Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2010-11-15

00:01:57 Quit ender` (Quit: In the beginning, there was nothing. And God said, 'Let there be Light.' And there was still nothing, but you could see a bit better.)
00:13:45 Quit DerPapst (Quit: Leaving.)
00:13:46 Quit wodz (Quit: Leaving)
00:16:25dodddummyi tried to browse the config file on the player from rockbox and it's frozen. how can i reset it?
00:16:40 Join JdGordon| [0] (
00:16:40 Quit JdGordon| (Changing host)
00:16:40 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
00:16:47krazykitdepends on the device. the manual should say how
00:20:01dodddummykrazykit, i didn't see it in the manual. but i just held the power button for a couple of minutes and that did the trick.
00:20:06dodddummynice manual by the way
00:21:22 Join dfkt_ [0] (dfkt@unaffiliated/dfkt)
00:22:32dodddummyif companies had smarts they'd use rockbox.
00:22:44*S_a_i_n_t gets an IRAM overflowed error building for Nano2G
00:23:29 Quit bertrik (Quit: :tiuQ)
00:24:00 Join jfc^2 [0] (
00:24:42saratogaTheSeven: we only have 2 level boosting in Rockbox, but some kind of 3 level scheme would be useful on a lot of targets
00:25:11 Quit dfkt (Ping timeout: 245 seconds)
00:25:48 Quit domonoky (Read error: Connection reset by peer)
00:26:29factorDoes rock box have a printer driver.
00:26:57factorso you can plug in your USB connector to a printer?
00:27:00amiconnsaratoga: We do have 3 levels, but boosting only switches between two of them
00:27:14saratogayes i know
00:27:20amiconnOn targets where core voltage is constant, more levels don't make sense
00:27:27keenerdfactor: Rockbox does not have USB host mode.
00:27:34factorahh ok
00:27:36 Quit stacker55 (Quit: ircN 7.27 + 7.0 for mIRC (2002/01/10 00.00))
00:27:51factorkeenerd, any work being done on one.
00:28:03amiconnOtoh, on a target with dynamic vcore, fine grained clock adjustment makes sense
00:28:09saratogaamiconn: on targets like AMS where the memory clock doesn't boost but the core clock does, it'd also be helpful
00:28:44saratogasince the CPU requires a lot more MHz boosted then unboosted to do the same thnig
00:28:59 Quit simonrvn (Ping timeout: 240 seconds)
00:29:06Tornefactor: do you mean a printer that accepts USB memory sticks/flash drives/etc? or just connecting it the same way you'd connect the printer to a PC
00:29:19Tornethe former should Just Work with rockbox, the USB device only needs to be a storage device
00:29:21keenerdfactor: Nope. Very few players even have usb OTG hardware, let alone host chips. Not going to happen. Though if I could hook a keyboard up to my player, I would not need a laptop for the most part.
00:29:23Tornewhich we are
00:29:33factorTorne:as a host
00:29:41saratogai was thinking of a system where the codec could request a higher boost speed (e.g. for APE) while by default it would just boost between the two lower clocks
00:29:46Tornethen no
00:29:53factorI agree
00:30:00factorthe sansa fuze would be a nice laptop
00:30:09factorTorne, ok
00:30:24S_a_i_n_tif you never wanted it to do anything except play audio, or tiny video...sure ;)
00:30:49keenerdS_a_i_n_t: Typing notes. And Lua.
00:30:52S_a_i_n_tit's a nice DAP, it's be an AWFUL laptop
00:30:59factorWell was wanting to make a D&D character sheet generator
00:31:16factorjust select your options via the screen and it would print out a pdf
00:31:33factorbut looks like that wont happen , but it could make a pdf file for the usb memstick option
00:33:18factorwould have liked the printer driver option since I could just carry the sansa and a small printer
00:33:28 Join ReimuHakurei [0] (~reimu@
00:33:42keenerdfactor: I think you'd be better off with a $100 Android tablet.
00:34:42 Quit ReimuHakurei_ (Read error: Connection reset by peer)
00:34:48factoris android java
00:35:38dodddummyfactor i think that's on the 'features never to be implemented' list
00:36:38dodddummysoon enough we'll have brain tops.
00:36:44dodddummywell maybe not soon enough
00:38:53factorcsansa can read pdf's correct? does it have the facility to save as a pdf?
00:39:13keenerdWe should be seeing players with USB OTG soon. There are plenty of photo frames and hard drives with it.
00:39:51saratogaamiconn: any chance you could find time to add some coldfire info here:
00:40:45saratogalots of devices have USBOTG
00:40:51keenerdEdit: There are plenty of players with usb otg. But it'll be a lot of work to write the drivers and do the port.
00:41:41 Quit Buschel (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630])
00:41:49Tornerockbox can't read PDF, no
00:42:52***Saving seen data "./dancer.seen"
00:43:19JdGordon|does anyone else want to look at bugfixes for tagcache?
00:45:01 Quit jdgord| (Quit: Bye)
00:45:05 Quit saratoga (Quit: Page closed)
00:45:59 Quit Azrael- (Quit: Leaving)
00:48:48factorthanks for the android tabel option
00:48:56factornice they are cheap
00:51:41 Quit Loto (Ping timeout: 276 seconds)
00:57:06 Quit chattr (Quit: gone)
00:57:30 Join chattr [0] (
01:01:13 Join Buschel [0] (
01:06:36 Quit Buschel (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630])
01:07:19preglowthe next time i'm i here pratling about "giving cygwin another try", just yell at me
01:09:47 Quit BHSPitMonkey (Read error: No route to host)
01:09:49keenerdpreglow: Enough ram for a linux VM?
01:10:35 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
01:11:10preglowkeenerd: certainly, and it's what i'm currently doing. in addition to uninstalling cygwin :>
01:19:17amiconnOh, hi preglow :)
01:19:35S_a_i_n_tPffffft! CygWin bashing.
01:19:37amiconnAny news wrt the rockbox resampler?
01:37:37 Join Loto [0] (~ctrlproxy@xbmc/user/Loto)
01:39:44preglowi didn't really get any wiser with my last tests, still don't have a good algo in mind
01:40:12preglowi pretty much just decided to take a look at the resampler the speex guy wrote, since he was so adamant it would be ok for our use
01:40:24preglowbut i haven't had time, and it's really not anything i need, so...
01:41:52preglowcurrently just working on getting my audio effects plug up and going, but i can't even seem to make full duplex audio work again
01:42:02preglowsince the api changed
01:50:53 Quit froggyman (Read error: Connection reset by peer)
01:51:43 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
01:58:38 Quit dfkt_ (Quit: -= SysReset 2.53=- Sic gorgiamus allos subjectatos nunc.)
02:01:47 Join mirak_ [0] (
02:23:50 Quit ReimuHakurei (Ping timeout: 255 seconds)
02:36:08kugelplayed with a different way of making lcd updates on android (using SurfaceView), not any better
02:36:38kugelit makes synchronous updates easier but that's it
02:42:53***Saving seen data "./dancer.seen"
02:44:24kugelotherwise it's equally fast (or slow) and cpu consuming
02:47:35 Join ReimuHakurei [0] (~reimu@
02:47:47*S_a_i_n_t wonders if there is a way to only display the "Custom View..." database menu if there is actually something there to display.
02:47:50JdGordon|kugel: do you have a samsung s?
02:48:22S_a_i_n_tanyone understand tagnavi.config enough to know if that is possible?
02:48:28JdGordon|one of the recent commits mentioned the S and i wasnt sure who had it
02:48:46JdGordon|I tried raa on my brothers S and the display is not very useable with its dpi
02:48:49 Join madalu [0] (~user@unaffiliated/madalu)
02:49:15kugela fellow student has one
02:49:38kugelit's about the same screen as the nexus one isn't it?
02:50:01kugelI found it very usable, but he complained that rockbox drains the battery a lot
02:50:54JdGordon|same resolution, bigger screen
02:51:03JdGordon|which should make it easier to use, but it looks smaller
02:51:08JdGordon|maybe it was just the font
02:51:20JdGordon|although even with the largets one it was crap
02:53:51 Quit liar (Quit: Leaving)
02:57:00JdGordon|kugel: have you looke at the resize patch yet?
02:59:03kugelwhat resize patch?
03:00:23JdGordon|dynamic screen size i mean
03:01:11kugelonly a bit
03:01:18kugelhow does it deal with themes?
03:02:35JdGordon|smartly :p
03:02:42JdGordon|well for cabbie anyway
03:03:30JdGordon|cabbie is he setting value, it then tries then
03:03:44JdGordon|ditto for the backdrop images
03:04:12kugelso all cabbie versions are installed?
03:04:27JdGordon|yeah, but we need to change that
03:04:30 Quit panni_ (Read error: Connection reset by peer)
03:04:32JdGordon|7MB apk is not very nice
03:04:45JdGordon|but really, neither is downloading 7MB of fonts on first load either
03:05:47kugel7MB apk means 21MB on disk
03:06:18JdGordon|actuslly it was worse... 40MB on disk before i moved the fonts and themes to the sdcard, then only 15
03:08:07JdGordon|we could put the theme downloader activity into the apk and then have it auto load on first run which would be a tad nicer
03:08:25JdGordon|and add smarts to only download the one required font for the cabbie for that screen szie
03:08:58kugelwe'd just have the right zips on the download server, has the smarts already
03:09:35JdGordon|slight pain to setup though, but still better than wasting heaps in the apk
03:12:21kugelI don't see a great advantage in resolution independant builds if themes are still resolution dependant, but well
03:12:54JdGordon|from a users POV it is much nicer
03:13:23JdGordon|well, unless we put each size on the market and make it only for that size which sucks for us
03:13:32JdGordon|but I want resizing for sdl anyway
03:15:36scorcheJdGordon|: if you drum up an 800x480 version, i can try it out on my G2 - it has a higher PPI than the galaxy s devices
03:16:09JdGordon|that is the other problem.... andorid currently forces portait
03:16:14JdGordon|is the g2 always landscape?
03:16:20scorcheof course not
03:16:41 Quit madalu (Remote host closed the connection)
03:17:09JdGordon| should work fine, except its 2 weeks old so doesnt have the battery life fix from last week(?)
03:18:32JdGordon|it would be nice to have a setting to chose landscape or portait
03:19:36scorchei think android users would generally be confused as to why there is a setting when most other apps auto-detect and rotate - i understand of course why we need a setting, but that is how the typical android user would probably feel
03:19:41kugelJdGordon|: most plugins will also be heavily resolution dependant I'm afraid
03:19:56JdGordon|no they wont
03:20:09kugelscorche: I've seen many apps which offer this setting
03:20:28kugelauto rotate is one of the options (and usually default) though
03:20:30 Join simonrvn [0] (
03:21:16JdGordon|we should be able to reload the theme on rotation though
03:21:21JdGordon|something worth playing with...
03:21:42JdGordon|and I dont see plugins coming to raa any time soon, not the graphical ones anyway
03:21:49kugelJdGordon: many plugins use bitmaps, I think those would be dependent, perhaps others too
03:23:59JdGordon|we could even get it to download the extras pack if needed (which would include screen dependant plugins)
03:24:27 Join Strife89 [0] (
03:25:45saratogasomeone on the forums says vorbis doesn't work on his Tegra 2 powered Android tablet
03:26:10S_a_i_n_tJdGordon: Is there a tag for "volume as percentile"?
03:26:16saratogaTegra 2 is a really weird ARM flavor in that it has armv7, a cortex a9 but no NEON
03:26:25saratogai wonder if it somehow lacks something else that our asm code needs
03:26:31 Join anewuser [0] (anewuser@unaffiliated/anewuser)
03:29:22scorcherockbox seems fine here
03:29:30scorcheis there a proper way to quit rockbox yet?
03:30:08kugelyou don't need to quit it
03:31:14scorchethe icon in the notifications bar can get annoying
03:31:25S_a_i_n_tJdGordon: I had assumed that it was possible to display volume as a percentile with a simple tag, but apparently not? the code must be there somewhat, for pregressbar style volume bars to work, but it seems like I have to do a really nasty long conditional statement like: %?pv<0|1|2|3|4|...|99|100|>
03:31:37S_a_i_n_tIs there a way to do that that doesn't suck?
03:31:49kugelscorche: stop music (not pause), then it goes away
03:32:55 Join InsDel [0] (
03:33:42JdGordon|S_a_i_n_t: yeah, no way to do it nicely
03:33:56S_a_i_n_tglarf. :/
03:34:15JdGordon|I can add it!'ll get no complaints here.
03:34:46JdGordon|I promise I will :D
03:34:47S_a_i_n_tI just discovered I wanted it, and it wasn't available.
03:34:55JdGordon|pixelma might just kill me D:
03:35:38kugelhaven't we discussed this a lot?
03:35:52JdGordon|various euopeans wanting to kill me?
03:36:17kugelpercentage for volume
03:36:30JdGordon|it used to be like that!
03:36:36S_a_i_n_tI'm really not sure what the problem is with adding the tag.
03:36:54S_a_i_n_tit's perfectly reasonable to want to display volume as a percentile
03:37:01JdGordon|no it isnt!
03:37:03saratogawe should just make all tags individually dependent on config file entries then people can turn off the ones they don't like
03:37:15JdGordon|saratoga: you mean, the skin file :p
03:37:29kugelsaratoga: we have that already
03:37:38saratogayeah but this way you can save 10 bytes of ram by disabling tags you don't want in the skin engine
03:38:24kugelS_a_i_n_t: percentage doesn't map easily to dB
03:38:52S_a_i_n_tprogressbar type volume display seems to do it well enough
03:41:00kugelwell, you're "supposed to" make the images a dB scale but most people don't
03:41:31S_a_i_n_tyou're not "supposed" to do anything with a theme
03:41:42S_a_i_n_tthat's the beauty of it, it's *your* theme
03:42:04kugelhence the "
03:42:39kugelbut with dB scale images it'll be beautyful and correct at the same time
03:45:53saratogaits really amazing how much I hate our test_codec sample file
03:45:57S_a_i_n_tmost users (speculation) expect a volume display (a percentile one at least) to be scaled from min-to-max volume...the real debate comes in when what 100% should indicate, 0db or "max volume"
03:46:00*JdGordon| remembers about making the peak meters skinable also
03:46:19Mode"#rockbox +o JdGordon|" by ChanServ (ChanServ@services.)
03:46:21Kick(#rockbox S_a_i_n_t ::specualtion and "most users" are not allowed in this channel) by JdGordon|!~jonno@rockbox/developer/JdGordon
03:46:22 Join S_a_i_n_t [0] (S_a_i_n_t@
03:46:24Mode"#rockbox -o JdGordon|" by ChanServ (ChanServ@services.)
03:46:25saratogaS_a_i_n_t: yeah but they also unknowingly expect the positions to be log spaced, not linear spaced
03:46:29 Quit feisar- (Ping timeout: 260 seconds)
03:46:30LloreanS_a_i_n_t: So is 50% half the range range or the point at which it's at half apparent maximum loudness?
03:46:32 Join simonrvn_ [0] (
03:46:39 Join feisar- [0] (
03:46:47saratogaif you linear space percentages people will think its broken :)
03:46:50LloreanIs 50% half as loud as 100%?
03:47:09saratogasince on linear scale 80, 90 and 100% have the same apparent loudness
03:47:31saratogaand 50% is just a tiny bit quieter then 100%
03:48:01saratogaactually since most of our devices use 1 or 1.5dB volume steps, 70, 80 , 100% would literally have the same volume
03:49:00 Quit simonrvn (Ping timeout: 240 seconds)
03:49:00 Nick simonrvn_ is now known as simonrvn (
03:50:20 Quit kugel (Remote host closed the connection)
03:50:36LloreanNot to mention none of our devices even *have* 100 steps do they?
03:50:57saratogano, but a lot of them don't line up exactly with the db scale we show anyway
03:51:13saratogasince so many have 1.5dB steps
03:51:34Lloreansaratoga: I was wondering about that. On a 1.5 step one should going down from 0db go "0, -2, -3, -5, -6, -8, -9"?
03:51:38saratogaso sometimes you'll change the voluem 1 dB on a lot of players and it doesn't actually change
03:51:47S_a_i_n_tJdGordon: Is there some way I can add "char X" to each menu entry, but have it right aligned without using spaces to align it artificially?
03:52:03S_a_i_n_tI'd like to add the ">" the iPod OF menu structure has.
03:52:23S_a_i_n_tand that's the most realistic way I have of doing it I think.
03:53:06JdGordon|not easily
03:53:18JdGordon|actually.... maybe
03:53:32JdGordon|yeah, no
03:53:54LloreanWhat happened to the icons-on-the-right idea?
03:54:09S_a_i_n_tthe line selector won't run over top of it
03:54:35S_a_i_n_tif the line selector can be placed overtop of icons one day...then that'll work fine.
03:55:23S_a_i_n_tI forgot the line selector doesn't go overtop of icons when you suggested, I felt rather defeated when I remembered it.
03:58:44JdGordon|if you *always* want extra text on th eline then the list could do it
03:58:49JdGordon|but you only want it in menus right?
03:59:22JdGordon|you could possibly do it in the menus list callback and just calculate how wide the current string is and how many spaces and > you need to concat
03:59:30JdGordon|but then scrollling lines will break
04:00:19 Join bluebrother [0] (
04:00:19 Quit bluebrother (Changing host)
04:00:19 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
04:00:36 Quit MethoS- (Remote host closed the connection)
04:03:19 Quit bluebroth3r (Ping timeout: 260 seconds)
04:05:51 Quit amiconn (Disconnected by services)
04:05:52 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:05:52 Quit pixelma (Disconnected by services)
04:05:54 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:05:56 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:06:11 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:07:09 Quit anewuser (Ping timeout: 245 seconds)
04:08:03JdGordon|S_a_i_n_t: you want icons on the right with colour inversion to do it properly?
04:08:09 Quit Strife89 (Quit: Switch!)
04:08:39 Quit InsDel (Read error: Connection reset by peer)
04:09:10S_a_i_n_tJdGordon: Icons would be the best way to do it, as the ">" char is only present on menu items that have a submenu...and there's an icon for that in Rb.
04:09:14 Join Strife89 [0] (
04:09:25S_a_i_n_tThe only shitter is that the line selector will mess it up.
04:09:46S_a_i_n_tI need it to pass over the icons as though they were a part of the actual menu entry
04:10:30S_a_i_n_tthat's why I asked about adding it to part of the menu entry string, as then it will do the appropriate colour change when selected also.
04:12:10LloreanDidn't the line selector used to pass over the icons? Since this is going to require a custom build anyway, why not make it do so again?
04:12:34 Quit Strife89 (Read error: Connection reset by peer)
04:12:47JdGordon|that could work
04:12:49 Join Strife89 [0] (
04:15:55 Join anewuser [0] (anewuser@unaffiliated/anewuser)
04:18:52JdGordon|of course, if you are doing a big custom build anyway you could just reimplement the list to just look like the of t begin with
04:25:18 Join leavittx [0] (~lev@
04:42:56***Saving seen data "./dancer.seen"
04:51:29 Join dys``` [0] (
04:51:29 Quit dys`` (Ping timeout: 276 seconds)
04:55:55 Quit TheSeven (Ping timeout: 245 seconds)
04:58:03 Quit ReimuHakurei (Ping timeout: 265 seconds)
04:58:57 Join froggyman [0] (
04:58:57 Quit froggyman (Changing host)
04:58:57 Join froggyman [0] (~seth@unaffiliated/froggyman)
04:59:59 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
05:01:37 Quit edboyer93 ()
05:07:50 Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier.
05:17:02alexbobPis there some kind of software hold for the clip+?
05:17:37alexbobPsaint told me there was earlier, but I just went through the manual and can't find anything about it
05:20:22mc2739alexbobP: clip+ software hold is: home+select
05:21:01*alexbobP tries it
05:21:43alexbobPworks, thanks!
05:21:45*alexbobP hugs mc2739
05:22:03saratogaits under 4.3.1. WPS Key Controls in the manual
05:22:09saratogaalthough not very easy to find
05:31:05saratogawow my new mp3 filterbank mostly works
05:31:09saratogaon the first try
05:31:26 Quit ps-auxw (Read error: Operation timed out)
05:37:29 Quit anewuser (Ping timeout: 265 seconds)
05:45:07saratogawell it mostly works but finding this bug is like a needle in a haystack :(
05:45:35 Join ps-auxw [0] (
05:55:58saratogaBuschel: FS #11759 - Rearrange libmad synthesis memory acceses for arm
05:56:08saratoganot quite working yet, but getting close
06:04:07 Quit Strife89 (Ping timeout: 240 seconds)
06:06:21dodddummythe main question is "who from the rockbox dev team should i propose to?"
06:09:14saratogapropose what?
06:21:28JdGordon|pick me!
06:21:35JdGordon|wait, can you cook?
06:25:51dodddummynot sure what. but i love them. and it's true love.
06:30:46 Quit S00row (Read error: Connection timed out)
06:42:59***Saving seen data "./dancer.seen"
06:50:59DEBUGEOF from server (Connection reset by peer) (snapshot: netstuff.c line 545)
06:50:59***No seen item changed, no save performed.
06:51:02***Started Dancer V4.16
06:51:02***Connected to on port 6667
06:51:02***Logfile for #rockbox started
06:51:03Mode"logbot :+i" by logbot
06:51:03CtcpVersion from frigg!~frigg@freenode/utility-bot/frigg
06:51:06***Server message 501: 'logbot :Unknown MODE flag'
06:51:06 Join logbot [0] (
06:51:06 Join crwl [0] (
06:51:06 Join aendruk [0] (
06:51:06 Join Dhraakellian [0] (
06:51:06 Join Zambezi_ [0] (Zulu@
06:51:06 Join jfc^3 [0] (
06:51:06 Join ps-auxw [0] (
06:51:06 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
06:51:06 Join froggyman [0] (~seth@unaffiliated/froggyman)
06:51:06 Join dys``` [0] (
06:51:06 Join leavittx [0] (~lev@
06:51:06 Join pixelma [0] (quassel@rockbox/staff/pixelma)
06:51:06 Join amiconn [0] (quassel@rockbox/developer/amiconn)
06:51:06 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
06:51:06 Join feisar- [0] (
06:51:06 Join simonrvn [0] (
06:51:06 Join S_a_i_n_t [0] (S_a_i_n_t@
06:51:06 Join mirak_ [0] (
06:51:06 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
06:51:06 Join Loto [0] (~ctrlproxy@xbmc/user/Loto)
06:51:06 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
06:51:06 Join chattr [0] (
06:51:06 Join JdGordon| [0] (~jonno@rockbox/developer/JdGordon)
06:51:06 Join Topy44 [0] (
06:51:06 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
06:51:06 Join mrkiko [0] (
06:51:06 Join factor [0] (
06:51:06 Join Rob2222 [0] (
06:51:06 Join dodddummy [0] (
06:51:06 Join JesusFreak316 [0] (
06:51:06 Join antil33t [0] (
06:51:06 Join grndslm [0] (
06:51:06 Join CaptainKewl [0] (
06:51:06 Join Kitar|st [0] (
06:51:06 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
06:51:06 Join timccc [0] (~timccc@
06:51:06 Join sasquatch [0] (
06:51:06 Join Horscht [0] (~Horschti@xbmc/user/horscht)
06:51:06 Join xavieran [0] (
06:51:06 Join pjm0616 [0] (~user@
06:51:06 Join Barahir [0] (
06:51:06 Join Zeiris_ [0] (~Zeiris@
06:51:06 Join markun [0] (~markun@rockbox/developer/markun)
06:51:06 Join merbanan [0] (
06:51:06 Join krazykit [0] (
06:51:06 Join AlexP [0] (~alex@rockbox/staff/AlexP)
06:51:06 Join gammy [0] (
06:51:06 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
06:51:06 Join avacore [0] (
06:51:06 Join Galois [0] (
06:51:06 Join Xerion [0] (
06:51:06 Join nano [0] (
06:51:06 Join elcan [0] (
06:51:06 Join jordan` [0] (
06:51:06 Join mystica555_ [0] (
06:51:06 Join mikroflops [0] (
06:51:06 Join keenerd [0] (
06:51:06 Join Guinness` [0] (
06:51:06 Join soap [0] (~soap@rockbox/staff/soap)
06:51:06 Join KiwiCam [0] (
06:51:06 Join riotz [0] (riot33r@gateway/shell/
06:51:06 Join CIA-7 [0] (~CIA@
06:51:06 Join Kohlrabi [0] (
06:51:06 Join captainkewll [0] (2669ecc2@gateway/web/freenode/ip.
06:51:06 Join Farthen [0] (
06:51:06 Join simabeis [0] (
06:51:06 Join bzed [0] (
06:51:06 Join alexbobP [0] (
06:51:06 Join literal [0] (
06:51:06 Join iq [0] (~iq@unaffiliated/iq)
06:51:06 Join FOAD [0] (~dok@
06:51:06 Join ved [0] (
06:51:06 Join scorche [0] (~scorche@rockbox/administrator/scorche)
06:51:06 Join thegeek [0] (
06:51:06 Join dionoea [0] (
06:51:06 Join advcomp2019_ [0] (~advcomp20@unaffiliated/advcomp2019)
06:51:06 Join z35 [0] (
06:51:06 Join shai [0] (
06:51:06 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
06:51:06 Join Hadaka [0] (
06:51:06 Join TBFOOL [0] (
06:51:06 Join yosafbridge [0] (
06:51:06 Join Bagder [0] (~daniel@rockbox/developer/bagder)
06:51:06 Join linuxguy3 [0] (
06:51:06 Join GodEater [0] (~bibble@rockbox/staff/GodEater)
06:51:06 Join rvvs89 [0] (
06:51:06 Join parafin [0] (
06:51:06 Join kadoban [0] (
06:51:06 Join Utchybann [0] (
06:51:06 Join powell14ski_ [0] (
06:51:06 Join noamsml [0] (
06:51:06 Join cjcopi [0] (
06:51:06 Join BlakeJohnson86 [0] (
06:51:06 Join rasher [0] (~rasher@rockbox/developer/rasher)
06:51:06 Join Amec [0] (
06:51:06 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
06:51:06 Join YPSY [0] (
06:51:06 Join scorche|sh [0] (
06:51:06 Join Battousai [0] (~bryan@gentoo/developer/battousai)
06:51:06 Join zu [0] (
06:51:06 Join miceh [0] (
06:51:06 Join sinthetek [0] (~sinthetek@unaffiliated/sinthetek)
06:51:06 Join tbox [0] (~tbox@
06:51:06 Join jepler [0] (~jepler@emc/developer/pdpc.professional.jepler)
06:51:06 Join aevin [0] (eivindsy@unaffiliated/aevin)
06:51:06 Join preglow [0] (
06:51:06 Join docgnome [0] (~docgnome@
06:51:06 Join n17ikh [0] (
06:51:06 Join tempname [0] (
06:51:06 Join jae [0] (
06:51:06 Join MagusG [0] (
06:51:06 Join Sudos [0] (
06:51:06 Join Rondom [0] (
06:51:06 Join tmzt_dg2root [0] (~tmzt@
06:51:06 Join ehntoo [0] (
06:51:06 Join Stummi [0] (
06:51:06 Join ranmachan [0] (
06:51:06 Join lostlogic [0] (~lostlogic@rockbox/developer/lostlogic)
06:51:06 Join Torne [0] (torne@rockbox/developer/Torne)
06:51:06 Join ThomasAH [0] (
06:51:06 Join knittl [0] (~knittl@unaffiliated/knittl)
06:51:06 Join @ChanServ [0] (ChanServ@services.)
06:51:06 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
06:51:06 Join Beta2K [0] (
06:51:06 Join ack [0] (
06:51:06 Join gevaerts [0] (~fg@rockbox/developer/gevaerts)
06:51:32 Quit JdGordon| (Quit: leaving)
06:51:56aendrukHello, I'm new. Would someone mind adding AndrewKvalheim to WikiUsersGroup?
06:54:05 Join ReimuHakurei [0] (~reimu@
06:54:22 Join maraz [0] (
06:55:03saratogaaendruk: added
07:03:54 Join S00row [0] (
07:05:52 Join JdGord [0] (~jd@
07:08:08 Part aendruk ("Leaving")
07:22:31 Quit saratoga (Quit: Page closed)
07:25:57 Join esperegu [0] (~quassel@
07:26:57 Quit JdGord (Ping timeout: 250 seconds)
07:28:06 Join n1s [0] (~n1s@rockbox/developer/n1s)
07:32:06dodddummywholly smack! it's got games, too?
07:35:09 Quit Horscht (Ping timeout: 265 seconds)
07:35:17 Join Horschti [0] (~Horschti@xbmc/user/horscht)
07:35:28 Join Buschel [0] (
07:51:11 Quit n1s (Quit: Lämnar)
07:51:20 Join stoffel [0] (
07:56:14BuschelTheSeven: I also played around with mp3, clocking and the buffering_thread debug screen. if you use the unboosted clock, mp3 needs ~43 MHz (220 kbps file). it can be played unboosted, if you simply reduce the frame rate of the debug screen from HZ/5 to HZ/2.
07:56:32 Quit BHSPitMonkey (Remote host closed the connection)
07:56:36BuschelTheSeven: the same was happening with the iPod Video that also has a slow screen driver.
07:58:29BuschelTheSeven: with the LCD sleep patch and a sleeping LCD (which also suppresses screen updates), there should be only few −− or even none −− boosts left on high bitrate files or when using software equalizer
08:09:21 Join fuzzybunny69y [0] (
08:11:08fuzzybunny69yHey guys! I am using Rockbox 3.7 and for some reason when I hook it up to some external speakers the volume is really low even if I put the volume on the ipod(30gb) and the external speakers to max. Does anyone know of anything I could do to correct the problem?
08:13:58sinthetekthe speakers sound ok when connected to other devices and/or the original firmware?
08:14:44Buschelfuzzybunny69y: maybe the line out is disabled (one of the system settings) or you have the ipod set to low volume (volume setting also changes the line out, if set to lower than -56 dB)
08:16:57 Join liar [0] (
08:17:05fuzzybunny69ysinthetek, hmm that is a good test I tried my brothers ipod which is the exact same model and the original firmware was fine but i will have to copy some songs onto the origin firmware
08:17:50fuzzybunny69yBushel it is connected using the bottom connector(the data cable) not through the headphone jack
08:18:14Buschelfuzzybunny69y: yes, that's clear. my comment is valid for the connector.
08:19:40Buschelmaybe you also need to switch on Accessory Power Supply?
08:19:41 Quit S00row (Read error: Connection reset by peer)
08:21:15 Join S00row [0] (
08:21:17fuzzybunny69yBushel ok I will give that I try ;)
08:23:07 Quit scorche (Ping timeout: 240 seconds)
08:24:44fuzzybunny69yoh wow Buschel I changed the line out setting and it is a lot louder now!!
08:24:49fuzzybunny69yhow nice
08:25:11fuzzybunny69ythanks so much you guys
08:25:13fuzzybunny69yyour so great
08:27:37Buscheljust keep it disabled, if you do not need it. it saves some power.
08:27:38 Quit JesusFreak316 (Ping timeout: 245 seconds)
08:28:26 Quit Buschel (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630])
08:28:29TheSeven[23:24] <Buschel> TheSeven: if I use FastMode, is CPU clock set to HCLK or to PCLK ? << FastBus? => HCLK
08:29:46TheSevenfrom a hardware point of view we could do a 2-level scheme for the system bus, and linear scaling for the CPU
08:33:37 Quit fuzzybunny69y (Quit: Leaving)
08:35:42CIA-7New commit by bluebrother (r28602): Set svn:mime-type on cabbie image files lacking it.
08:36:13 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
08:37:29CIA-7r28602 build result: All green
08:44:24 Join JdGord [0] (~jd@
08:46:42 Join ender` [0] (
08:50:31 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
08:50:41 Join hebz0rl [0] (
08:51:04***Saving seen data "./dancer.seen"
08:58:08 Quit leavittx (Ping timeout: 255 seconds)
09:08:54pixelmaLlorean: "Not to mention none of our devices even *have* 100 steps do they?" Archos devices have even more (-100dB to +12dB), except maybe the Player). And about the line selector going "over" the icons, it ones did yes, but is gone for a long time since some rework and I believe it was before adding gradient line selectors, maybe even line selector colours at all. I guess it won't be easy to "just" add it again
09:13:18amiconnPlayer has -78 dB to +18 dB, in 1.5 dB steps, below -54 dB even in 3 dB steps
09:15:21amiconnBut then expressing volume as a percentage is nonsense anyway
09:16:26JdGordAdding line selectors under the icon probably would be trivial
09:16:36JdGordOr at least easy enough
09:18:20pixelmawell, and why did it go?
09:21:23 Quit bertrik (Quit: :tiuQ)
09:24:08JdGordYeah should be pretty simple. Pretty sure it's been that way at least as before I was interested in GUI code
09:25:37amiconnAfair the selection bar never overlapped the icon
09:25:44pixelmaI'm sure it did
09:26:09pixelmaand was changed during some rework, I believe when icons became bitmaps or somesuch
09:26:11 Quit S00row (Read error: Connection reset by peer)
09:26:38pixelmaI mean "external" ones
09:26:49amiconnIi don't think so
09:27:17amiconnMaybe when the selector bar got different modes on colour targets
09:27:34amiconnBut then I still think it never did...
09:28:04pixelmaI find the discussion silly as no-one has facts and won't continue this way
09:28:09 Join S00row [0] (
09:29:11 Join xxcv [0] (
09:29:14*S_a_i_n_t is pretty sure that SVN has facts on what used to be/didn;t used to be.
09:29:18 Join einhirn [0] (
09:29:54 Nick S_a_i_n_t is now known as [Saint] (S_a_i_n_t@
09:30:45pixelmaprobably but it won't be easy to find
09:31:29 Join scorche [0] (~scorche@rockbox/administrator/scorche)
09:32:27[Saint]svn diff <revision you're pretty sure is old enough to have the feature/change> > patch.patch
09:32:34[Saint]how far back am I looking?
09:32:50 Join Rob2223 [0] (
09:34:25pixelmaand what's your diff supposed to do? If you want to try older versions, just check those out
09:35:06pixelmaI doubt you can spot such a change in a mega patch you get with that. If you are lucky it is mentioned in a commit message
09:35:17 Join petur [0] (d408b802@rockbox/developer/petur)
09:36:12 Quit S00row (Read error: Connection reset by peer)
09:36:26[Saint]'twould be pretty easy to filter out the junk.
09:36:38[Saint]it's not like noone knows the rough area to start looking in the code.
09:36:41 Quit Rob2222 (Ping timeout: 276 seconds)
09:37:47 Join S00row [0] (
09:38:20pixelmaerrr... I bet the structure has changed a lot since then. If you think you can, I won't hold you back
09:44:32 Join swilde [0] (
09:45:26 Join LIONFURY [0] (
09:45:27 Quit factor (Read error: Connection reset by peer)
09:46:23 Quit LIONFURY (Client Quit)
09:47:16mrkikoit's the morning here :)
09:48:09 Quit mrkiko (Quit: leaving)
09:57:52 Join LinusN [0] (~linus@rockbox/developer/LinusN)
09:59:34 Join lestatar [0] (
10:02:03 Quit sasquatch (Quit: WeeChat 0.3.2)
10:02:27 Join sasquatch [0] (
10:02:32 Join factor [0] (
10:09:59 Quit xavieran (Remote host closed the connection)
10:11:57 Join jdgord| [0] (~jd@
10:12:36 Quit stoffel (Ping timeout: 260 seconds)
10:15:28 Join xavieran [0] (
10:15:32 Quit JdGord (Ping timeout: 255 seconds)
10:28:33 Join lixxus99 [0] (
10:29:24 Quit lixxus99 (Remote host closed the connection)
10:31:48 Join lixxus99 [0] (
10:32:54 Part lixxus99
10:33:07 Quit TheSeven (Ping timeout: 240 seconds)
10:34:51 Join DerPapst [0] (~Alexander@
10:40:48 Join crow [0] (
10:49:16 Join casainho [0] (
10:51:05***Saving seen data "./dancer.seen"
10:59:30 Quit xxcv ()
11:13:50 Join stoffel [0] (
11:36:56 Quit esperegu (Remote host closed the connection)
11:39:50 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
11:49:07 Join dfkt [0] (dfkt@unaffiliated/dfkt)
11:53:03 Quit [Saint] (Ping timeout: 255 seconds)
11:53:43 Join S_a_i_n_t [0] (S_a_i_n_t@
11:59:29S_a_i_n_tTheSeven: where is the main stack defined for Nano2G?
11:59:32S_a_i_n_tI seem to have lost it.
12:00:04TheSevenlook for ". += 0x4000"
12:00:54S_a_i_n_taha I would have found it but I grep-failed
12:02:18 Quit scorche (Ping timeout: 276 seconds)
12:06:01 Quit Zeiris_ (Read error: Connection reset by peer)
12:06:37 Join Zeiris_ [0] (~Zeiris@
12:08:08 Join InsDel [0] (
12:09:23S_a_i_n_tTheSeven: What would you suggest I change the stack to to give me enough room for aa fonts?
12:09:39TheSeven3e00 works
12:12:45 Join scorche [0] (~scorche@rockbox/administrator/scorche)
12:15:09 Nick S_a_i_n_t is now known as [Saint] (S_a_i_n_t@
12:15:21 Quit jdgord| (Quit: Bye)
12:18:03 Quit scorche (Ping timeout: 264 seconds)
12:20:20 Join scorche [0] (~scorche@rockbox/administrator/scorche)
12:32:09 Join teru [0] (
12:36:10 Quit stoffel (Remote host closed the connection)
12:36:43 Join stoffel [0] (
12:43:38JdGordondo we have a setting to draw icons on the right already?
12:45:57 Quit InsDel (Read error: Connection reset by peer)
12:47:53[Saint]JdGordon: Not that I know of.
12:49:10JdGordonI'm having trouble finding how the selector is actually drawn, I think what needs to happen is the list needs to draw the rectandle (filled or whatever) and then the text ontop of that instead of leaving it to the lcd text drawing code
12:49:25JdGordonwhich means scrolling might cause problems
12:49:48[Saint]scrolling always seem to be the bain of cool list stuff. :/
12:50:20[Saint]a themed list/line selector would be awesome.
12:51:08***Saving seen data "./dancer.seen"
12:51:20JdGordonactually maybe not...
12:53:10JdGordonyeah, putsxyofs_style() needs fixing... (IIUC) it only draws under text even if the x offset is inside the viewport
12:53:25JdGordonit should be able to make that use the whole viewport instead of just the text width
13:01:42JdGordonthat may not actually be all that difficult... just a tiny bit of refactoring
13:09:50 Quit dodddummy (Quit: Leaving)
13:15:22 Join JdGordon1 [0] (
13:16:54 Quit JdGordon (Ping timeout: 255 seconds)
13:18:31 Quit JdGordon1 (Client Quit)
13:19:45 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
13:25:51 Quit JdGordon (Ping timeout: 264 seconds)
13:32:49 Quit Zeiris_ (Quit: Leaving)
13:33:49CIA-7New commit by teru (r28603): FS #11751: increase buffer size of settings to store path.
13:36:12CIA-7r28603 build result: All green
13:42:38 Quit TheSeven (Ping timeout: 240 seconds)
13:43:42Stummiteru, btw, I splittet the rockboy-patch into 5 files now and attached them to a comment to the task. Is this ok so?
13:45:49Stummi(I hope, you excuse my mistakes. Rockbox is the first time i am contributing to an OSS-Project :) )
13:48:33 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
13:49:39terui think it's ok.
13:50:21Stummiok, thanks
14:01:24 Quit antil33t (Read error: Connection reset by peer)
14:01:34 Join antil33t [0] (
14:20:17 Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
14:22:13JdGordonteru: you probably should have used MAX_PATH instead of MAX_PATHNAME
14:22:22JdGordon80 is just as arbitrarily wrong as 32 was
14:23:48 Quit crow (Quit: ( :: NoNameScript 4.22 :: ))
14:23:56 Quit hebz0rl (Quit: Leaving)
14:29:18 Join crow [0] (
14:29:19 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
14:31:07 Join LambdaCalculus37 [0] (~3f74f70d@rockbox/staff/LambdaCalculus37)
14:36:03LambdaCalculus37Bagder: Ping
14:40:54 Quit shai (Quit: Leaving)
14:48:38 Join Kitr88 [0] (
14:49:57 Part LinusN
14:50:27 Quit Kitar|st (Ping timeout: 240 seconds)
14:50:31 Join lixxus99 [0] (
14:51:09***Saving seen data "./dancer.seen"
14:51:12 Join leavittx [0] (~lev@
14:53:48 Quit Kitr88 (Ping timeout: 272 seconds)
14:58:13 Join Kitar|st [0] (
15:09:53 Quit teru (Quit: Quit)
15:14:18 Join casainho_ [0] (~chatzilla@
15:15:48 Join jgarvey [0] (
15:22:28 Join wodz [0] (
15:24:03wodzMPIO engineers are the masters of complicate things. Slide strip of the HD300 output signals like rotary encoder. Why the hell the didn't use functions provided by synaptics chipset present in touch module.
15:28:23wodzTouchstrip, when pressed at the very bottom provides stable two square waveforms shifted by 90o by each other with the frequency of around 20Hz. On the rising edge of gpio6 the gpio7 is low. When pressed at the very top on the rising edge of gpio6 the gpio7 is high. Scrolling changes frequency and proportion of high to low periods (but this may be artifact of my oscilloscope).
15:29:03 Join anewuser [0] (anewuser@unaffiliated/anewuser)
15:30:30 Join panni_ [0] (
15:31:54wodzok at least they connected this to gpios with interrupt capabilities
15:36:15 Join b0hoon [0] (~quassel@
15:39:08 Join JesusFreak316 [0] (
15:45:17 Join komputes [0] (~komputes@ubuntu/member/komputes)
15:45:20 Join hebz0rl [0] (
15:48:06LambdaCalculus37b0hoon: Have you played around with the touchpad code for the HDD63x0 as of late? I've noticed that the touchpad's behavior is still a bit funky.
15:48:32LambdaCalculus37b0hoon: Sometimes the touchpad becomes non-responsive if you press the View button.
15:52:09b0hoonLambdaCalculus37: yes, i've had some sporadic problems with it, but i wasn't able to reproduce it, the view button is handled with the same method like the other buttons, it's a normal button packet.
15:52:21 Quit JesusFreak316 (Ping timeout: 240 seconds)
15:54:29LambdaCalculus37b0hoon: But for some reason, and I can't quite place my finger on it, pressing View causes the other buttons to start behaving strange.
15:55:49LambdaCalculus37b0hoon: For example, on mine, both the PREV and the Left buttons stop responding in the WPS after I press View.
15:56:01b0hoonLambdaCalculus37: i think these sporadic problems may be related to the touchpad driver or the hardware, this synaptics is very weird, comparing to the other targets
15:56:16LambdaCalculus37b0hoon: Yes, lowlight mentioned that to me once.
15:56:44LambdaCalculus37b0hoon: I'm going to email him again later on to ask him if maybe he has any bits of advice to give.
15:57:45wodzheh weird synaptics module :-)
15:58:01b0hoonLambdaCalculus37: strange.. i don't remember exactly but the View button worked for me however it stoped working under the other actions
16:00:29b0hoonLambdaCalculus37: ok, i must test it a bit more, find a solution, but it's hard to reproduce for me.
16:02:19 Quit CaptainKewl (Ping timeout: 276 seconds)
16:03:54b0hoonLambdaCalculus37: i'll try to play a bit more with the View button, thanks.
16:06:25LambdaCalculus37b0hoon: No problem. :)
16:06:42 Join kevku [0] (~kevku@2001:7d0:0:f000::135d)
16:06:49LambdaCalculus37wodz: Yeah, bit of a weird behavior on my HDD6330's touchpad. My HDD1630 doesn't exhibit this behavior.
16:07:43LambdaCalculus37b0hoon: I know that most of the HDD6330 port's code was just adapted from the HDD1630's port due to the similarities between the two. I wonder if lowlight probably hasn't cleaned out some HDD1630-specific bits from the HDD6330 port yet.
16:08:14 Part Zagor
16:09:00 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
16:10:52b0hoonLambdaCalculus37: no, the touchpad code was the same (almost) on the beginning, it was changed by me :P according to my findings
16:11:05 Join freddyb [0] (
16:12:23b0hoonLambdaCalculus37: btw: i don't think that these players are similar, except maybe SOC, RTC, everything else is different
16:14:06 Quit TheSeven (Remote host closed the connection)
16:14:08 Join MethoS- [0] (~clemens@
16:14:39LambdaCalculus37b0hoon: The ports were similar only in the sense that most of the same code ran on both targets in the beginning. The HDD16x0 port was first into SVN, and in order to compile for the HDD63x0, you had to just change the source to use 64MB RAM, 220x176 LCD and that was it.
16:14:56LambdaCalculus37b0hoon: But this was *waaaaay* back when the port started about 2 years ago.
16:16:12 Quit casainho_ (Quit: ChatZilla 0.9.86 [Firefox 3.6.10/20100922073514])
16:18:41b0hoonLambdaCalculus37: yes i know. but when you wan't to run everything right now on the hdd6330 the code must be (had to be - touchpad) strongly adapted for it and changed.
16:20:24freddybOn a Fuze V1 would cutting the boost frequency from 248MHz to 186 give any significant power savings? Keeping it under 200 means the voltage doesn't have to be boosted and then there's no need to wait for voltage adjustment.
16:21:18LambdaCalculus37b0hoon: Yup.
16:21:58freddybI'm also playing around with a lower pclk and it seems to work fine... Anyone know a reason not to do this?
16:24:51 Quit b0hoon (Quit: Back to work.)
16:29:59 Join TheSeven [0] (578b86bd@rockbox/developer/TheSeven)
16:30:47Zagorfreddyb: it's all just a balancing act. too low idle clock can affect ui rsponsiveness.
16:31:19Zagorlowering boost could mean some cpu-intensive codecs fail to run.
16:32:29Zagorbut of course avoiding voltage boost is a benefit too
16:36:25freddybWhich codecs are more demanding?
16:42:35 Part lixxus99
16:43:20 Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.
16:43:57saratogafreddyb: its worth looking into
16:45:16freddybI've been playing around with my Fuze v1 but I've only checked vorbis and mpeg video so far.
16:46:32saratogaIMO on fuze v1 ideal would be a way to reclock the PCLK for boost, and then do 31 idle, ~100 boosted, and then have a third level for mpegplayer and ape
16:46:45saratogawith plck of 31, 62 and 62 respectively
16:48:54 Join theseven_ [0] (578b902a@rockbox/developer/TheSeven)
16:49:01 Quit TheSeven (Disconnected by services)
16:49:04 Nick theseven_ is now known as TheSeven (578b902a@rockbox/developer/TheSeven)
16:49:14freddybHmm. How to tell if the cpu is waiting for memory?
16:49:47 Part Zagor
16:50:09 Join n1s [0] (
16:50:09 Quit n1s (Changing host)
16:50:09 Join n1s [0] (~n1s@rockbox/developer/n1s)
16:50:43saratogafreddyb: in general or in a specific block of code?
16:51:09saratogafor codecs, I would say disable boosting in test_codec, then run with boosting and see how much the MHz to decode changes
16:51:10***Saving seen data "./dancer.seen"
16:51:19saratogasince boosting increases teh CPU clock but not the memory clock on AMS
16:51:39saratogaon amsv2 it spends quite a lot of time waiting on memory when boosted
16:52:13freddybOK. Maybe boosting pclk is important then.
16:57:30alexbobPis the sansa clip the smallest rockbox target?
16:58:58 Quit simonrvn (Ping timeout: 240 seconds)
17:00:40 Join simonrvn [0] (simon@
17:02:20CIA-7New commit by nls (r28604): libtremor: correct constraint for coldfire asm, swap can only be used on data registers.
17:03:19LambdaCalculus37alexbobP: In terms of physical size? AFAIK yes.
17:04:45CIA-7r28604 build result: All green
17:06:16alexbobPLambdaCalculus37: woot
17:06:24alexbobPI have the smallest mp3 player that doesn't suck :D
17:06:40alexbobPactually I have the clip+, which I think is slightly smaller than the original clip (if I'm not mistaken)
17:07:28LambdaCalculus37alexbobP: I have a Clip v1 and a Clip v2.
17:07:39 Quit wodz (Quit: Leaving)
17:09:21saratogafreddyb: another fun idea that we never got anywhere with on the fuzev1 was seeing if it was possible to run at the memory controller at 2x pclk
17:09:39alexbobPLambdaCalculus37: have you compared to a clip+? do you know how the size compares?
17:09:39saratogathe datasheet suggests its possible
17:09:44saratogapretty similar
17:09:49saratogamaybe very slightly larger
17:09:50alexbobPLambdaCalculus37: I owned a clip, I think v1, but I dunno, a good long time ago...
17:09:57alexbobPLambdaCalculus37: I lost it a long time before rockbox had support
17:09:57 Quit Topy44 (Ping timeout: 240 seconds)
17:10:00saratogaif you have a clip and remove the clip part thats as small as you get
17:10:18alexbobPwell I meant smallest retail
17:10:32alexbobPotherwise we get into case-modding territory...
17:10:49alexbobPcheckitout guys, rockbox on a clip+ in ipod nano touch chassis >_>
17:11:58pixelmaalexbobP: the difference is I think that on the v1 and v2 Clips the clip is removable and on the Clip+ it's (usually) fixed
17:12:02alexbobPLambdaCalculus37: well it looks like sandisk publishes the exact same dimensions for the two products
17:12:06pixelmaso no case-modding needed
17:12:10alexbobPpixelma: yeah, it's fixed on mine :/
17:12:43alexbobPpixelma: but I meant, if you are allowed to take off parts, you can make it even *smaller* with case-modding
17:12:48alexbobPI'm not saying removing the clip is a case-mod :P
17:13:11alexbobPanyways it would be silly. the clip *already* has an awesome form factor!
17:19:21 Quit casainho (Ping timeout: 250 seconds)
17:23:30 Join EvanCarroll [0] (~ecarroll@
17:25:17n1swow, increasing the codebook decoding cache size in tremor speeds up decoding of high bitrate files a lot on coldfire
17:28:43n1snow, if it wasn't for all these dynamically sized allocs i could know if this would fit in the codec buffers of the targets with smaller codec buffers :/
17:32:03n1son the 500kbps file in test_files this increases the malloced memory size ~28k
17:32:31n1scould be #ifdef'ed for targets with small buffers of course
17:33:51 Join binaryhermit [0] (
17:34:48freddybsaratoga: does that need external memory?
17:36:26 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
17:39:30 Join bmbl [0] (~bmbl@unaffiliated/bmbl)
17:40:03 Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
17:42:30 Join lionfury [0] (
17:42:34lionfuryanyone here
17:42:46lionfuryi gotta question
17:42:58lionfuryobviously no one cares
17:43:32Mode"#rockbox +o Torne" by ChanServ (ChanServ@services.)
17:43:34lionfuryI JUST WANNA GET HIM BLOCKED!!!!!!!!!!
17:43:38 Quit lionfury (Client Quit)
17:49:54Mode"#rockbox -o Torne" by Torne (torne@rockbox/developer/Torne)
17:52:41LambdaCalculus37My, my, such behavior...
17:52:58 Quit petur (Quit: Page closed)
17:54:51saratogafreddyb: yeah the fuzev1 has an external 8 MB DRAM rated for 120MHz, and the included memory controller is rated at 90 MHz according to the datasheet
17:55:12saratogaso in theory is could be run higher then pclk, maybe improving performance
17:55:34*LambdaCalculus37 goes back to looking through the romfs file of the Gigabeat MEG50JS
17:55:54 Join toffe82 [0] (
17:57:21freddybI see now, although I'm exploring how slow I can make things without breaking the Rockbox Experience(TM).
17:57:53saratogaits probably worth trying a battery bench with a fast codec like flac with stock rockbox, and then again with lowered clocks
17:57:59saratogato see what kind of savings is available
17:58:27EvanCarrollIs the 200% pitch limiter there for a reason?
17:59:48EvanCarrollI assume the greater the pitch the more it takes to decode the track, but is that the reason why it is there? or is it just because it is assumed useful at 200%
18:00:41n1sit's probably just an arbitrary limit
18:01:34EvanCarrollI'm loving the feature, it makes the tape sound like a chipmonk but I can swing through audio books I've heard much faster than 200%, and ones that I haven't heard comfortably at 200%.
18:01:54saratogahow does the pitch thing work anyway?
18:02:00saratogafft based or something else?
18:02:22n1si'd guess linear interpolation
18:02:32EvanCarrollno idea, I just like it because it makes the track play "faster"
18:02:32n1ssimiöar to the resampler
18:02:58EvanCarrollIt's surprisingly awesome, if you can get over not ridiculing the chipmonk sound.
18:03:08n1sEvanCarroll: did you know there's a timestretch feature that let's you change the tempo while keeping pitch the same
18:03:58EvanCarrollno I didn't know that ;)
18:04:01EvanCarrollI'll have to seeit.
18:04:52EvanCarrollAnyway, I'm addicted for the time being to The Teaching Company, this feature which I only started using a week ago is totally addicting.
18:07:31EvanCarroll Due to the limitations of the algorithm, speed is limited to be between 35% and 250% of the current pitch value. Pitch must maintain the same ratio as well as remain between 50% and 200%.
18:07:37EvanCarrollI guess that's why it is stuck at 200%
18:08:31EvanCarrollenabling time stretch and changing the pitch does not make a difference on the e200v2 target
18:08:57freddybEvanCarroll: I think I would squish my audio book before I loaded onto the player if I were you. More battery life, smaller filesizes, less screwing around with the pitch screen...
18:10:22EvanCarrolltimestretch adds this humming noise in the background, if it is more pitch correct, it only sounds slightly better.
18:10:43EvanCarrollfreddyb: If I do end up doing that, I would have never considered it without the feature built into rockbox.
18:11:20freddybPoint taken.
18:13:28freddybThere's "better" time stretch algorithm posted on FS: that's been sitting for a while.
18:33:14n1sah, almost ~10% speedup on the 500kbps file on PP too
18:34:58 Join domonoky1 [0] (
18:36:45 Quit domonoky (Ping timeout: 240 seconds)
18:40:54 Quit binaryhermit (Read error: Connection reset by peer)
18:41:19 Join binaryhermit [0] (
18:48:11 Quit tchan (Quit: WeeChat 0.3.3-dev)
18:49:46 Join simonrvn_ [0] (simon@
18:49:52 Quit simonrvn_ (Read error: Connection reset by peer)
18:51:11***Saving seen data "./dancer.seen"
18:52:01 Quit benedikt93 (Quit: Bye ;))
18:52:58 Quit simonrvn (Ping timeout: 240 seconds)
18:54:43 Join simonrvn [0] (
19:02:55 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
19:04:21 Join user890104 [0] (
19:05:43n1sinterestingly made almost no diff on the fuze...
19:06:51saratogawhich fuze?
19:07:14 Join bertrik [0] (
19:07:15 Quit bertrik (Changing host)
19:07:15 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
19:07:17saratogado you have a diff?
19:07:37n1si can post one, 1sec
19:08:59n1sthe CODEC_SIZE condition is kind of arbitrary
19:09:19 Quit Zambezi_ (Changing host)
19:09:19 Join Zambezi_ [0] (Zulu@unaffiliated/zambezi)
19:09:31 Nick Zambezi_ is now known as Zambezi (Zulu@unaffiliated/zambezi)
19:10:17saratogais that buffer in IRAM?
19:10:20saratogaprobably not...
19:11:12saratogafuzev1 performance is extremely weird
19:11:14 Join TheLemonMan [0] (~lemonboy@
19:11:27saratogai can't figure out why it behaves the way it does
19:11:36 Quit swilde (Remote host closed the connection)
19:11:38n1sincreasing this saves a lot of searches in the codelist with is not in iram either so it saves a bunch of loads from dram, especially on high bitrate files
19:11:43n1ssaratoga: indeed
19:18:29AlexPbluebrother, domonoky1: In rbutil when the dialog pops up telling you you need to download an OF (for e.g. fuze), are you supposed to be able to click on the bits in underlined blue that look likes links? (Sansa Forums, wiki, manual). If so, I can't :)
19:27:28 Join Buschel [0] (
19:30:54saratogawe need to play around with the test_mem plugin on amsv1 until someone understands whats going on there
19:31:03saratogait seems to be unlike all other arm9 targets
19:31:12saratogai wonder if it has a broken cache like the old PPs or something similar
19:31:30BuschelTheSeven: I have got severe problems with the voltage scaling on my local build
19:31:32domonoky1AlexP: yes, it should be clickable (its a URL in richtext), but it probably depends on the OS if it really is clickable.
19:31:54AlexPdomonoky1: Isn't clickable here in linux (kde on Arch)
19:33:40 Quit bmbl (Ping timeout: 240 seconds)
19:34:01BuschelTheSeven: I have got this patch applied to the build (gui boost):
19:34:09domonoky1AlexP: Thanks for the info, but we probably can not fix that, its controlled by Qt.
19:34:31AlexPOk, it isn't a big deal - it just confused me
19:34:53AlexPI'm a simple soul :)
19:35:56BuschelTheSeven: As soon as I start to scroll (which will boost the CPU) the nano crashes. If I comment the pmu_write's and the sleep it works fine... A mutex does not help either... Any idea?
19:35:59 Join Zagor [0] (~bjst@rockbox/developer/Zagor)
19:36:21TheSevenand this doesn't happen without that gui boosting patch?
19:37:30Buschelno. it must be either some timing issue (that's why I though a mutex could help) or any other issue I did not recognize so far...
19:40:39 Quit DerPapst (Quit: Leaving.)
19:41:51*TheSeven reboots to windows again
19:45:24 Quit LambdaCalculus37 (Quit: CGI:IRC 0.5.9 (2006/06/06))
19:46:27 Quit TheSeven (Ping timeout: 265 seconds)
19:47:19 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
19:48:01 Join Topy44 [0] (
19:58:16 Quit jgarvey (Ping timeout: 264 seconds)
19:58:26 Join stacker55 [0] (
19:58:45 Join jgarvey [0] (
20:01:22amiconnsaratoga: I may fill in coldfire info later. For now I've fixed the PP info :)
20:02:21 Quit Xerion (Quit: )
20:02:24 Quit stoffel (Remote host closed the connection)
20:03:06BuschelTheSeven: when inserting the sleep() in set_cpu_frequency() the lockup happens
20:03:37TheSevenhm, I can't see how yielding in a UI thread can hurt
20:03:57TheSevenhave you tried keeping the i2c writes and throwing the sleep out?
20:04:07Buschelyes, just doing it -> works
20:04:28TheSevenso you can actually run it at 192MHz with 0.900V Vcore for short periods of time?
20:05:02Buschelit seems so, but I am not sure what will happen when this is done over a longer period of time
20:06:29TheSeventry it :)
20:06:39TheSevenmaybe yours is stable under these conditions
20:08:09*Buschel is normally lucky with his targets
20:08:18Buschelmy iPod Video runs stable at 100MHz :)
20:08:26Buschelfor years now
20:09:36Buschelwhen inserting sleep(HZ/100) the tick-task crashes...
20:09:58Buschel*to the tick-task
20:10:52 Join JesusFreak316 [0] (
20:12:07 Join InsDel [0] (
20:12:31TheSevenoh, this is called from the tick task?
20:12:37TheSevenforget about blocking on anything from there
20:12:57Buschelyes, after some debugging... not the best idea, isnt' it ;)
20:13:07TheSeventhat's interesting
20:13:26TheSeveni just tried to do I2C from the tick task yesterday, without sleeps, and it failed
20:13:37TheLemonMananyone worked on freescale SoC ?
20:13:42Buschelit is called from "button_tick" task
20:13:42TheSevenit must fail, because an IRQ handler can't possibly wait on an IRQ
20:14:10Buschelso, we might use active wait instead?
20:14:16n1sTheLemonMan: the gigabeat s has a freescale SoC
20:14:31 Quit liar (Ping timeout: 255 seconds)
20:14:36TheSevenfirst of all we should figure out how much delay we actually need
20:14:45TheLemonMancool, i have a 3700 device and i dont know how to power on the heaphones
20:15:04TheSeveni just chose 10ms because our sleep implementation doesn't have more resolution anyway, and because it should be sufficient
20:15:23TheLemonManthe dsp is controlled via /dev/dsp but i cant hear any sound, so i guessed headphones were teh problem
20:15:26 Join Xerion [0] (
20:16:05 Join liar [0] (
20:16:25n1siirc, the gigabeat s has a gpio controlled headphone mute
20:17:52TheLemonMani should avoid using direct hw access, theres a nice /dev/gpio interface tho :P
20:17:59TheLemonManlink to the gigabeat code ?
20:19:16 Join DerPapst [0] (
20:19:47n1sit's in svn
20:20:09n1sare you planning to run rockbox as an app on that player?
20:21:03 Quit TheSeven (Ping timeout: 260 seconds)
20:21:45TheLemonManas long as i get working audio
20:22:22n1scan't you check what the app running on it usually does?
20:24:09TheLemonManive already done that (samsung left a diagnostic tool with debug symbols in the player)
20:24:24TheLemonManbut theres no reference to gpio writes nor memory writes
20:25:34saratogathis audio corruption in my libmad synth_filter patch is driving me nuts
20:25:54saratogai double checked all the filter coefficients as they're loaded and they seem identical to the svn version
20:26:03saratogabut theres still some glitching so something must have changed
20:26:33n1sTheLemonMan: well, it's hard to know what' s going on without having it to poke on but that diagnostic tool can output sound?
20:27:11TheLemonManyeah, outputs a sine wave
20:27:54n1sand you diassembled this too or ran it through gdb or something?
20:28:42TheLemonManive disassembled it
20:31:11 Join Kupop [0] (
20:32:29BuschelTheSeven: if I read the spec of a pcf5606 correct a stepwise change from 0.9 to 1.0 V takes 4*8 us, instantaneous will be faster. so, let's assume the switch will take<0.1 ms
20:33:48n1sudelay should be fine then
20:35:50 Quit JesusFreak316 (Ping timeout: 276 seconds)
20:37:01 Join Jerom [0] (~jerome@
20:41:17 Nick YPSY is now known as Ypsy (
20:44:00TheLemonMan[STMP SOUNDTEST] dev/dsp mmap for sram enable of codec success..
20:44:22TheLemonManthis string looks intresting, might be a dsp codec issue ?
20:46:10 Quit Jerom (Quit: Leaving.)
20:48:38 Join lixxus99 [0] (
20:49:22 Part lixxus99
20:51:15***Saving seen data "./dancer.seen"
20:54:08 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
21:02:51 Join Horscht [0] (~Horscht@xbmc/user/horscht)
21:04:21 Quit Horschti (Ping timeout: 240 seconds)
21:10:50 Quit DerPapst (Quit: Leaving.)
21:11:34 Join DerPapst [0] (
21:13:50 Quit n1s (Quit: Lämnar)
21:15:25 Join Zarggg [0] (
21:18:25soapJdGordon, have you been paying attention at all to the "Playlists" thread in the forums. It is not exactly in line with some of your recent work, but if I am understanding the description of the Apple Firmware correctly it is a very interesting and perhaps useful interface design. Just curious as to your thoughts on it.
21:20:41 Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
21:22:38keenerdSilly thought, has there ever been a patch for set operations on playlists? E.g., add to playlist "beatles" subtract from playlist "yellow submarine".
21:23:17Lloreankeenerd: You can add files to playlists easily using the playlist catalog.
21:23:37keenerdI've read the manual :-)
21:23:42LloreanTo remove I think they playlist needs to actually be loaded (since a file can appear multiple times, for example, so you'd be removing an explicit line)
21:23:54Lloreankeenerd: If you've read the manual, why did you ask about adding to playlists?
21:24:09keenerdI did not. I asked about set operations.
21:24:16alexbobPkeenerd: I've always wanted a media player to do things like that! I haven't even found a desktop media player that does...
21:24:38Lloreankeenerd: Your description doesn't differentiate your "add to playlist "beatles"" from the existing add to playlist functionality though...
21:24:41 Join krabador [0] (
21:24:48alexbobPunion, intersection, and not...
21:24:55keenerdLlorean: /me facepalms.
21:24:56alexbobPthey'd be totally awesome
21:25:03keenerdWe've got union by default.
21:25:09keenerdThat is just "Play next"
21:25:22alexbobPnot quite... you can't slap two playlists together and call it a union, a union would remove duplicates
21:25:53Lloreankeenerd: Don't give me that 'facepalm' crap. I gave an honest response. Maybe your assumption that people would know your intent was fine, but I clearly didn't. Maybe try being poilte and explain things rather than being like that?
21:26:06keenerdalexbobP: Fair enough. I do everything directory-wise, so the playlists don't overlap.
21:26:13LloreanalexbobP: Playlists can already have more than one copy of the same song, so why should duplicates be removed when merging?
21:26:31alexbobPLlorean: because that's what union is, and we're talking about set operations?
21:26:36keenerdLlorean: Not everyone wants the song twice in the dynamic playlist.
21:26:37alexbobPLlorean: I realize playlists are *not* sets, so it's not entirely consistent
21:26:51LloreanalexbobP: Yes, it's not. So why talk about treating it as such?
21:26:52alexbobPLlorean: in my use case, the original playlists would not have any duplicates, but they might have tracks in common, and I don't want dupes in the output
21:27:12LloreanIf list X has 3 copies of song A and Y has 2 copies, how many does the resultant playlist get?
21:27:20alexbobPthat's a good question
21:27:31alexbobPI'd go with 3, I guess
21:27:46alexbobPbut someone who already has duplicates in a playlist would probably just not be using the union feature
21:27:49alexbobPeither 3 or 1
21:28:12LloreanWhy waste binary size on a feature that's not going to make sense to the majority of users anyway?
21:28:27alexbobPwell I wouldn't really add a union, tbh
21:28:37alexbobPI'd just add a "remove all duplicates" if there isn't already one :P
21:28:52keenerdLlorean: Hypothetical situations require a set up. The first half of the hypothetical situation will be obvious/simple/common. The second half will be impossible. Reread my entire statement, no just the first half.
21:28:59alexbobPbut intersection would be nice... which tracks are in *both* playlists... and subtract all songs in playlist foo would be nice too
21:29:47Lloreankeenerd: I responded to both halfs...
21:30:03keenerdLlorean: Nope. Not one word about set subtraction.
21:30:06Lloreankeenerd: But you don't ask "wouldn't it be nice if we could add and remove" when you know we can already add. You'd ask "wouldn't it be nice if we could also remove"
21:30:09*gevaerts thinks that keenerd's original description was clear
21:30:25Lloreankeenerd: "(2:23:31 PM) Llorean: To remove I think they playlist needs to actually be loaded (since a file can appear multiple times, for example, so you'd be removing an explicit line)" <−− Did you not see this line?
21:30:42keenerdLlorean: Yup. It was nonsense.
21:30:58LloreanIt's an accurate description of what is necessary currently in Rockbox to remove a song from a playlist...
21:31:04keenerdLlorean: If you add a playlist first, then it is already "actually loaded".
21:31:13Llorean"add a playlist"?
21:31:24alexbobPI think he means to your "current" playlist
21:31:24LloreanThat doesn't even mean anything. Do you mean "add a song to a playlist?" Because that can be done without loading a playlist.
21:31:35keenerd*dynamic playlist
21:32:01LloreanYou didn't mention that this function was explicitly focused on dynamic playlists.
21:32:08LloreanDo you mean all playlists, or just the dynamic playlist?
21:32:44keenerdWhatever is easier to implement or already exists :-)
21:32:57LloreanYou can add songs to static playlists without loading them into playback, but removal requires (currently) that they be the playing playlist.
21:33:23LloreanIt also requires that you explicitly save the playlist after removing songs from it, which is a hassle last time I checked.
21:33:35*Llorean wouldn't mind an "update playlist on stop" option.
21:33:36pixelmaor the text editor (though not very handy)
21:34:30LloreanA proper playlist editor plugin might be nice.
21:35:33keenerdI'm guessing the current playlist would be more useful. Rewording the initial story: Files -> Beatles -> Playlist -> Insert. Files -> Beatles -> Yellow Submarine -> Playlist -> Subtract.
21:37:01alexbobPLlorean: so you *can* remove tracks from the *current* playlist? Would it then be doable to make a "remove all tracks in foo playlist from current playlist" feature?
21:37:02keenerd(Or from the DB interface.)
21:37:03pixelmathe one thing I once missed after doing a mistake in playlist manipulation a few times in a short while was undoing the last step (inserted a whole folder next instead of at the end or so), that hasn't happened to me for a long time though and so I forgot about it
21:37:08Lloreankeenerd: Does that remove the first instance of the song, or all instance of the song? Queued instance, actually inserted instances, both?
21:37:30alexbobPwhat's the difference between queued and inserted?
21:37:47keenerdOne is popped off the playlist after it is played.
21:37:50LloreanalexbobP: Queued remove themselves after first playback (they aren't actually saved if you save the playlist, basically "play without really adding")
21:38:34alexbobPLlorean: ah, that's good to know.
21:38:47 Join JesusFreak316 [0] (
21:38:59keenerdLlorean: We have seven different types of "add to playlist" so why not a different "subtract" for each one you've named?
21:39:01Lloreanpixelma: An undo like that would be great. I've accidentally insert-shuffled something I wanted in order and it was quite frustrating.
21:39:46Lloreankeenerd: There's already a sentiment that there's too much stuff in the playlist menu (I'm pretty sure it it could be reduced to 2/3 the current size actually)
21:40:06BuschelTheSeven: can USEC_TIMER be used for the udelay() implementation on nano2g? s, simply copying udelay from PP?
21:40:32TheSevenit only has five usecs resolution, but that shouldn't matter
21:40:38keenerdSo I take it there are no patches for any sort of "subtract" floating around?
21:40:43Lloreankeenerd: If you're just removing one song from the playlist, why is finding it in the filetree and removing it preferable to finding it in the playlist and removing it?
21:41:10keenerdLlorean: It is not one song. It is entire directories or DB searches.
21:41:50keenerd(I figured Beatles were well enough known that people knew their album titles. My bad.)
21:42:01BuschelTheSeven: 5 usec precision, but still usecs? in this case the implementation should be the same
21:42:42TheSeventhere is also a 64bit FIVE_USEC_TIMER thing if you should need that one day
21:42:50 Join wodz [0] (
21:43:04CIA-7New commit by wodz (r28605): mkmpioboot - Add support for HD300, increase verbosity.
21:43:04TheSevenUSEC_TIMER is based on that, multiplied by five and truncated to 32 bits
21:43:45Lloreankeenerd: I guess I don't really see it as a significant problem. I don't generally add music that I don't want to listen to in the first place, and I set up playlists in advance that I re-use. It might be a useful feature, but it's probably the sort someone would have to put up a patch for and demonstrate that it won't get in the way / increase size much
21:44:35pixelmaLlorean: probably not easy though as you need to store the state of the playlist before the last "action" somewhere
21:45:23CIA-7r28605 build result: All green
21:45:27keenerdLlorean: Some people are bad at planning in advance (making playlists) or are completionists who insist on carrying around a band's bad music too.
21:45:48Lloreanpixelma: Or store the last action. If you know that you inserted 5 songs at indices 5, 7, 9, 11, 12, then you just remove 5, 7, 9, 11, and 12 from the playlist and you've restored its state.
21:46:16TheSevenBuschel: seems to be ok, but 20µs probably aren't sufficient for the voltage to stabilize
21:46:17pixelmayeah, just realised this myself
21:46:40Lloreankeenerd: Rockbox has a history of not trying to cater to everyone, but rather preferring to aim for a good cross section of "functionality" without going into the too minor usages.
21:46:42keenerdpixelma: An easy hack solution would be to half the max playlist size. Though you on;y get one level of undo.
21:47:16amiconnLlorean: Not that simple for delete actions though
21:47:25Lloreanpixelma: an action like "shuffle playlist" might be more of a hassle though
21:47:45pixelmakeenerd: I don't understand what max playlist size has to do with undo
21:47:48Lloreanamiconn: For delete you would need to store a list of deleted files too, yes
21:47:54BuschelTheSeven: then I'll choose 50us, still far less than svn's 10000us ;)
21:47:57keenerdpixelma: Store two copies of it, one for rollback.
21:48:01Lloreanpixelma: If you halved the max size, you could store two copies of the playlist.
21:48:10amiconnShuffling stores the random seed, so that is undoable
21:48:11TheSeveni don't think that will be sufficient either
21:48:31TheSevenwhat's the lowest stable (besides of sampling rate drops) voltage for 48MHz?
21:48:40TheSevencan you try using that when not boosted?
21:48:57TheSevenif it survives the transition from that to boosted, it should hopefully be fine on all devices
21:51:01keenerdamiconn: Well, if we used a reversable PRNG....
21:51:28keenerd(All are, but it is a PITA.)
21:52:19*pixelma wonders if the "undoable" was an "undo possible" or an "impossible"
21:52:54amiconnmersenne twister
21:53:19keenerdamiconn: Nuts. If it was a LFSR, it would be trivial.
21:55:05amiconnOne could go back to the original seed, re-generate the sequence, store and then reverse it. But given the memory requirements for that it's probably easier to keep two copies of the playlist
21:55:38TheSevenif it's a linked list, one could just attach it the other way round
21:56:01amiconnIt's not
21:56:16TheSevena plain array with constant element size?
21:56:33TheSevenwell, reversing that is trivial :)
21:57:06amiconnThen you get the shuffled list in reverse order, not the original unshuffled list
21:57:34*TheSeven thinks we'll need to agree on the definition of "reversing" first
21:57:41amiconnI'm not sure - there might be a reversible shuffle algorithm that yields the original list if the same sequence of pseudo-random swaps is applied a second time
21:57:59amiconnSimilar to how burrows-wheeler is reversible
21:58:54saratogaBuschel: what are you working on with the timers?
21:59:25Buschelsaratoga: I need a udelay for the nano 2g
22:01:11BuschelTheSeven: 0.8 V worked for 48 MHz, 0.7 V did not work
22:01:36Buschelwill try the udelay now
22:02:13saratogaah for changing voltage or clock settings? thought maybe you were benchmarking
22:02:17TheSeventry 0.800=>0.975 transitions, switching to 192MHz immediately without any delay
22:02:24TheSeveni'd expect that to fail
22:03:22TheLemonManif i run rb as an app do i need to provide the interface to the SI4703 tuner ?
22:04:01Buschelhmm, 0.8 -> 0.975 with 20 us fails after several switches from 48 -> 192 MHz
22:04:47Buschelw/o udelay it does not start = first boost fails
22:05:22TheSevenis it 100% stable with 100µs?
22:05:24Buscheltrying 50 us now
22:05:41TheSevenif it's apparently stable at 50, i'd say commit 100
22:05:45 Join xxcv [0] (
22:05:46TheSeventhis is still damn fast
22:07:39 Quit benedikt93 (Quit: Bye ;))
22:08:48saratogaTheLemonMan: no, if you're running as an application then you cannot access any hardware directly or provide your own drivers
22:09:05BuschelI am playing with test_boost plugin for a while now. stable for 0.8 V (48 MHz) -> 0.975 V (192 MHz) with 50 us delay
22:09:29saratogawhats the max clock on the nano2g?
22:09:33TheLemonMansaratoga, wait, i dont get it
22:09:37Buschel200 MHz
22:10:07saratogaif you guys want to add support for 3 level boosting, this might be a good time
22:14:01CIA-7New commit by Buschel (r28606): Implement udelay() for S5L870x. Exchange sleep() with udelay() during CPU voltage scaling. Voltage scaling was measured stable with 50us delay, to ...
22:15:46 Part Zagor
22:15:55CIA-7r28606 build result: All green
22:16:59 Join lixxus99 [0] (
22:17:04 Part lixxus99
22:18:08BuschelTheSeven: how low can you go @48 MHz on your target?
22:24:59TheLemonManFUCKIN HELL YEAH
22:25:08TheLemonMangot first sound output
22:25:31bertrikTheLemonMan, nice
22:25:53bertriknot completely sure what you're doing though :)
22:26:03TheSevenBuschel: it was <=0.875, don't remember the exact value
22:26:25TheLemonManbertrik, porting rockbox to samsung yp-q2 :D
22:28:13bertrikit runs linux as the original firmware?
22:28:55saratogathey were nice enough to open source the entire firmware, and then give instructions for building it from scratch
22:29:18saratogaso i think you can pretty much do anything you want on the deivce, although i haven't tried so its possible they've got some limitations
22:29:23TheLemonMannope :) i had to hack the upgrade process and hijack the boot process
22:29:37saratogayou tried building the original firmware?
22:29:57TheLemonManthats for YP-R0, not YP-Q2
22:30:35 Quit kevku (Quit: KVIrc 4.0.2 Insomnia
22:32:03saratogaoh? i thought you linked the firmware for your player before?
22:32:22saratogaoh you're right
22:32:28saratogai just realized it was the R0
22:33:01saratogayes sorry I'm mixing you up with slate's project on a different linux mp3 player
22:39:55keenerdThat Samsung player looks nice. Though at $99, I'd get an Archos 28 (yay wifi) instead :-)
22:43:18TheLemonManhows that in € ?
22:43:29keenerdHrm, Q2 has twice the battery life of the 28. Tougher choice.
22:44:12dfktwait for the samsung r0 port.. this one has tactile buttons and an sd slot ;)
22:44:56wodzAre we about to see explosion of RaaA ports on linux based players?
22:46:55 Quit Buschel (Ping timeout: 272 seconds)
22:48:57saratogadepends how much kugel gets done :)
22:51:19***Saving seen data "./dancer.seen"
22:53:45 Nick Ypsy is now known as YPSY (
23:00:53 Quit xxcv ()
23:00:53 Quit bertrik (Quit: :tiuQ)
23:08:04 Quit TheLemonMan (Quit: Help me, i got shot! *DIES*)
23:18:51 Quit jgarvey (Quit: Leaving)
23:19:56JdGordonsoap: think I gave it a read yesterday, didnt pay much attnetion...
23:20:00JdGordonI'm not a UI guy :p
23:20:06JdGordonI'll look ltaer
23:21:39 Join kazaik [0] (
23:23:59 Quit leavittx (Ping timeout: 255 seconds)
23:24:40 Join stripwax [0] (
23:25:14 Quit Kupop (Ping timeout: 265 seconds)
23:25:57 Quit simabeis (Quit: Lost terminal)
23:27:38 Join T44 [0] (
23:30:46 Quit Topy44 (Ping timeout: 240 seconds)
23:31:24 Join krazykit` [0] (
23:31:41 Quit krazykit (Ping timeout: 240 seconds)
23:42:01 Quit wodz (Quit: Leaving)
23:42:23 Quit freddyb (Quit: ChatZilla 0.9.86 [Firefox 3.6.8/20100723085406])
23:49:50 Part domonoky1

Previous day | Next day