#rockbox 2010-11-10

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 2010-11-10

00:00:35	saratoga	Buschel: ping
00:00:50	saratoga	46.59 for SVN, vs. 34.62 for the patched version :)
00:00:52		Quit Zarggg_ (Quit: Zarggg_)
00:01:19	Buschel	wow
00:01:20	saratoga	and i think theres still some stalls in the asm that can be improved
00:01:50	Buschel	+26%
00:02:26		Join Ijon_Tichy [0] (~username@188.46.178.229)
00:02:57	saratoga	i guess i should also check for any artifacts too since the precision is reduced a bit
00:04:08	Buschel	yes, but if the precision is the same than with the c-implementation, I have measured it and placed this information in yur fs-entry. the results were acceptable...
00:04:28	saratoga	yes it should be the same
00:04:36	saratoga	how much rounding error was there?
00:04:51	Buschel	you should of course add some rounding to the coefs instead of just truncating the least 16 bit −− like done in my interim patch
00:05:10		Quit utanapischti (Ping timeout: 240 seconds)
00:05:13	Buschel	From FS #11709: "Difference to svn decoding is max +/-4 of +/-32768 (~ -78 dB), and average ~ -110 dB. This is quite few."
00:05:53	saratoga	so better then 16 bits precision
00:05:55	saratoga	not bad
00:05:59	Buschel	this was including rounding instead of truncation: win = (s + (1<<15))>>16
00:06:20	saratoga	IIRC the Sony reference ATRAC3 decoder is said to do less then that
00:07:03	saratoga	do you remember how scheduling works on ARM11?
00:07:17	Buschel	no, sorry.
00:07:27	saratoga	i think it has a 64 bit load/store so the ldmia r1!, {r6, r7} lines should take a single cycle if in cache
00:07:39	saratoga	but i bet theres a pipeline stall because I try to use them on the next cycle
00:09:15		Join sasquatch [0] (~username@188.46.178.229)
00:09:31	saratoga	i can use r10 right?
00:09:37	Buschel	yes
00:09:49	Buschel	gotta leave now, it's getting late here :)
00:10:05	saratoga	sure
00:10:07	saratoga	night
00:10:21	Buschel	see you!
00:10:24		Quit Buschel (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630])
00:11:04		Quit captainkewlllll (Quit: Page closed)
00:11:34		Join teba [0] (~teba@p5B20C9C8.dip.t-dialin.net)
00:12:08		Quit Ijon_Tichy (Ping timeout: 255 seconds)
00:13:37		Part teba
00:13:43		Join Zero [0] (~chatzilla@96.18.130.90)
00:13:54		Quit wodz (Quit: Leaving)
00:14:18		Nick Zero is now known as Guest91383 (~chatzilla@96.18.130.90)
00:14:41	saratoga	huh divide by zero when i try to use r10
00:14:43	saratoga	i wonder whats wrong
00:15:00	Guest91383	Hello
00:16:11	bertrik	hm, the hdd1630 seems to spend 10% of its time doing an udelay to generate a key click
00:17:11	bertrik	oh wait, only if it's clicking of course
00:17:23	saratoga	oh huh i guess i should stack r10 if i overwrite it :(
00:17:40	Guest91383	I'm having a volume problem with the fuzev2. If the volume goes up to high, it starts skipping/stuttering. Lower the volume and it goes away. Is this a common problem? And is there any way I can fix it?
00:18:29	bertrik	No, it's not a common problem as far as I know
00:18:34	saratoga	you mean it clips if you go above 0 dB?
00:18:53	Guest91383	the volume threshold seems to be about -14
00:19:02	bertrik	It could be some kind of headphone short circuit detection
00:19:07	saratoga	it should start clipping at 0 dB
00:19:52	Guest91383	hmm... well as soon as it goes from -14 to -13, it starts. So I should try different headphones then?
00:21:17		Part toffe82
00:21:56	Guest91383	weird, different headphones and it now starts from -11 to -10.
00:22:08	saratoga	which revision are you running?
00:23:48	Guest91383	I just used the installer today. And the stuttering kicked back in at -11
00:24:12	Guest91383	as for revision.... a bit over my head....
00:26:16	pixelma	did you chose to install the release (3.7) or a "current build"?
00:26:37	Guest91383	3.7, all done automatically
00:27:16	saratoga	then you're running 3.7
00:28:04	pixelma	do you have the equaliser enabled?
00:28:08	Guest91383	ah, my noobiness is shining bright.
00:28:15	saratoga	how do I check for ARMv5E or higher?
00:28:32	Guest91383	I tried with the equalizer on and off.
00:29:05	Guest91383	I mainly was concerned that others might have been having the same problem. If not, then there's a really good chance it's the player itself. It's a refurb from newegg.
00:29:06	bertrik	The equalizer could cause distortion but never skipping AFAIK
00:29:26	gevaerts	Does it work properly in the original firmware?
00:29:47	Guest91383	I installed rockbox as soon as I made sure it turned on. I'll check really quick.
00:30:36	saratoga	#elif defined(CPU_ARM) && (ARM_ARCH >= 5)
00:30:40	saratoga	nice
00:30:50		Quit pamaury (Remote host closed the connection)
00:31:17	Guest91383	crap, I wasted your time. Yeah, it's the player... sorry about that
00:32:35	Guest91383	Thanks for your time, off to newegg support
00:32:41	gevaerts	Good luck!
00:33:00		Quit Guest91383 (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630])
00:34:24		Join MethoS- [0] (~clemens@134.102.106.250)
00:37:02		Quit bertrik (Quit: :tiuQ)
00:39:01		Join InsDel [0] (~haqr.net@c-98-231-87-43.hsd1.fl.comcast.net)
00:40:20	***	Saving seen data "./dancer.seen"
00:48:34		Quit mystica555_ (Read error: Connection reset by peer)
00:53:02		Join mystica555 [0] (~mike@m482636d0.tmodns.net)
00:53:03		Quit mystica555 (Excess Flood)
00:53:35		Join mystica555_ [0] (~mike@m382636d0.tmodns.net)
00:54:18		Quit ender` (Quit: Come to think of it, there are already a million monkeys on a million typewriters, and IRC is nothing like Shakespeare.)
00:56:41		Quit Kitar\|st (Ping timeout: 276 seconds)
01:00
01:02:43		Quit kazaik (Quit: Leaving)
01:02:47		Join Kitar\|st [0] (~Kitarist@BSN-182-80-211.dial-up.dsl.siol.net)
01:07:08		Join CaptainKewl [0] (~jason@207-38-215-126.c3-0.nyr-ubr1.nyr.ny.cable.rcn.com)
01:10:23	saratoga	updated this: FS #11709 - ARMv5E iQMF for libatrac
01:10:36	saratoga	feel free to let me know if anything breaks on coldfire
01:10:48	saratoga	(don't see how it would though)
01:11:56		Join kazaik [0] (~kazaik@pool-71-179-65-27.bltmmd.east.verizon.net)
01:24:14	saratoga	adding extra cycles between loads and register accesses has only a very small effect on arm9e
01:24:22	saratoga	i suppose the pipeline is too small for stalls to matter much
01:34:02	saratoga	is there some way to ensure that a buffer is double word aligned?
01:36:34		Quit ved (Quit: sepukku)
01:38:52		Join ved2 [0] (ved@ddsbox.co.cc)
01:39:24		Nick ved2 is now known as ved (ved@ddsbox.co.cc)
01:39:50		Join Dhraakellian [0] (~ntryon@cpe-67-240-248-41.rochester.res.rr.com)
01:41:47		Join checker [0] (~8b373152@giant.haxx.se)
01:42:23	checker	can i format gigabeat F then simply put ".rockbox" folder on root level?
01:44:50	krazykit	no, you need the bootloader and at least the dummy firmware files
01:46:01	saratoga	anyone with a beast available to test something?
01:49:57	saratoga	ok i have to run, but I'd love to see Beast benchmarks for the stock decoder, the v7 patch and the v8 patch here: http://www.rockbox.org/tracker/task/11709
01:50:07	saratoga	theres a test file linked in case you don't have atrac3 files handy :)
01:51:04	saratoga	now to figure out how to get such a huge speedup for mp3 . . .
01:51:19		Quit checker (Quit: CGI:IRC)
02:00
02:14:09		Join fdinel [0] (~Miranda@modemcable235.127-131-66.mc.videotron.ca)
02:15:03		Join anewuser [0] (anewuser@unaffiliated/anewuser)
02:15:41		Quit DerPapst (Quit: Leaving.)
02:20:52		Quit saratoga (Quit: Page closed)
02:40:23	***	Saving seen data "./dancer.seen"
03:00
03:08:35		Quit kazaik (Quit: Leaving)
03:16:15		Quit thegeek (Read error: Connection reset by peer)
03:21:01		Quit BHSPitMonkey (Ping timeout: 265 seconds)
03:34:25		Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
03:40:09		Quit BHSPitMonkey (Read error: Connection reset by peer)
03:41:07		Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
03:58:05		Join bluebrother [0] (~dom@f053154106.adsl.alicedsl.de)
03:58:05		Quit bluebrother (Changing host)
03:58:05		Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
03:59:55		Quit BHSPitMonkey (Ping timeout: 265 seconds)
04:00
04:01:03		Quit bluebroth3r (Ping timeout: 272 seconds)
04:09:34		Quit InsDel (Read error: Connection reset by peer)
04:12:12		Quit pixelma (Disconnected by services)
04:12:14		Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
04:12:16		Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
04:13:31		Quit amiconn (Disconnected by services)
04:13:32		Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
04:13:52		Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
04:27:29		Join Barahir_ [0] (~jonathan@frnk-590fe458.pool.mediaWays.net)
04:27:35		Quit MethoS- (Remote host closed the connection)
04:30:30		Quit Barahir (Ping timeout: 245 seconds)
04:32:38		Quit mystica555_ (Read error: Connection reset by peer)
04:40:26	***	Saving seen data "./dancer.seen"
04:44:07		Quit factor (Quit: Leaving)
04:44:26		Join factor [0] (~factor@r74-195-220-23.msk1cmtc02.mskgok.ok.dh.suddenlink.net)
04:47:57		Quit antil33t (Read error: Connection reset by peer)
04:53:23		Join antil33t [0] (~Mudkips@124-197-51-80.callplus.net.nz)
04:56:25		Join dys` [0] (~andreas@krlh-5f720dfd.pool.mediaWays.net)
04:58:07		Quit sasquatch (Ping timeout: 255 seconds)
04:58:35		Quit dys (Ping timeout: 276 seconds)
04:59:15		Quit TheSeven (Ping timeout: 260 seconds)
05:00
05:04:46		Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
05:05:36		Quit fdinel (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
05:30:17		Quit anewuser (Ping timeout: 276 seconds)
05:33:25		Quit ps-auxw (Ping timeout: 245 seconds)
05:36:57		Quit Horscht (Quit: Verlassend)
05:44:56		Join ps-auxw [0] (~arneb@p4FF7EE46.dip.t-dialin.net)
05:56:32		Quit panni_ (Quit: ( www.nnscript.de :: NoNameScript 3.81 :: www.XLhost.de ))
06:00
06:00:49		Quit factor (Read error: Connection reset by peer)
06:02:35		Quit CaptainKewl (Ping timeout: 245 seconds)
06:07:35		Quit KiwiCam (Ping timeout: 255 seconds)
06:09:17		Join Seed [0] (~ben@bzq-84-108-232-45.cablep.bezeqint.net)
06:11:25		Join jfc^3 [0] (~john@dpc6682208002.direcpc.com)
06:12:03		Join factor [0] (~factor@r74-195-220-23.msk1cmtc02.mskgok.ok.dh.suddenlink.net)
06:15:04		Join riotz_ [0] (riot33r@gateway/shell/shellium.org/x-ldbwroelemthymro)
06:15:10		Join TBFOOL [0] (~tb@c-f03271d5.09-42-73746f22.cust.bredbandsbolaget.se)
06:15:14		Join markun_ [0] (~markun@rockbox/developer/markun)
06:15:43		Join Naked [0] (~naked@naked.iki.fi)
06:15:56		Join mc2739_ [0] (~mc2739@rockbox/developer/mc2739)
06:16:35		Join banan_ [0] (~banan@c-94-255-221-206.cust.bredband2.com)
06:16:45		Quit riotz (.net .split)
06:16:45		Quit markun (.net .split)
06:16:45		Quit Judas_PhD (.net .split)
06:16:45		Quit Galois (.net .split)
06:16:46		Quit merbanan (.net .split)
06:16:46		Quit jfc^2 (.net .split)
06:16:46		Quit TBCOOL (.net .split)
06:16:46		Quit mc2739 (.net .split)
06:16:46		Quit Hadaka (.net .split)
06:16:50		Nick Naked is now known as Hadaka (~naked@naked.iki.fi)
06:17:26		Join Judas_PhD [0] (~kevin@misterfluffy.dsl.xmission.com)
06:30:30	S00row	I have a failure failure for RaaA in svn current:LD librockbox.so ... icon.o ... icon.c(.text.load_icons+0x9c): undefined reference to 'get_user_file_path'
06:31:31	S00row	s/failure/link/
06:40:28	***	Saving seen data "./dancer.seen"
06:48:01		Join mystica555_ [0] (~mike@c-75-70-179-25.hsd1.co.comcast.net)
06:48:01		Quit mystica555_ (Excess Flood)
06:48:33		Join mystica555_ [0] (~mike@c-75-70-179-25.hsd1.co.comcast.net)
06:49:00	JdGordon\|	S00row: sdl or android?
06:50:27	S00row	Android
06:54:04	JdGordon\|	clean svn?
06:54:15	S00row	JdGordon\|: Just did a clean rebuild and checked the config options - def Android. 480x800 and Normal
06:54:50	JdGordon\|	thats slightly messed up.. does icon.c have #include "rbpaths.h" ?
06:54:50	S00row	svn update didn't complain. How would i check that svn is clean?
06:55:10	S_a_i_n_t	svn stat
06:55:28	S_a_i_n_t	if that complains about anything, do "svn revert -R ."
06:55:32	S00row	no #include "rbpaths.h"
06:55:45	JdGordon\|	add it then
06:55:53	JdGordon\|	are there any compile warnings?
06:56:55	S00row	There were a couple
06:57:19	S00row	Looks like i had the remnants of a previous manually added patch. Reverted and building again now
06:57:47		Quit factor (Quit: Leaving)
07:00
07:01:06	S00row	Oops... sorry about that. Building fine
07:01:39		Join Galois [0] (djao@efnet-math.org)
07:17:49		Nick dys` is now known as dys (~andreas@krlh-5f720dfd.pool.mediaWays.net)
07:24:10		Quit shai (Quit: Leaving)
07:30:43		Join shai [0] (~Shai@l192-117-110-233.cable.actcom.net.il)
07:32:28		Join einhirn [0] (~Miranda@bsod.rz.tu-clausthal.de)
07:39:44		Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
07:42:50		Join factor [0] (~factor@r74-195-220-23.msk1cmtc02.mskgok.ok.dh.suddenlink.net)
07:49:47		Quit BHSPitMonkey (Remote host closed the connection)
07:50:22		Join stoffel [0] (~quassel@p57B4D3EA.dip.t-dialin.net)
08:00
08:06:38		Quit z35 (Ping timeout: 265 seconds)
08:08:59		Join z35 [0] (~z35@ool-18bdad71.dyn.optonline.net)
08:12:19		Quit tchan (Ping timeout: 240 seconds)
08:12:23		Join Zagor [0] (~bjst@rockbox/developer/Zagor)
08:16:07	CIA-7	New commit by Buschel (r28546): Introduce macros to ARM asm atrac3 filterbank for better readability.
08:16:47	amiconn	Zagor: ping
08:17:04	Zagor	yes?
08:17:38	amiconn	I have some mpeg layer 1 and layer 2 testfiles for inclusion in http://download.rockbox.org/test_files/
08:17:44	CIA-7	r28546 build result: All green
08:17:57	Zagor	amiconn: ok
08:21:40		Quit factor (Read error: Connection reset by peer)
08:22:20		Quit Judas_PhD (Quit: This is a quitting message)
08:24:43		Join factor [0] (~factor@r74-195-220-23.msk1cmtc02.mskgok.ok.dh.suddenlink.net)
08:29:10		Join wodz [0] (~wodz@chello087206240131.chello.pl)
08:29:18		Quit TheSeven (Ping timeout: 245 seconds)
08:29:20	wodz	Bagder: ping
08:31:17	wodz	Zagor: ping
08:31:28	Zagor	yes?
08:33:11	wodz	I asked Badger a few times but he seems to forget, could you upload mkmpioboot binaries to download server? http://www.datafilehost.com/download-63e5b4f1.html (windows) http://www.datafilehost.com/download-0118a6e2.html.(linux x86)
08:33:21	wodz	Zagor, ^
08:34:52	Zagor	should I put them in "utils"?
08:35:43	wodz	hmm I think other such tools are put in appropriate bootloader dir
08:36:04		Join ender` [0] (krneki@foo.eternallybored.org)
08:36:17	Zagor	indeed. I'll do that then.
08:37:20	wodz	btw. I think bootloader for mpio should be put in v1.0 subdir
08:38:02	wodz	I plan to release v2 in the future with graphics for battery charging and bootloader USB mode
08:39:20	Zagor	generally we only keep the latest bootloader on download.rockbox.org. I'm not sure why some ports have multiple versions.
08:40:28	Bagder	we only have that for safety, ability to step back
08:40:30	***	Saving seen data "./dancer.seen"
08:40:37	Bagder	but we shouldn't need to keep any particular history
08:41:06	wodz	ok no problem than
08:41:46	Zagor	wodz: bootloader/mpio/mkmpioboot added
08:43:55	wodz	win32 subdir seems to be empty
08:46:03	Zagor	the mirrors sometimes update in mysterious ways. give it a little time.
08:46:58	wodz	ok thanks
08:56:11		Join Judas_PhD [0] (~kevin@misterfluffy.dsl.xmission.com)
08:56:58	wodz	Zagor: something strange is happening - I get 404 or empty dir in random order
08:57:20	Zagor	that's because we have several mirrors and they are not updated yet
08:57:38	Zagor	each request can be handled by different mirrors
08:58:04	wodz	now I understand
08:58:59		Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
09:00
09:09:14	S_a_i_n_t	ml/csfe'hl/sgddbl/ffffffffffffffffffffffehhh/hiiii ffffffffffff qpuhidcv efhdwffffffffffffffffffffffffeje2pasjk.gkgk.23r8y0000hhhhhhhhhhhhhhhhhhhhhhhhhhhv3vvvvvvvvvvv4r4r4r4r4r4r4r4r4r4rasd
09:09:20	S_a_i_n_t	kj.jkjkjkjkjkjkjkjkjkjkjkjkjk.
09:09:56	S_a_i_n_t	gah! cat typing ^ :/
09:12:29		Join tchan [0] (~tchan@lunar-linux/developer/tchan)
09:14:22	CIA-7	New commit by wodz (r28547): MPIO manual - add link to mkmpioboot binary
09:16:01	CIA-7	r28547 build result: All green
09:19:28		Join petur [0] (d408b802@rockbox/developer/petur)
09:30:08		Quit bertrik (Ping timeout: 245 seconds)
09:39:09		Join Rob2222 [0] (~Miranda@p4FFF05B6.dip.t-dialin.net)
09:43:04		Quit Rob2223 (Ping timeout: 272 seconds)
09:49:57	wodz	pixelma, S_a_i_n_t: http://www.youtube.com/watch?v=O08JS9jy9kA
09:51:16	wodz	pixelma, S_a_i_n_t: http://www.youtube.com/watch?v=BUe9ocqEjBM
09:51:43	wodz	the animation is smooth actually the jumps comes from my capture program
09:54:19	S_a_i_n_t	I think mine fills the screen a bit better...the perhaps (just perhaps ;)) I'm biassed.
09:54:29	S_a_i_n_t	s/the/though/
09:55:30		Join LinusN [0] (~linus@rockbox/developer/LinusN)
09:56:07	wodz	honestly I like the size of pixelma's version and overall look of yours
09:56:25	S_a_i_n_t	I can scale it.
10:00
10:03:05	wodz	S_a_i_n_t: http://yfrog.com/45sizecompj <- comparision with USB logo size
10:04:47	wodz	and the 'fill area' has to have even width because the filling image is 2pix width
10:06:05	S_a_i_n_t	Hmmm...ok, so the internal width of the battery need be /2 ?
10:06:19	pixelma	wodz: there is a slightly smaller greyscale USB logo in the apps/bitmaps/remote_native, I guess I'd prefer that to something that fills the complete screen (width)
10:06:40	S_a_i_n_t	pixelma: Read my mind.
10:06:48	wodz	pixelma: ok I'll look at it
10:06:57	S_a_i_n_t	I was just going to say I don;t really like the way the USB logo fills the screen.
10:07:59	S_a_i_n_t	verry messy desktop btw wodz ;p
10:08:05	S_a_i_n_t	-r
10:09:09	wodz	S_a_i_n_t: not at all - on the left side there are my icons on the right side my wife's. The windows fills empty space on this screenshot :-)
10:09:39	S_a_i_n_t	I just have the recycle bin on mine ;)
10:09:43	pixelma	I guess I'm biased too but I like my battery icon better overall - it is the same style as the USB logo and doesn't look as "angled" (?), I also don't like the simple outline where the rest is 3D-ish. Not sure this should turn into art critics here though
10:10:46	wodz	pixelma: It's not that your work is bad - it just look a bit to clumsy IMO
10:11:16	S_a_i_n_t	making it "bland" looking was kinda deliberate...I don't really think that trying to do "nice" shading ever really looks right in 4-grey
10:11:49	S_a_i_n_t	I think it needs to be more, "clean"...but pixelma's right, this 'aint art school ;)
10:12:00	pixelma	huh?
10:12:24	wodz	well USB logo looks perfect
10:17:14		Quit JdGordon\| (Quit: leaving)
10:21:28		Join JdGord [0] (~jd@pa58-109-129-213.pa.nsw.optusnet.com.au)
10:36:38		Join DerPapst [0] (~Alexander@178.4.16.73)
10:40:32	***	Saving seen data "./dancer.seen"
10:42:14		Quit Seed (Ping timeout: 245 seconds)
10:44:20		Join sasquatch [0] (~username@p4FF2D445.dip.t-dialin.net)
10:46:55		Join Seed [0] (~ben@bzq-84-108-232-45.cablep.bezeqint.net)
10:50:05		Join xxcv [0] (~IMNOTADEV@c211-30-174-99.carlnfd1.nsw.optusnet.com.au)
10:50:29		Quit JdGord (Quit: Bye)
10:50:30	wodz	what is the point of firmware/drivers/rtc/rtc_tcc77x.c in its current form?
10:54:17		Quit wodz (Quit: Leaving)
10:59:27		Quit stoffel (Remote host closed the connection)
11:00
11:07:18		Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
11:12:17		Join JdGord [0] (~jd@pa58-109-129-213.pa.nsw.optusnet.com.au)
11:20:19		Join casainho [0] (~chatzilla@pal-213-228-181-14.netvisao.pt)
11:49:22		Quit ajb (Remote host closed the connection)
11:49:57		Quit Seed (Ping timeout: 265 seconds)
11:54:31		Join Seed [0] (~ben@bzq-84-108-232-45.cablep.bezeqint.net)
12:00
12:02:51		Join user890104_ [0] (~Venci@212.233.239.19)
12:03:03		Quit user890104 (Ping timeout: 272 seconds)
12:11:24		Quit JdGord (Ping timeout: 245 seconds)
12:12:10		Join InsDel [0] (~haqr.net@c-98-231-87-43.hsd1.fl.comcast.net)
12:15:21		Quit xxcv ()
12:19:02		Join JdGord [0] (~jd@pa58-109-129-213.pa.nsw.optusnet.com.au)
12:20:21		Join teru [0] (~teru@KD059133111160.ppp.dion.ne.jp)
12:21:34		Join MethoS- [0] (~clemens@134.102.106.250)
12:22:45		Quit MethoS- (Remote host closed the connection)
12:30:01		Quit user890104_ ()
12:30:09		Join user890104 [0] (Venci@Venci-Notebook-LAN.ipv6.6bez10.info)
12:40:35	***	Saving seen data "./dancer.seen"
12:51:11		Quit InsDel (Read error: Connection reset by peer)
12:52:33		Join n1s [0] (~n1s@nl118-174-240.student.uu.se)
12:52:33		Quit n1s (Changing host)
12:52:33		Join n1s [0] (~n1s@rockbox/developer/n1s)
12:57:45		Quit JdGord (Quit: Bye)
13:00
13:01:24		Quit petur (Quit: Page closed)
13:08:29		Join dfkt [0] (dfkt@unaffiliated/dfkt)
13:24:45		Quit JdGordon (Ping timeout: 272 seconds)
13:25:36	*	TheSeven wonders why apple calls the i2s interface or codec "MeCCA"
13:26:00	*	TheSeven found strings like "MeCCAAudioIOWorkerThread"
13:26:02	n1s	Buschel: There was no need to revert r28488 for the ARM7TDMI case if it only slowed down your nano's core, i did test on my fuzev1 which has a ARM9TDMI and saw no speed diff there, also that last add insruction that just increases a pointer in a caller saved reg that is just subsequently discarded is still not needed.
13:26:26	S_a_i_n_t	TheSeven: try facing it in the correct direction and see if it prays?
13:29:47		Quit krazykit (Quit: awe yeeeeeee)
13:36:30	pixelma	teru: did you test compile the unified image viewer for targets with small plugin buffer?
13:37:38	teru	pixelma: yes.
13:37:54	pixelma	not sure how your patch actually works - if everything is in one imageviewer binary or if "decoders" get loaded as needed
13:39:57	teru	the latter. decoders are loaded when needed.
13:40:30	pixelma	ah, nice. Thanks for the info :)
13:40:50		Quit avacore (Ping timeout: 252 seconds)
13:41:36		Join avacore [0] (~avacore@1008ds1-rdo.0.fullrate.dk)
13:45:51		Join krazykit [0] (~kkit@99-126-205-52.lightspeed.cicril.sbcglobal.net)
14:00
14:02:56	*	S_a_i_n_t wonders if someone has a target handy to attempt to reproduce a bug with pictureflow
14:03:00		Quit riotz_ (Changing host)
14:03:00		Join riotz_ [0] (riot33r@unaffiliated/riotz)
14:03:00		Quit riotz_ (Changing host)
14:03:00		Join riotz_ [0] (riot33r@gateway/shell/shellium.org/x-ldbwroelemthymro)
14:03:11		Nick riotz_ is now known as riotz (riot33r@gateway/shell/shellium.org/x-ldbwroelemthymro)
14:07:11		Quit fyrestorm (Quit: Ur skills' fireproof like a wooden panel -- U got feds talking leet on your IRC channel!)
14:12:06		Quit Seed (Ping timeout: 255 seconds)
14:12:19	S_a_i_n_t	If someone with a target with database enabled could test this, it's be great as it seems plain wrong to me:
14:12:21	S_a_i_n_t	Commence playback from either the database or filebrowser (set your WPS hotkey to Pictureflow first), then bring up Pictureflow with the hotkey, exit pictureflow using whatever means (which should drop you back to the WPS), stop playback, this launches the pictureflow plugin...why?
14:13:15	S_a_i_n_t	I would expect it to launch pictureflow on playback stop if I commenced playback with pictureflow, but then and only then.
14:14:12	S_a_i_n_t	dropping a user into pictureflow on stop just because they (may have even accidentally) opened pictureflow with the hotkey after commencing playback with db or filebrowser seems quite odd behaviour to me.
14:17:43		Join Seed [0] (~ben@bzq-84-108-232-45.cablep.bezeqint.net)
14:19:46	mc2739_	S_a_i_n_t: I cannot reproduce that behavior with e200v2
14:20:22		Nick mc2739_ is now known as mc2739 (~mc2739@rockbox/developer/mc2739)
14:20:54		Join LambdaCalculus37 [0] (~3f74f70d@rockbox/staff/LambdaCalculus37)
14:20:55	S_a_i_n_t	Hmmm...quite odd. I thought there may have been a chance one of my patches was messing something up, so I tried an svn build and got the same results.
14:20:58	teru	pictureflow seems to be counted as a "browser" when you launched with PicutreFlow in wps context menu or hotkey.
14:22:22	S_a_i_n_t	I would find that behaviour quite normal if I launched playback with pictureflow, but it seems to be very strange - if not a bug, to do so when playback is launched from the database.
14:22:57		Quit evilnick (Quit: Leaving)
14:24:32	S_a_i_n_t	Hmmm...ok, even weirder. It seems like it is one of my config settings messing things up.
14:24:40	S_a_i_n_t	the plot thickens.
14:29:34		Join CaptainKewl [0] (~jason@207-38-215-126.c3-0.nyr-ubr1.nyr.ny.cable.rcn.com)
14:32:10	S_a_i_n_t	Ok, it turns out that "Follow Playlist" causes this behaviour.
14:32:35	S_a_i_n_t	I'm not 100% about what follow playlist does, but it doesn't seem like it should have that effect.
14:33:27	n1s	it is that setting that meks the file browser end up in the dir that the current playing track lives in
14:33:36	n1s	s/meks/makes/
14:33:57	S_a_i_n_t	ah...then based on what I'm experiencing it's surely a bug.
14:34:03	S_a_i_n_t	just need someone to verify it.
14:34:24	S_a_i_n_t	mc2739: Care to try again with "Follow Playlist" enabled?
14:34:32	S_a_i_n_t	+please
14:40:39	***	Saving seen data "./dancer.seen"
14:44:06	mc2739	S_a_i_n_t: turning on "Follow Playlist" does reproduce your described problem
14:44:28	S_a_i_n_t	mc2739: Good to know, thanks.
14:44:36	S_a_i_n_t	I shall flyspray the sucker.
14:48:12		Quit krazykit (Quit: awe yeeeeeee)
14:54:37		Quit teru (Quit: Quit)
14:55:42	mc2739	S_a_i_n_t: something else I noticed with that configuration, after stopping playback and going back to pictureflow, I exit pictureflow again to return to the menu, now the down button acts like a hotkey and returns me back to pictureflow
14:56:07	S_a_i_n_t	mc2739: yes, very strange.
14:56:29	S_a_i_n_t	Something is definitely quite wrong...I had a brief look but it is so far beyond me it isn't even funny.
15:00
15:01:09		Part LinusN
15:03:05		Join esperegu [0] (~quassel@145.116.15.244)
15:06:24	LambdaCalculus37	Can someone with a non-iPod PP target take a look at the battery reading in System > Rockbox Info? On my HDD6330, I'm noticing that it's reading 0h -1m at all times.
15:06:36	LambdaCalculus37	I want to see if this is also affecting other PP targets.
15:08:00	n1s	looks fine on my c200 88% 15h 32m
15:08:35	n1s	although the build on that is from before the recent code juggling in the debug menu
15:08:48		Join designate72 [0] (~aaron@adsl-065-013-002-216.sip.asm.bellsouth.net)
15:09:30	n1s	i'll update and check
15:09:40	LambdaCalculus37	n1s: Okay.
15:10:13	*	LambdaCalculus37 should check on his m:robe 100 later on
15:10:32		Join komputes [0] (~komputes@ubuntu/member/komputes)
15:10:57	n1s	still fine with an up to date build, and when i think about it, that isn't the debug menu :)
15:11:21	n1s	does the battery reading work and is the runtime thing calibrated for that player?
15:11:24	S_a_i_n_t	TheSeven: I know it's quite late now...but I was just reading your comments on FS #5111 and I couldn't get it to work on Nano2G myself.
15:12:47	LambdaCalculus37	n1s: Probably related to inaccurate battery measurement on the HDD63x0 itself. Although IIRC my m:robe 100 exhibited the same issue the last time I looked at it.
15:13:18		Quit LambdaCalculus37 (Quit: CGI:IRC)
15:13:36		Join wodz [0] (~wodz@chello087206240131.chello.pl)
15:13:44		Join LambdaCalculus [0] (~3f74f70d@giant.haxx.se)
15:13:50	wodz	LambdaCalculus: ping
15:14:04	LambdaCalculus	wodz: Pong
15:14:14		Nick LambdaCalculus is now known as LambdaCalculus37 (~3f74f70d@giant.haxx.se)
15:14:24	TheSeven	S_a_i_n_t: what exactly was the problem?
15:14:42	wodz	LambdaCalculus37: have you tried patch on your AJBR ?
15:15:06	n1s	LambdaCalculus37: i see the problem
15:15:07	LambdaCalculus37	n1s: I was about to say before I was rudely interrupted that I'm going to ask lowlight and b0hoon about how far along battery callibration readings are on the HDD63x0.
15:15:14	LambdaCalculus37	n1s: What's the problem?
15:15:16	S_a_i_n_t	No click...speaking of which (and possibly I have asked you this before) do you have SW keyclicks on your Nano2G without playback?
15:15:19	LambdaCalculus37	wodz: I have. Works there too.
15:15:19	S_a_i_n_t	TheSeven: ^
15:15:38	S_a_i_n_t	For me, on all my Nano2Gs SW keyclick only works during audio playback.
15:15:49	*	TheSeven hasn't checked sw keyclicks at all
15:16:01	wodz	LambdaCalculus37: nice
15:16:06	wodz	thank you
15:16:26	n1s	LambdaCalculus37: the gogears don't define the CURRENT_NORMAL thing that tells powermanagement how much power it sucks so battery_status_update() never updates the remaining runtime, see powermgmt.c:278
15:17:22	n1s	so powermgmt_est_runningtime_min is always -1 which is where your -1m remaining runtime comes from
15:17:38	wodz	hdd6330 which b0hoon own is defective IIRC. So he couldn't work on battery code.
15:18:25	LambdaCalculus37	wodz: And I have to look at the service manual for the HDD63x0 to see what the battery specs are.
15:18:28		Join LinusN [0] (~linus@rockbox/developer/LinusN)
15:18:56	wodz	service manual - what a luxury
15:19:09	S_a_i_n_t	amen.
15:19:33	S_a_i_n_t	I can't even get one for my VCR
15:19:51	LambdaCalculus37	S_a_i_n_t: Why, you planning on porting Rockbox to it? ;)
15:20:14	S_a_i_n_t	.....<looks around shiftily>...maybe ;)
15:21:07		Quit Seed (Ping timeout: 260 seconds)
15:21:22		Join JdGordon [0] (~jonno@rockbox/developer/JdGordon)
15:22:10	LambdaCalculus37	n1s: I think I can get this fixed.
15:22:27	Stummi	is there a special reason why the lamp-plugin has only white and one more color? I would find it better if there are at least a simple palette of basic colors (ff000, ffff00, 00ff00, 00ffff, 0000ff and ff00ff)
15:22:56	S_a_i_n_t	Stummi: I had a patch in the tracker for that, but it'll be out of sync now.
15:22:57	n1s	LambdaCalculus37: yes, you should only need to do a battery bench to calculate the normal runcurrent
15:23:40	S_a_i_n_t	and fwiw, I asked the same question to myself many time, which is why I added a full ROYGBIV colour-set
15:24:04	S_a_i_n_t	oh, and black...which was kinda cool too ;)
15:24:12	Stummi	S_a_i_n_t, can you point me to that? A search for "lamp" has only 3 results and this what you means don't semms there
15:25:02	S_a_i_n_t	Stummi: 'twas probably closed, two of its features got committed, and the ohter is out of sync now.
15:25:26	S_a_i_n_t	if you look at lamp.c, it really isn't hard to see how it's done.
15:25:42	S_a_i_n_t	just remember to update the total colours ammount.
15:26:21		Join Seed [0] (~ben@bzq-84-108-232-45.cablep.bezeqint.net)
15:26:58	S_a_i_n_t	I shall add a patch to the tracker again with another colour-set for lamp. but it most likely won't get committed.
15:27:20	Stummi	yes i saw already how it's done
15:27:22	Stummi	why not?
15:27:56	S_a_i_n_t	If it was going to get committed, I figure it would have been committed when the other features from the same patch were added.
15:28:27	S_a_i_n_t	I added brightness, and made it not time out from idle timeout.
15:29:43	Stummi	hm, and the developer who commited it does not give you a reason for that?
15:30:07	Stummi	if there is not a special reason for this, i just would try it a second time
15:30:40	S_a_i_n_t	It's such a simple thing to do, I would assume if it was wanted in core it would have been done already.
15:30:44		Part LinusN
15:30:55	S_a_i_n_t	It's not like it's sitting waiting for a f33t coder to do it ;)
15:31:06	S_a_i_n_t	hehe... *l33t
15:31:53		Quit n1s (Quit: L�mnar)
15:32:05	Stummi	yes, i know. Adding new colors is realy easy (since you can edit c-source-files. For "normal" rb-users, it won't be easy :))
15:32:50	LambdaCalculus37	Stummi: There were "normal" Rockbox users that eventually became developers.
15:32:59	S_a_i_n_t	"normal users" get what they're given ;)
15:34:29	Stummi	hehe
15:35:29		Quit CaptainKewl (Ping timeout: 240 seconds)
15:38:42	wodz	I tend to think that 'normal
15:38:50	wodz	' rb user IS developer
15:39:47	Stummi	hm, i think there were also users, whitch don't whant have much to do with the developing-process
15:40:46	gevaerts	Plenty of those
15:42:41		Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
15:50:18		Quit advcomp2019_ (Read error: Connection reset by peer)
15:51:07		Join advcomp2019_ [0] (~advcomp20@unaffiliated/advcomp2019)
15:52:37		Quit dionoea (Ping timeout: 255 seconds)
15:53:28		Join dionoea [0] (~dionoea@yop.chewa.net)
15:56:21		Part Seed
16:00
16:03:39		Join kugel [0] (~kugel@rockbox/developer/kugel)
16:05:05		Join jgarvey [0] (~jgarvey@cpe-065-190-066-089.nc.res.rr.com)
16:08:03		Join MethoS- [0] (~clemens@134.102.106.250)
16:08:21		Join Alchimysta [0] (~97513d5e@giant.haxx.se)
16:08:30		Quit Alchimysta (Client Quit)
16:10:39		Quit MethoS- (Remote host closed the connection)
16:21:39		Join liar [0] (~liar@clnet-p09-185.ikbnet.co.at)
16:23:53		Join TheLemonMan [0] (~lemonboy@ppp-151-213.32-151.iol.it)
16:25:22	CIA-7	New commit by kugel (r28548): Touchscreen: Improved scroll threshold ...
16:26:58	CIA-7	r28548 build result: All green
16:29:01	kugel	oh, I forgot to mentioned: I fixed LCD_WIDTH/LCD_HEIGHT for onda vx767 as mcuelenaere suggested (hence the binsize increase)
16:30:18		Join captainkewll [0] (2669ecc2@gateway/web/freenode/ip.38.105.236.194)
16:30:19	kugel	wodz: r28544 removed the pitchscreen on the hd200, I think that should've been a separate commit or at least be mentioned in the ci message. What was the reason?
16:30:46		Join saratoga [0] (9803c6dd@gateway/web/freenode/ip.152.3.198.221)
16:31:19	kugel	I guess it's awkward to use without front buttons
16:31:55	saratoga	is " __attribute__ ((aligned (8)))" the correct way to make sure a buffer is double word aligned?
16:32:20	kugel	saratoga: yea, see CACHE_ALIGN macros
16:33:10	saratoga	looking at atrac, i see a lot of things are 16 byte aligned, I guess thats so they line up with a cacheline?
16:33:56	kugel	I think 16byte for cache line is only correct on PP (if at all), the standard arm caches need 32 byte
16:34:15	saratoga	good to know
16:34:39	kugel	could be 32 byte for PP as well, I don't know for sure
16:35:24	saratoga	on arm11, are ldm instructions going to be slow if I don't ensure that they're even word aligned?
16:35:39	saratoga	hmm yes they are
16:36:24	saratoga	so i start off by loading three words, then always load two sequentially, and the buffer is aligned, thats going to screw up access times on arm11 right?
16:38:33	kugel	hm we don't seem to set CACHEALIGN_ATTR correctly for the standard arm caches
16:38:42	kugel	nobody sets it, and it defaults to 16
16:39:55	Torne	i thought we were fairly confident it was 16 for PP
16:40:00	Torne	based on someone's benchmark sometime
16:40:11	Torne	but yeah, it is 32 for most ARM caches on ARM9/ARM11
16:40:43	***	Saving seen data "./dancer.seen"
16:41:16	saratoga	anyone feel lik ebenchmarking on the beast?
16:45:21		Nick markun_ is now known as markun (~markun@rockbox/developer/markun)
16:50:56		Join domonoky [0] (~Domonoky@rockbox/developer/domonoky)
16:51:31		Quit Judas_PhD (Quit: This is a quitting message)
16:52:03		Part Zagor
16:53:25	wodz	CFs also like 16 alignment as it speeds up dram access (it can utilize bursts than)
16:54:17	pixelma	kugel: not the pitchscreen but the quickscreen, I agree about the seperate commit though. There are other targets without it in case of the Ondios because of "no spare button to access it easily" or in the case of the Player because of no space on the display
16:54:41	kugel	oops yes quickscreen, typo
16:55:48	wodz	It was separate in my repo but I must make mistake when rebasing :/
16:56:15	pixelma	I also find it a bit silly on the c200 where it's easily accessible from the WPS but on some weird button combo from the menus but you can't do much to improve that if you want to maintain an intuitive moving through the lists
16:58:07		Join Judas_PhD [0] (~kevin@misterfluffy.dsl.xmission.com)
17:00
17:07:03	TheSeven	kugel/saratoga: S5L8701 cacheline size is 16 bytes. maybe that's one of the causes of Buschels weird performance measurements?
17:08:13		Quit kugel (Ping timeout: 255 seconds)
17:15:09	saratoga	i got my old gigabeast booting, but when i connect USB it often freezes on disconnect, anyone else noticing that?
17:17:42	saratoga	oh its not crashed, the backlight just didn't turn on
17:18:07		Quit TheSeven (Ping timeout: 255 seconds)
17:18:15	Torne	that's a known bug i think
17:18:23	Torne	certainly i have heard people complaining about it before :)
17:21:01	saratoga	GigabeatS ATRAC3: 61.78MHz (ARMv4) 46.69(ARMv5E)
17:21:03	saratoga	not bad
17:21:49	saratoga	i need to play around with scheduling more on arm11
17:21:53	saratoga	i bet i can get it faster
17:25:03	saratoga	wow it still takes 15MHz to run
17:25:09	saratoga	just for the QMF
17:25:42	saratoga	Torne: can I use r11 if I stack it?
17:25:47	Torne	sure
17:25:54	Torne	r11 isn't special
17:27:47	saratoga	hmm ok then I can probably do this a little better
17:28:05	saratoga	is it just the stack pointer and pc that are special for us?
17:29:05	Torne	yeah
17:29:26	Torne	r0-r3 and r12 are free to use for anything; r4-r11 you have to save/restore, but they're not special
17:30:33	saratoga	what about r14?
17:31:13	Torne	r14 you can overwrite, but you need to store the return address somewhere
17:31:23	Torne	you don't have to put it back into r14, though, you ca nrestore it from the stack directly to pc
17:35:34		Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
17:41:32		Join ajb [0] (~user@cbnluk-gw0.cambridgebroadband.com)
17:41:48	saratoga	ensuring that all ldm's were double word aligned saved another 3MHz on the beast
17:50:17		Quit jgarvey (Ping timeout: 250 seconds)
17:52:38		Join stoffel [0] (~quassel@p57B4BCDF.dip.t-dialin.net)
17:54:12		Join jgarvey [0] (~jgarvey@cpe-065-190-066-089.nc.res.rr.com)
17:57:09		Quit antil33t (Read error: Connection reset by peer)
17:57:19		Join antil33t [0] (~Mudkips@124-197-51-80.callplus.net.nz)
18:00
18:02:19	saratoga	forcing 32 byte alignment of all my buffers doesn't seem to make a further difference
18:04:16	saratoga	anyone feel like testing on coldfire or something just to make sure I didn't screw up atrac?
18:04:22		Quit Guinness (Read error: Connection reset by peer)
18:11:21		Quit designate72 (Quit: Leaving)
18:17:42		Join thegeek [0] (~nnscript@172.80-203-148.nextgentel.com)
18:20:47		Join benedikt93 [0] (~benedikt9@unaffiliated/benedikt93)
18:28:28		Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
18:30:41		Join domonoky1 [0] (~Domonoky@agsb-4d0494dd.pool.mediaWays.net)
18:31:51		Quit swilde (Remote host closed the connection)
18:32:15		Quit domonoky (Ping timeout: 245 seconds)
18:36:17		Quit markun (Read error: Connection reset by peer)
18:39:16		Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
18:40:19		Quit DerPapst (Quit: Leaving.)
18:40:44	***	Saving seen data "./dancer.seen"
18:46:16		Join lixus00 [0] (~lixxus@78-86-253-21.zone2.bethere.co.uk)
18:49:51		Part lixus00
18:59:53		Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
19:00
19:07:13		Join markun [0] (~markun@5ED33C2C.cm-7-4a.dynamic.ziggo.nl)
19:07:14		Quit markun (Changing host)
19:07:14		Join markun [0] (~markun@rockbox/developer/markun)
19:10:02		Quit evilnick_B (Quit: Page closed)
19:12:42		Join GeekShadow [0] (~Antoine@44.181.204-77.rev.gaoland.net)
19:12:43		Quit GeekShadow (Changing host)
19:12:43		Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
19:15:06		Quit markun (Read error: Connection reset by peer)
19:15:58		Quit casainho (Remote host closed the connection)
19:16:55		Join evilnick_B [0] (0c140464@rockbox/staff/evilnick)
19:19:35	saratoga	do i need to set the svn properties on new files?
19:22:55	saratoga	ah nevermind found the guide
19:25:18		Join markun [0] (~markun@rockbox/developer/markun)
19:29:31	CIA-7	New commit by saratoga (r28549): Commit FS #11709 by me. Introduces ARMv5E optimized iQMF for atrac3 based on packed multiply instructions. Additionally, improves scheduling on arm9 ...
19:31:14	CIA-7	r28549 build result: 28 errors, 64 warnings (saratoga committed)
19:31:22	saratoga	bah
19:34:29	CIA-7	New commit by saratoga (r28550): Remove some benchmarking code accidentally included in the previous commit.
19:34:44		Quit factor (Quit: Leaving)
19:35:51		Join Jerom [0] (~jerome@95.171.136.54)
19:36:21	CIA-7	r28550 build result: All green
19:36:36	saratoga	don't know how i managed to commit the version with the new code disabled :(
19:39:09	LambdaCalculus37	CRAP... MajorChanges page is borked. =/
19:39:48	LambdaCalculus37	How does adding a R% tag cause an entire section to disappear?
19:40:13	*	LambdaCalculus37 meant a linebreak tag
19:41:46		Join factor [0] (~factor@r74-195-220-23.msk1cmtc02.mskgok.ok.dh.suddenlink.net)
19:41:49	saratoga	yeah i didn't get that
19:41:56	saratoga	but you clicked edit before me, so its your problem now
19:41:56		Quit factor (Read error: Connection reset by peer)
19:43:23		Join factor [0] (~factor@r74-195-220-23.msk1cmtc02.mskgok.ok.dh.suddenlink.net)
19:45:03		Join Buschel [0] (~chatzilla@p54B6697D.dip.t-dialin.net)
19:45:58		Quit LambdaCalculus37 (Quit: CGI:IRC)
19:47:43		Join LambdaCalculus37 [0] (~3f74f70d@rockbox/staff/LambdaCalculus37)
19:56:03	CIA-7	New commit by bertrik (r28551): tcctool: add support for Cowon J3 - FS #11747
19:56:06		Join DerPapst [0] (~Alexander@93.229.183.167)
19:57:50	amiconn	saratoga: If you typedef the respective int type used for the qmf window, you'll save several ifdefs
19:58:06	saratoga	amiconn: ah good idea
19:58:09	amiconn	I did it this way in libdemac for the filters
19:58:13	CIA-7	r28551 build result: All green
19:59:42	amiconn	saratoga: Also, has this been tested on arm11?
19:59:55	saratoga	amiconn: yes, its much faster
20:00
20:00:05	saratoga	theres numbers in the commit message for the beast
20:01:42	amiconn	Ah, somehow missed those
20:02:06	saratoga	i should have said "gigabeat S" instead of "arm11" in that message since I bet the beast and an android phone aren't quite the same performance-wise
20:02:42	saratoga	but yes i put a good deal of effort into scheduling on the beast, and the final version was much faster then the optimal arm9e version
20:07:25		Join Horscht [0] (~Horscht@p4FD4D8C8.dip.t-dialin.net)
20:07:25		Quit Horscht (Changing host)
20:07:25		Join Horscht [0] (~Horscht@xbmc/user/horscht)
20:08:39		Join b0hoon [0] (~quassel@public-gprs106885.centertel.pl)
20:12:15		Join EA [0] (chipper@chipsworld.llamas.net)
20:13:43		Join InsDel [0] (~haqr.net@c-98-231-87-43.hsd1.fl.comcast.net)
20:15:47		Join panni_ [0] (hannes@ip-178-203-81-220.unitymediagroup.de)
20:16:56	b0hoon	LambdaCalculus37: ping
20:18:13	LambdaCalculus37	b0hoon: pong
20:18:53		Join sideawake [0] (~590008c1@giant.haxx.se)
20:19:31	b0hoon	LambdaCalculus37: hi, hdd1630 has 32MB of RAM, page 19 on the scheme : 256Mbit
20:20:01	b0hoon	LambdaCalculus37: so it looks like there is a typo on the diagram
20:20:52	b0hoon	LambdaCalculus37: about the battery: there is no estimation because a few targets dont have a CURRENT defines.
20:20:53	sideawake	Hi. A quick question about themes: I switched to cni_aquamarine and back to cabbiev2, and everything is fine except the folder names in the file browser which now seem to have permanently turned to blue, with no obvious way to change that. Is there any way to restore the default colors without deleting the entire .rockbox folder and reinstalling Rockbox? (I have r28518 installed. One of the two skins was built-in and the other was freshly downloaded; th
20:21:20	LambdaCalculus37	b0hoon: That may be so. I was wondering why the block diagram in the service manual says 64MB of RAM. Must've been a copy+paste from the HDD6330, perhaps.
20:21:59	saratoga	does anyone actually have an atrac3 encoder?
20:22:08	LambdaCalculus37	b0hoon: I also remember trying to build a HDD1630 build with 64MB of RAM configured and having the whole thing crash like the stock market.
20:23:11	b0hoon	LambdaCalculus37: i'm working on disassembling OF of the hdd6330 but there is no any other interesting code for the LCD
20:24:36	b0hoon	LambdaCalculus37: PPs LCD2 bridge is too slow for this size of LCD, i suppose
20:25:27	LambdaCalculus37	b0hoon: We only have one other PP target with a 220x176 LCD, and that's the iPod Color.
20:27:01	bertrik	hm? e200 has 220x176 too and has a PP5024
20:27:05	b0hoon	LambdaCalculus37: yes, and i think here is solution for speed up the bridge :) i will look at it later.
20:27:30	Buschel	interesting, using CACHEALIGN_ATTR on some of the relevant mpc arrays delivers more consistent performance results. e.g. the negative impact of r28488 becomes hardly measurable
20:27:37	pixelma	sideawake: there probably was a config file that changed filetype colours and since cabbiev2 doesn't set it to something else they stay. Unfortunately I don't remember how the settings nam is called exactly but I guess you can find out when looking through the config.cfg in the .rockbox folder and delete the line with it, save the file and reboot
20:27:44	Buschel	...for S5L8701
20:27:51	LambdaCalculus37	b0hoon: Okay, no problem.
20:28:00	pixelma	bertrik: the e200 is 176x220
20:28:33	pixelma	oh it was about the amount of data then it shouldn't matter
20:28:34	*	b0hoon is looking at the e200...
20:29:42		Join Kupop [0] (~Kupo@cpc2-bsfd7-2-0-cust220.5-3.cable.virginmedia.com)
20:32:17	b0hoon	bertrik: transfers are done through the DMA somehow
20:32:54	saratoga	Buschel: on arm11 ensuring that all ldm instructions were double word aligned made a surprisingly large difference
20:33:13	sideawake	pixelma: thanks, there was a line "filetype colours", I deleted it and everything is fine : )
20:33:20	saratoga	i suppose because of the 64 bit cache bus, but also probably because of better caching
20:34:02		Quit sideawake (Quit: CGI:IRC 0.5.9 (2006/06/06))
20:34:20	Buschel	yep, I read this in the logs. on the S5L there is not much differen (~0.15 MHz saved), but r28488 now only makes a minor difference (0.08 MHz slower instead of 0.7 MHz)
20:40:45	***	Saving seen data "./dancer.seen"
20:52:33		Quit kevku (Ping timeout: 272 seconds)
20:54:14		Quit LambdaCalculus37 (Quit: CGI:IRC)
20:55:48		Join LambdaCalculus37 [0] (~3f74f70d@rockbox/staff/LambdaCalculus37)
21:00
21:00:24		Quit stoffel (Remote host closed the connection)
21:02:26		Quit feisar- (Ping timeout: 276 seconds)
21:04:22		Join bertrik_ [0] (~bertrik@rockbox/developer/bertrik)
21:04:59		Quit bertrik (Read error: Connection reset by peer)
21:05:04		Nick bertrik_ is now known as bertrik (~bertrik@rockbox/developer/bertrik)
21:07:34		Join anewuser [0] (anewuser@unaffiliated/anewuser)
21:08:14		Quit b0hoon (Quit: GTG, bye.)
21:26:29		Join MItaly [0] (~chatzilla@151.57.195.254)
21:27:05	MItaly	Hi,
21:27:08	MItaly	just a question
21:27:23	MItaly	is it likely that FS #8961 will be merged in SVN at some time?
21:28:40	MItaly	FWIW, I use it since August without experiencing any issue, and in my opinion it's a really nice enhancement for the rockbox look
21:30:30		Join freddyb [0] (~chatzilla@216.8.239.112.etczone.com)
21:35:29	freddyb	MItaly: I would like to see that, too. I think the core patch is pretty solid. I (or someone smarter) needs to go over convttf and fix it up. I've been busy with other stuff. :(
21:35:29	saratoga	freddyb: you were working on the fonts patch just mentioned?
21:36:42	MItaly	freddyb: again, FWIW, the latest version worked fine for me, at least with DejaVu Sans, but indeed I remember having some problems in several occasions with "strange" fonts
21:37:52	freddyb	Saratoga: I basically resycned it back in July/August because it had been idle for quite some time.
21:38:33	saratoga	ah ok
21:38:38	freddyb	MItaly: There were some versions of convttf that made bad font widths
21:38:40	saratoga	what was actually missing from that patch?
21:39:55	MItaly	IIRC the last one worked out of the box
21:40:04	freddyb	Saratoga: I don't think there's anything functional missing from the patch but convttf has been handled by several different people and could use some clean up. It works, tho.
21:40:20	pixelma	I wonder how it deals with greyscale and monochrome displays. Also, I see that Unhelpful took part in development, would be curious to see his opinion
21:40:35	freddyb	I haven't touched it for a while, though.
21:41:22	freddyb	I don't remember about grey/mono.
21:43:23	Unhelpful	pixelma: greyscale displays don't offer enough levels to do decent AA and also tend to be of relatively low spatial resolution. i've wanted to support producing mono fonts from convttf for a while, as something i think could be useful, but i kind of fell out of rockbox dev work last time i took up work on audiomangler again
21:43:24		Join Topy [0] (~Topy44@f049078140.adsl.alicedsl.de)
21:44:43	Unhelpful	basically right now antialiasing is only of value for color targets, and i guess possibly greylib, and convttf is only useful for producing 4-bit fonts for the antialiasing patch.
21:46:05	pixelma	well, I was more curious about it not breaking anything on these displays or add unnecassary code then. For usual targets I don't see the need for AA fonts at all, maybe for very high resolution android devices running on RaaA, just a personal opinion
21:46:21	pixelma	unnecessary too
21:47:16		Quit T44 (Ping timeout: 245 seconds)
21:47:25	MItaly	looking at http://www.rockbox.org/wiki/BuyersGuide
21:47:43	Unhelpful	ah, no, it shouldn't break anything on mono/greyscale devices, but obviously they can't do anything useful w/ 4-bit fonts.
21:47:43	MItaly	I see that all the grayscale targets offer just 2-bit grayscale
21:48:05	MItaly	so I agree it can't be useful at all on grayscale targets
21:48:14	MItaly	well, not even on BW screens :D
21:48:49	saratoga	is there much binsize hit on grayscale?
21:49:31	freddyb	The AA fonts patch allows the use of TTF fonts for themes, which is nice. It's not easy for the average dude to get a nice looking mono font from TTF.
21:50:37	freddyb	Not much, but the font buffers should be 4x if AA is going to be used.
21:56:32		Quit Buschel (Ping timeout: 276 seconds)
21:56:55		Join Buschel [0] (~chatzilla@p54A3B3F3.dip.t-dialin.net)
21:57:52		Quit LambdaCalculus37 (Quit: done)
22:00
22:04:37		Join Quelsaruk [0] (~jose@89.226.117.91.dynamic.mundo-r.com)
22:04:43	Quelsaruk	hi
22:05:36	pixelma	lo :)
22:10:44		Quit domonoky1 (Read error: Connection reset by peer)
22:13:57		Join soap_ [0] (~soap@rockbox/staff/soap)
22:14:37		Quit soap (Disconnected by services)
22:14:40		Nick soap_ is now known as soap (~soap@rockbox/staff/soap)
22:20:18		Join lixus001 [0] (~lixxus@78-86-253-21.zone2.bethere.co.uk)
22:20:30		Part lixus001
22:24:10		Quit benedikt93 (Quit: Bye ;))
22:28:20		Quit Buschel (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101026210630])
22:30:43		Quit komputes (Quit: I haven't slept for ten days, because that would be too long.)
22:31:36		Join T44 [0] (~Topy44@f048172081.adsl.alicedsl.de)
22:34:46		Quit Topy (Ping timeout: 245 seconds)
22:40:49	***	Saving seen data "./dancer.seen"
22:45:03		Quit riotz (Ping timeout: 260 seconds)
22:46:42		Join riotz [0] (riot33r@gateway/shell/shellium.org/x-fhfxosuvkshmtgts)
22:47:16		Quit Quelsaruk (Quit: Konversation terminated!)
22:52:39		Quit riotz (Changing host)
22:52:39		Join riotz [0] (riot33r@unaffiliated/riotz)
22:52:39		Quit riotz (Changing host)
22:52:39		Join riotz [0] (riot33r@gateway/shell/shellium.org/x-fhfxosuvkshmtgts)
22:56:06		Quit TheLemonMan (Quit: Help me, i got shot! DIES)
22:58:20	freddyb	I'm confused. I have a Fuze V1 that garbles vorbis files. My Fuze V2 does fine and my iPod works. Any ideas?
22:59:31	pixelma	maybe you suffer from that bug that only affects some AMS devices. Is this with an SVN build and could you compare to the release?
23:00
23:00:50	freddyb	I'm using current SVN. I'll try the release and see if it makes any difference. Is there an FS# ?
23:01:24	pixelma	http://www.rockbox.org/tracker/task/11608
23:08:38		Quit MItaly (Quit: ChatZilla 0.9.86 [Firefox 3.6.12/20101027124735])
23:10:22	freddyb	Thanks
23:24:10		Join LambdaCalculus37 [0] (~rmenes@static-71-249-221-104.nycmny.east.verizon.net)
23:24:20		Quit LambdaCalculus37 (Changing host)
23:24:20		Join LambdaCalculus37 [0] (~rmenes@rockbox/staff/LambdaCalculus37)
23:25:01	saratoga	fucking hell i still can't get real producer to make an atrac file
23:25:06	saratoga	why are these people such assholes
23:26:14		Quit riotz (Quit: out!)
23:31:56		Quit panni_ (Read error: Connection reset by peer)
23:36:33	S_a_i_n_t	freddyb: FWIW, convttf (your version of it at least) doesn't respect font output height.
23:36:48	S_a_i_n_t	It seems a random numbers game trying to get a precise height.
23:37:20	S_a_i_n_t	And, if there's a way to add ascent/descent...I would LOVE for you to tell me how.
23:37:25		Join panni_ [0] (hannes@ip-178-203-81-220.unitymediagroup.de)
23:46:25		Join JdGord [0] (~jd@pa58-109-129-213.pa.nsw.optusnet.com.au)
23:53:31		Quit saratoga (Quit: Page closed)
23:54:01	freddyb	Saint: convttf -p [the height you want] font.ttf
23:54:12	S_a_i_n_t	yeah, broken.
23:54:46		Quit jgarvey (Quit: Leaving)
23:55:03		Join Guinness [0] (Slayer@c-68-55-111-159.hsd1.va.comcast.net)
23:55:21	S_a_i_n_t	convttf -p 16 Helvetica.ttf produces a 22px font :/
23:55:34	S_a_i_n_t	-5 produces a 19px font.
23:55:39	S_a_i_n_t	it's a crazy game ;)
23:55:53	S_a_i_n_t	s/5/15/
23:57:18		Quit Kupop (Ping timeout: 276 seconds)
23:58:47	S_a_i_n_t	freddyb: I'd also how to know how to add X lines of blank pixels in ascent/descent to a font using convttf....convbdf does it with ease, perhaps it's silly to aliken the two, but it should be possible.
23:59:01	S_a_i_n_t	I just can't seem to find how to do it :/

Previous day | Next day