
19:00:45preglownew? old?
19:01:03webguest59hmm it came out a while ago. not sure if before or after the h1x0
19:03:11bluebrother^ is really annoying.
19:03:22webguest59how come
19:03:43bluebrother^ok, found it.
19:04:00 Join einhirn [0] (
19:04:35bluebrother^ is crappy because it is completely overloaded with flash, I need to search hours for 1 piece of information and in most cases the information I want isn't even there.
19:05:44webguest59hmm most times i just click no to all the permission requests
19:05:59bluebrother^but the device looks nice.
19:06:33webguest59yep, controls same way as h100 too but with buttons and knobs
19:06:53 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
19:07:51webguest59wouldnt say its exactly more conpact than some harddrive players but i run around with this and if the hw inside is same as h100 then it seems like a easy win for port
19:07:52preglowit _is_ flash
19:08:02webguest59ermm flash?
19:08:10webguest59no its a 1" hd
19:08:25webguest59you can hear the head moving and parking etc
19:08:47webguest59oh maybe you mean the site uses flash..
19:11:02*bluebrother^ meant the site using flash
19:12:06webguest59hey try going to and looking for idp-100
19:12:22webguest59ive never seen that before. must of died quietly
19:14:33bluebrother^what format is QDX?
19:14:55 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
19:15:21 Quit darkskiez ()
19:16:55webguest59never heard of it
19:17:16 Join LinusN [0] (
19:17:31bluebrother^the mentioned idp is stated to support it.
19:17:51webguest59"In addition to compression quality and user convenience features, QDX offers copyright owners the benefit of an unpublished format with integrated encryption and watermarking which contribute to preventing access to content by unauthorized players"
19:17:56webguest59from that page the link goes to
19:18:14webguest59maybe thats why havent heard much of it
19:18:34bluebrother^I guess so. And at least me doen't need this :)
19:18:42webguest59hey linus
19:18:52Mr_Chey, quick question, will there be an official 2.5 rockbox release for iriver ?
19:18:56 Join bagawk [0] (
19:19:04 Nick Mr_C is now known as [-AIR-] (
19:21:02webguest59hey does anybody here used to have an amiga??
19:21:29webguest59amiga 500/600/1200
19:21:58webguest59you remember the game scorched tanks? (not scorched earth or anything else, but tanks)
19:22:32webguest59not much fancy graphics or moving stuff, perhaps someone could make a version of scorched tanks for the iriver
19:22:40LinusNMark_`away: there are wps tags for the hold switches
19:22:53LinusNwe just forgot to document them
19:23:00LinusNi have added them to the wps docs
19:24:55webguest59LinusN: hi what are the chances of a port to igp100 based on h1x0 version of RB? the hardware seems to be the same basically other than smaller hd, no linein socket, and different button layout
19:25:21LinusNwebguest59: the chances are slim unless you start working on it
19:26:21webguest59k not even if i donated a igp100 player?
19:27:36LinusNwebguest59: donating one will of course help, but before you do that we need to know it the hardware really is similar
19:28:32LinusNotherwise you would have donated one in vain
19:28:35webguest59hmm realised that page is out of date cos and my own experience shows it can play oggs too
19:29:07LinusNopen it up and take digital photos of the circuit boards
19:29:25webguest59ok i can try
19:29:52 Join linuxstb [0] (
19:30:35linuxstbLinusN: What's the "WPS Conditionals" bug you've added to the ReleaseTodo page?
19:32:26LinusNit's a bug where you can't do %?Fc<> conditionals to test if the id3 info for the next file is available or not
19:32:26webguest59if this isnt the same ill eat my hat :) but yep ill look at opening it up. just hope i dont bollock something up doing it heh
19:33:35linuxstbSo it just applies to the %Fc tag? Is this on both archos and iriver?
19:34:31LinusNyes, but the iriver %Fc is totally broken because of a flaw in the playback code
19:34:46linuxstbI know about that - resetting codectype to zero?
19:35:19LinusNusing codec type as a flag if the codec is loaded or not
19:36:03linuxstbIs it an easy fix (for the Archos)?
19:36:20LinusNi don't know yet
19:37:15amiconnLinusN: Would you mind me changing the playback time calculation to frame based? (in order to avoid the roundoff error for 44.1/22.05/11.025 kHz)
19:37:29linuxstbShouldn't we just return NULL instead of the "???" string for codec type zero?
19:37:36LinusNamiconn: iriver?
19:37:44amiconnAll platforms
19:38:05amiconnI think this would be helpful for implementing the recording frame count estimation
19:38:17LinusNamiconn: how do you count frames on the archos?
19:39:02LinusNlinuxstb: no, then the enums don't work
19:39:10amiconnI don't want to count them, I just want to base the total playtime calculation on the #of samples per frame and the sampling frequency
19:39:24amiconn...instead of the msec based calculation we have now
19:39:48LinusNamiconn: i don't mind
19:40:12linuxstbWhat's wrong with measuring time in samples?
19:40:12amiconnThe playtime is just a realtime counter iirc, and I don't plan to change that
19:40:26LinusNwhat would the song length unit be then?
19:41:18amiconnlinuxstb: Nothing; it's just that the calculation is currently done differently
19:41:30LinusNamiconn: aha, you just want to change the calculation
19:41:37LinusNsure, go ahead
19:41:49amiconnI want to avoid the roundoff error
19:42:26amiconnThen I could add another parameter to the create_xing_header() - the playtime in seconds
19:42:54amiconnThe function would then use this time to calculate the frame count in case the frame count parameter is zero
19:43:36amiconnI could do this right now, but it would suffer from the same roundoff error as playback if I would build upon the frame_time field
19:44:23amiconnI have to look where these values are needed and/or generated in the iriver playback code. I might need help in doing this...
19:44:49amiconnMy goal is to eliminate the frame_time field, and have a frame_samples field instead
19:45:06amiconnI'm not sure whether this will work with all formats on iriver...
19:46:15amiconnLinusN: Btw, regarding recording, I did a number of test recordings for my new id3v2+xing header code, among which were some s/pdif recordings
19:46:38amiconnThe only recordings where I got bitshifted frames were −− s/pdif recordings!
19:46:50LinusNhardly a surprise
19:47:34amiconnI have no real idea what to do about it...
19:47:48LinusNneither have i
19:48:01amiconnI had two occasions where the frame bits were shifted 2 bit positions to the right
19:48:31amiconnUndoing the shift lead to perfectly valid frame headers
19:48:50amiconnI did my tests with timecode switched on
19:49:10LinusNto the right?
19:49:15LinusNso it added 2 bits?
19:49:23amiconnYes, to the right. I had to shift left to compensate
19:49:50amiconnOf course it could be that it was shifted 6 bits to the left, but I think this is less likely
19:51:24amiconnAn additional effect is that mem_find_next_frame() can't find a valid frame header, so file splitting happens at an arbitrary position
19:52:34LinusNof course
19:52:43amiconnI wonder whether we have enough CPU power to monitor all frames, and reset the mas if it starts delivering bad data
19:53:08amiconnWhile this would cause a short glitch, it would still be better than hours of invalid data
19:53:10LinusNi think we have, but it will be a major effort
19:54:26amiconnAnother observation is that recordings don't end at a frame boundary most of the time
19:54:35amiconnSplitting is frame exact
19:55:18amiconnI've had a look at the code, and it seems that it tries to stop the MAS in a defined order. However, the last frame is a few bytes short
19:55:42amiconnI think this is caused by the DMA buffer granularity of 30 bytes
19:55:50LinusNyes, the mas refuses to give us the last bytes
19:55:55amiconnThe mas won't give us the last partial buffer
19:56:15amiconnWe could do a frame header search at stop as well
19:56:36amiconnIt would be useful to have backwards search in addition to forward search....
19:58:21LinusNdefinitely a lot harder
19:58:42amiconnNot necessarily least for memory search
19:59:03webguest59linusn: ok i opened it up with some trouble. haha wonder how ill get it together again. how do you want the pictures linus?
19:59:30webguest59i noticed indeed that the chip also says scf5249
19:59:36LinusNas high resolution as possible, so i can see the markings on the chips
19:59:59LinusNthat's in fact all i need to know
20:00:04LinusNbut the pics would be nice
20:00:08webguest59hmm ok
20:00:12 Quit matsl (Read error: 110 (Connection timed out))
20:00:32LinusNnow if you are serious about donating one...
20:00:32webguest59youll have to forgive my amateur pics ill take
20:00:47webguest59i got a ok camera cannon a80
20:00:48amiconnLinusN: Of course the frame header matching would always be done forward, only the search for the first sync byte (0xFF) would run backward
20:00:58webguest59but crap at photos
20:01:11amiconn...starting at current_position -4 to compensate for the header length
20:01:44LinusNamiconn: the trick would be to determine if it's a real header or not
20:02:01LinusNthen we'd have to search another frame backwards and check the length
20:02:26amiconnI don't see how this is different from a forward search
20:02:27 Join matsl [0] (
20:02:57LinusNwell, we search 2 frames backwards and one forward
20:02:57amiconnAfaik we don't check the postion of the next header, do we?
20:03:16LinusNhmmm, no we don't
20:03:30LinusNas long as it's the same version and sample rate...
20:04:32***Saving seen data "./dancer.seen"
20:04:48amiconnYes, as long as it's the same version and sample rate as the supplied template header
20:05:20amiconnHmm, that's probably stuff that has to wait until after 2.5
20:05:21 Join darkskiez [0] (
20:06:07LinusNamiconn: guess so
20:17:28webguest59linusn: taken pics at 2000 resolution or something it was high. now checking them out
20:21:25webguest59seems ok except 1 chip that appears to say 'S3' or something at the bottom underneath the coldfire
20:33:19 Quit matsl (Read error: 60 (Operation timed out))
20:34:59LinusNwebguest59: good
20:35:10LinusNi gotta go, bbl
20:35:10webguest59i tyhink its sst
20:35:19webguest59where should i put pics?
20:35:22LinusNsst? probably the flash
20:35:32LinusNwebguest59: the wiki
20:35:33webguest59i retook some pics of the sst chip
20:35:43webguest59hmm dunnoi how to but ill ask others
20:35:51 Part LinusN
20:38:03 Join webguest85 [0] (
20:38:40webguest85webguest59: Are you registred on the wiki?
20:40:20webguest59sry was afk packing up
20:40:22webguest59no im not
20:40:31webguest85Register here:
20:41:06webguest85Then go to and click the "create" button
20:41:23 Join matsl [0] (
20:41:44webguest85There'll be help links at the bottom of the edit page on the syntax
20:41:49webguest85help for the syntax, that is
20:42:45webguest59ok registered
20:43:06webguest59do i need to click a link to confirm in an email or is it merely notification and i can get started?
20:43:21webguest85I don't think you have to confirm
20:44:10webguest59k well i got several more shots second tiem round as i got a pic of that sst chip so ill need to sort them out first
20:45:04webguest85Have a look at and for inspiration on how to arrange the page
20:46:05webguest59whoa my picture is nowhere near as clear as that single pic of the circuit board
20:46:17webguest59i had to take several pictures from different angles
20:46:21webguest85Better than nothing
20:46:46webguest85You can always try and get better shots if needed
20:47:08webguest59well i put it back together :)
20:47:14webguest59but ill try followthat pages example
20:47:26webguest59and put 'board images'
20:47:46webguest59first and then see from there where to go
20:47:56 Join TCK- [0] (
20:48:25webguest85Just add another section to that page
20:51:18 Join Mxm`Pas`Bien [0] (
20:51:18 Quit bagawk ("Leaving")
20:51:27 Quit Maxime (Read error: 104 (Connection reset by peer))
20:51:56 Join edx [0] (
20:52:01webguest59hmm perhaps you can look at these pictures if i send them to you? and tell me what you think?
20:52:12 Join DangerousDan [0] (
20:53:46 Join solex_ [0] (
20:56:44 Nick paugh is now known as AliasCoffee (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
20:58:07 Quit edx ()
20:59:23 Join bearin [0] (
20:59:36bearini have a Q:...
21:00:44bearinon recorder wps, is there a way to show the full playlist time
21:01:15amiconnRockbox has no idea about the playtime of tracks that are not yet loaded
21:01:17bearini know you can song track #/ total track (this is default)
21:01:47bearinwell, at least my question was solved
21:02:49 Quit bearin (Client Quit)
21:02:55Wettdoes anyone knows if there are any knows issues with pcm_playback using getmore callback ?
21:03:01webguest59for adding to wiki what is attribute 'h' in the table?
21:03:18webguest59does it mean they checked the hide in normal topic view box?
21:03:29 Quit TCK (Read error: 110 (Connection timed out))
21:03:44amiconnwebguest59: yes
21:04:11webguest59k thanks. i found out by clicking link to 'attribute' as well
21:05:21webguest59ok if i am uploading pics i take it the link box must be checked too? or is this an extra options to gather the links together in one section?
21:05:38webguest59oh nvm
21:05:44webguest59stupid didnt read it all
21:08:05webguest5912.5 meg rar file. some shots are quite bad, others not that bad in info they reveal about the chips (which seems to be basically that the major components and chips are all in h100). perhaps someone can check them out once they are uploaded and give feedback
21:09:30 Quit solex (Connection timed out)
21:11:49Wettor... what king of bugs can create randomly stkov audio, stkov ata, *panic*, etc. ? It's very strange..
21:13:37 Quit solex_ (Read error: 110 (Connection timed out))
21:14:13webguest59grrr wtf cant be that bad can it. my upload speed is crap
21:18:02 Quit matsl (Read error: 60 (Operation timed out))
21:20:57webguest59ah its done
21:21:05webguest59anybody here?
21:21:10 Nick Mxm`Pas`Bien is now known as Maxime (
21:21:16 Nick Maxime is now known as Maxime` (
21:22:06webguest59oh just not sure if i done it right the placement on the page somehow doesnt seem right at the bottom of in the 'Hardware' section
21:22:33webguest59but i managed to upload some shots of igp100 chips
21:22:58Maxime`what's the problem? o_O
21:22:59webguest59and was wondering if anyone wanted to check them out, comment on the pictures etc
21:23:06webguest59and i could try improve them
21:24:10 Nick Mark_`away is now known as Mark_ (
21:25:25Mark_linus, thanks very much
21:27:23 Nick AliasCoffee is now known as paugh (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
21:27:28webguest59you can notice philips tea, coldfire, sst flash, and some other components shown on that page in the shots, though not all at once since my lighting and camera focus wasnt quite right i think
21:28:30 Join matsl [0] (
21:34:08webguest85A good trick is to stuff it in a flatbed scanner if you have access to one
21:34:25webguest59:( no go there
21:34:47webguest59dont have one
21:35:05zeyou can get decent scanners for dirt cheap these days, why wouldn't anybody have one? heh
21:35:32webguest59well i dont have anything to scan :)
21:35:52zethere's always stuff to scan... case in point, IC's :p
21:36:27 Join Coldtoast [0] (
21:36:29zeintegrated circuits?
21:36:36zeyou're talking about chips on a circuit board no?
21:36:40webguest59well i didnt consider that
21:36:58webguest59i might need a scanner for such a purpose one day
21:37:24zeyou got a printer?
21:37:34webguest85Better lighting would be good
21:37:38webguest85They're a bit dark
21:38:00webguest85not that I have any clue what to do about them
21:38:02webguest85but they seem dark
21:38:32webguest59i thought so too. but i think some of the chip writing is faded so hard to notice
21:38:37webguest59like that SST chip
21:38:59webguest59that the page ssays is the flash
21:39:05zei haven't looked, but you can probably adjust levels and contrast to get them to show up better
21:39:32webguest59damn paintshop pro expired...
21:40:43webguest59i used it for like 600 days after the 28 or 30 day trial ended then i installed new version big mistake, it really was a trial and ended after a short while
21:41:13zeyou can probably find a registry entry to reset or something
21:41:24webguest59well uninstalled now
21:41:46webguest59if i could access my old computer i got photoshop and various stuff on there
21:42:55 Quit Coldtoast (Remote closed the connection)
21:52:46webguest59well comparing my pics, the SC5249 coldfire, TEA5767 FM chip, LVCH16245A 16-Bit Bus Transceiver, SST flash rom, MEC-AA3k4 30.0000 oscillator are all there. just seems to be all stuffs from h100 mainly. next time i must see what is under the lcd. the edges are stuck down with some sticky tape like stuff that also covers up more chips
21:53:34amiconnWould be very interesting to know whether the LCD is the same
21:53:44amiconnWhat's the resolution?
21:53:55webguest59hmm the picture is this:
21:54:00webguest59(lemme find it)
21:54:02webguest85How aobut firmware upgrade - how is that done?
21:54:43webguest59look at that picture
21:54:48webguest59if thats not he h100 screen...
21:55:02webguest59there have been no f/w upgrades for this
21:55:17 Nick Mark_ is now known as Mark_`afk (
21:55:28webguest85Hm.. that's a problem... is it possible?
21:55:36webguest59but the flash is the same i think (sst)would think it should be possible
21:56:44webguest85Is there a firmware upgrade menu?
21:57:24webguest59arg whats this bloody music
21:57:32webguest59i got that open heh
21:58:01webguest59ill check
21:58:26webguest85It'll be troublesome to say the least, if there isn't.
21:58:32webguest59yes there is the same f/w upgrade menu
21:58:35webguest59it is there
21:59:00webguest85Right.. still a problem if there is no available firmware file
21:59:36webguest59to look at and see how it does things?
21:59:56 Join edx [0] (
22:00:16webguest85No, to patch and load to enable a bootloader
22:00:42webguest85Like it's done on the h100
22:01:02webguest85We take a firmware upgrade file, add some code to it and flash it
22:01:06webguest59so it wouldnt be possible to have a standalone rockbox/complete replacement with no bootloader?
22:02:10webguest59or bootloader that enables a person to add the rockbox files in completely wiping out original f/w?
22:02:17webguest85Sure - it's just a problem creating a file that the firmware upgrade will accept, without having an example
22:02:58webguest59what is needed to take the f/w file off the flash rom for analysis?
22:03:33webguest85No idea - That probably won't help much anyway, since the firmware upgrade file is "packaged"
22:03:41webguest85"Nach den ersten Bildern zu urteilen, fällt das im Gerät eingebaute LCD-Display mit 160 x 128 Pixel sehr großzügig aus"
22:03:44webguest85160x128 it seems
22:04:34***Saving seen data "./dancer.seen"
22:04:51webguest59for resolution of igp screen?
22:04:59webguest59is that same as h100?
22:05:29webguest59so probably is the same
22:05:30webguest85Good sign
22:08:41 Join webguest90 [0] (
22:08:43amiconnWPS is looking identical to H1x0 too
22:08:43 Quit webguest59 ("CGI:IRC (EOF)")
22:09:24webguest90trying to look up anything about a f/w upgrade in past
22:13:44webguest90bah all i could find is a quote from a custoemr review on cnet "Customer Support could not give a date for the firmware upgrade to fix the above issue."
22:15:29 Join Mxm`Pas`Bien [0] (
22:15:48 Quit Maxime` (Read error: 104 (Connection reset by peer))
22:18:57webguest90"Doesn't play DRM WMAs, right now. iRiver web site states a firmware upgrade to play DRM WMAs will be available later this year. " hmm i wonder what happened to that expected f/w. disappeared into the electronic ether i guess if it was ever completed or started
22:19:55amiconnThe UI does indeed look like the H1x0 UI. Just tried radio and mp3 playback
22:20:12amiconnTook me quite some time to navigate this mess :/
22:20:21webguest90just tried? hmm you have one there somehow?
22:21:00amiconnNo, I have a H-140, and compared with the photos in the tomshardware article
22:21:27amiconnIt's the first time I tried the radio in the iriver firmware
22:21:27webguest90yeah i noticed the 'circle' button below the play and stop buttons on the right, would be for recording but here it only does eq changes
22:21:35webguest90ah ok
22:22:34webguest90well it seems at one point a f/w was planned. what progress was made who knows, but they decided to phase the igp out
22:23:21*HCl reads the topic
22:24:21 Quit Lear ("Chatzilla [Firefox 1.0+/undefined]")
22:24:44amiconnwebguest85: I'm quite convinced this is a H1x0 in disguise, with that small 1" HD and different casing
22:24:50webguest90hmm that 1" cornell harddrive... if i replaced it with something else... far far bigger
22:24:55amiconnI bet I also know the button equivalences
22:25:25webguest90two wheels on left and right can both be depressed
22:25:48webguest90the right one if you keep it pressed down takes you to normal h100 confog menus
22:25:52amiconnSame type of construction as on the H1x0 remote control
22:26:39amiconnThe right wheel is equals joystick left/right/click
22:26:49webguest90hmm the h10s use 1" hd dont they?
22:27:01amiconnThe left wheel equals joystick up/down and A-B button
22:27:19amiconnON/FUNC equals play/pause
22:27:33amiconnPWR OFF equals stop
22:27:44amiconnEQ equals record
22:28:00webguest90but since there is no line in on igp it only does eq
22:28:19amiconnYes, but if it's wired it could be possible to record from FM
22:28:30amiconn...something that the iriver fw doesn't implement
22:28:43webguest90i tried looking for that uda chip but i think i missed it. several other chips on top covered by the black sticky plastic aroundthe lcd
22:29:08 Quit edx ()
22:29:08amiconnFound the RAM chip somewhere?
22:29:59webguest85Wonder if it's 16mb or maybe 8mb?
22:30:29webguest90i wrote this earlier from irc log
22:30:32webguest90"<webguest59> well comparing my pics, the SC5249 coldfire, TEA5767 FM chip, LVCH16245A 16-Bit Bus Transceiver, SST flash rom, MEC-AA3k4 30.0000 oscillator are all there. just seems to be all stuffs from h100 mainly. next time i must see what is under the lcd. the edges are stuck down with some sticky tape like stuff that also covers up more chips"
22:30:34amiconnH-120/H-140 is 32MB, H-110 and H-115 is 16MB
22:30:46webguest90hmm anyway to check?
22:30:48webguest85amiconn: that's what i was thinking of
22:31:38webguest85I guess you could do some guessing based on how often the harddrive spins up while playing a large dir
22:31:44webguest85(of equally sized files)
22:32:01webguest90or perhaps a playlist onm constant play?
22:32:02webguest85except I don't know how iriver does buffering
22:32:10amiconnDepends on the buffering mechanism
22:32:28amiconnIf it uses the same silly scheme as H1x0, you wouldn't be able to tell
22:32:39webguest90hmm what silly scheme is that
22:32:51amiconnH1x0 always only buffers a single files
22:32:59webguest90then load new one?
22:33:03amiconnThat's why it can't do gapless
22:33:21amiconnWell, that's why we have, and love, rockbox!
22:33:35amiconn(one reason only)
22:33:40webguest90it would be amazing in that case if this didnt load 1 by 1
22:33:53webguest90because they go on about how you can jog with it etc
22:34:00dpassen1then try loading a very large mp3
22:34:08webguest90i know the one hehe
22:34:16amiconnDoes it play wav?
22:34:42webguest90should do
22:35:07amiconnWould be easier to use a large wav for testing - wav is large even when not too long, and bitrate is guaranteed to be constant
22:35:14webguest85Okay, then play a really large file (>64mb) and see
22:35:25webguest85Okay, wav is easier
22:35:32amiconnJust play a >64MB wav file and measure the time between 2 disk spinups
22:35:36webguest90hmm ok ill load a wave on
22:36:21amiconnOf course you should know the sample frequency, sample width and whether it's stereo
22:37:19webguest90hmm by coincidence havea 63 meg wav at hand
22:38:09webguest9048 khz/16bit/stereo
22:39:15webguest90hmmm is it me or no wav support... plays bloody 500 kbps ogg but no wav
22:39:38webguest90ok one more double check
22:40:30 Join edx [0] (
22:40:32webguest90gah no wav support i guess
22:40:44webguest90cbr mp3 of a whole album do?
22:41:04webguest85That'll work just ifne
22:41:17webguest85make it 192kbps or something
22:43:12 Join LinusN [0] (
22:44:41LinusNthe igp100 pics look really promising
22:45:01LinusNeven the BDM debug connector looks the same as the H100 series
22:46:28amiconnI wonder where the RAM is, and/or what the D7416498GGU chip is
22:47:00HClhow much storage does the igp100 have..?
22:47:09dpassen11.5 GB
22:47:39HClit looks pretty neat.
22:47:43amiconnLinusN: Where's the BDM connector (I'm curious)
22:48:47HClhow new is it
22:48:55 Quit TCK- (Read error: 110 (Connection timed out))
22:49:31webguest90there was news in march 04 they plan to phase igp out
22:49:33webguest85It's about as old as h100
22:49:38webguest85I think
22:49:42HClhm, okay
22:49:42webguest90 but i think h140 only come out in feb 04
22:50:03webguest85maybe I dont know
22:50:04LinusNthe bdm connector is the unpopulated 20-pin ZIF at the lower right corner of the cpu
22:50:30 Nick webguest85 is now known as flack (
22:51:10 Quit matsl (Remote closed the connection)
22:51:20amiconnLinusN: Ah.
22:51:48amiconnwebguest90: Does the igp-100 have a RTC
22:52:03webguest90dont know
22:52:21webguest90certainly it mimics the h100 f/w in that there is no option for it
22:53:03LinusNi doubt that it has an artc
22:53:06webguest90man making this cbr mp3 will take some time. compressing 550 meg wav
22:53:36HCldoes the iron thing on top actually work like a lock?
22:54:05flackwebguest90: does it have a clock?
22:54:33webguest90the 'iron' thing on top seems to be some light tough plastic with silvery finish
22:54:44webguest90no clock just like h100
22:55:26webguest90hmm i think id have to take it apart again to check (and even then not sure what im looking for)
22:55:42webguest90but i think probably not likely as they base design off h100?
22:55:59LinusNrtc is very unlikely
22:56:51flackI never saw the point of having an rtc in a dap anyway
22:56:51webguest9064% encoded... grrr
22:57:24HClan rtc is very handy when generating last played runtime data..
22:57:36webguest90hmm i suppose from 64 meg mp3 you can check for multiples of 8/16/32 by how many times it accesses?
22:59:04amiconnLinusN: A chinese site states the D7416498GGU might be a TI chip, D7416 series
22:59:21amiconnI didn't find anything about this @ ti...
23:00:03webguest90done, now time to upload to igp
23:00:27HClwhere are the pics of igp we're talking about?
23:00:34flackamiconn: wonder what they're doing on that page.. seems to be a naked muvo
23:00:36flackhcl: iriverinfo
23:00:39flack(wiki page)
23:00:43HClo.o; hm. okay.
23:00:49HCli went there but it didn't have a link to igp
23:00:50 Nick Mark_`afk is now known as Mark_ (
23:00:59amiconnflack: Yes, it seems they dissected a MuVo...
23:01:03flackthey're tacked at the bottom, a rar file
23:01:26flackamiconn: Sounds like "our" type of guys
23:02:58webguest90hmm 69 meg 192 cbr mp3 rough length 50 mins or so
23:03:02webguest90starting my test
23:03:23webguest90ah f*ck
23:03:31webguest90what idiot left volume a t 40
23:03:34 Join muesli- [0] (
23:03:44webguest90blew my ears off
23:04:07Mark_lol sux2bu :P
23:04:14 Join solex [0] (
23:04:42flackMan, seems the chinese are all about picking mp3 players apart
23:08:44webguest90hmm coming up to 5.48... which should be almost 8 meg at 1.38 meg/min
23:08:54OctalENPHSorry, wrong channel.
23:08:54webguest90yep and it accesses once
23:09:25webguest90it started accessing around 5:35 for 5+ secs or so
23:09:52webguest90hey guys it seems to be 8 meg...
23:10:18muesli-somebody interested in a german party spot (worth it!)
23:10:19flackCould be pre-buffer
23:10:24flacklet it go for a while longer
23:10:41flackjust to be sure
23:10:59webguest90what you mean by prebuffer?
23:11:10flackMaybe they buffer "half the ram" or something first
23:11:16webguest90ah right
23:11:25flackprobably not, but just to be more certain
23:11:55webguest90well 5:40 or so x 2 is 11:25-30 maybe i should watch for red hd access light
23:12:20webguest90should be 5:48 really but it started accessing at 5:35
23:12:42flackThey probably have some of the RAM used for other things
23:12:48webguest909:13 accesses for 9 seconds
23:12:52webguest90to 9:22
23:13:25webguest905:35... 9:13... hmm not exactly spread out is it
23:13:42webguest90this is a single cbr mp3 too
23:13:58webguest90by spread out i mean multiples of 5:35
23:14:48flackProbably 8mb minus something
23:15:32flackStill.. the lack of a firmware update file is upsetting
23:16:13LinusNthat would simplify things, yes
23:16:30webguest9012:49-13:00 access hd
23:16:35LinusNmy guess is that the D7416498GGU chip is the hard disk controller
23:17:19webguest90hmm from 5:35 its multiples of 3:40 ish??
23:18:26webguest90so if this is true... next is at 16:30
23:18:35webguest901 min to find out
23:19:05flackwhat's that, 5-6mb?
23:19:45webguest90bingo as i though
23:19:53webguest9016:30 access
23:20:12amiconnflack: 8 MB ram minus firmware code+other data
23:20:25 Join Wett_ [0] (
23:20:30webguest90seems first time it start playing the song it takes up to just under 8 meg for first access
23:20:31flackSounds reasonable
23:20:57webguest90cos 8 meg / 1.38 meg per min means 5:48
23:21:02webguest90 it accessed at 5:35
23:21:09LinusNwebguest90: does it have a firmware upgrade menu option?
23:21:12amiconnwebguest90: In fact I guess there was an additional access near the start, around 2 minutes
23:21:55webguest90subsequent accesses are at 3:40 ish or 5 meg
23:22:26webguest90ill restart test see if there is an access i missed at 2 mins
23:22:57flackLinusN: I asked - it has
23:22:57webguest90y it got f/w upgr menu linus
23:23:10flackBut no firmware updates were ever released, it seems
23:23:38webguest90one supposedly planned for wma drm support but nothing came of it since igp was phased out
23:24:34HClisn't that going to be tricky? without having a flashable firmware to start with
23:25:11HClwe could probably come a long way with dumping the flash with the bdm and reverse engineering how it flashes, but still.
23:25:19flackwith a bit of luck it could be similar enough to h100 to be useful
23:25:39amiconnI would expect the same scrambling algo...
23:26:10LinusNme too
23:26:33LinusNwe'll dump the mem with the bdm and disassemble it
23:26:53webguest90hmm odd... i stopped at 19:20 switched off
23:26:57webguest90switched on
23:27:02amiconnThis looks like a design reuse, just like archos reused the recorder design for the Ondio (with more modifications, I'd say)
23:27:14webguest90restarted... and it accesses at 20:10
23:27:24flackamiconn: apparently the other way round
23:27:39webguest90but now i restart from beginning 0:00 and switch off/ switch on
23:27:47webguest90but no access after 50 secs
23:27:49flackaccording to webguest90, igp-100 was pre-h100
23:28:10flackbut yeah, looks like massive design reuse
23:28:16webguest90wait.... but 16:30 + 3:40 = 20:10!!!
23:28:22 Quit Wett (Read error: 110 (Connection timed out))
23:28:32preglowdesign reuse is good
23:28:34preglowfor us, that is
23:28:34webguest90so even though i switched it off and it still needed to rebuffer all back into menory
23:29:01webguest90it still used old 16:30 + 3:40 to access at 20:10
23:29:14flackThat sounds.. strange
23:29:25flackoh well
23:29:29flackno telling what they're doing there
23:29:47webguest90switched on restarted at 19:20 but still used old 20:10 from before it was switched off
23:30:10webguest90ok i predict this will access hd in 1 min,,, at 5:35
23:30:27flackmaybe they buffer at some preset times
23:30:32webguest90hmm maybe
23:30:49webguest90or preset amount of memory is used up?
23:30:52flackor file-positions (size), more likely
23:30:59webguest90k here it comes...
23:31:11webguest90like i said it came
23:31:13 Quit Sucka ("a bird in the bush is worth two in your house")
23:33:33LinusNtime to sleep
23:33:36LinusNcu tomorrow
23:33:40 Part LinusN
23:35:57 Quit Zagor ("Client exiting")
23:39:55webguest90hmm dunno wtf that is in the 'stock photo' doesnt look right
23:40:13webguest90the wps or whatever screen
23:44:16 Join ashridah [0] (
23:45:06 Join Wett [0] (
23:45:48webguest90 $4.99 starting bid.. right im off. night
23:46:34 Quit webguest90 ("CGI:IRC (EOF)")
23:51:18 Join webguest96 [0] (
23:53:15webguest96Hi, Just thought you could use the FCC application details for the IGP-100:'QDMIGP100'
23:53:21 Quit webguest96 (Client Quit)
23:53:22 Join webguest96 [0] (
23:53:52 Quit webguest96 (Client Quit)
23:58:07flackThat's handy.
23:59:00 Nick Mark_ is now known as Mark_`away (

Previous day | Next day