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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2013-04-16

00:02:48 Join fyrestorm [0] (~nnscript@24.90.158.3)
00:03:41***Saving seen data "./dancer.seen"
00:04:36 Quit DexterLB (Read error: Connection reset by peer)
00:05:24 Quit pamaury (Ping timeout: 246 seconds)
00:06:05 Quit ps-auxw (Ping timeout: 264 seconds)
00:07:37 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
00:09:55 Join DexterLB [0] (~dex@95.43.98.183)
00:12:27 Quit shamus (Read error: Connection reset by peer)
00:13:27 Join shamus [0] (~shmaus@ip-206-192-195-49.marylandheights.ip.cablemo.net)
00:20:33 Quit Gareth (Ping timeout: 240 seconds)
00:21:17 Join Gareth [0] (~gareth@2607:ff38:2:83::3)
00:27:11 Join jhMikeS [0] (~jethead71@50.4.247.132)
00:27:11 Quit jhMikeS (Changing host)
00:27:11 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
00:27:33 Quit ender` (Quit: On the other hand, you have different fingers.)
00:29:35 Quit mikroflops (Ping timeout: 264 seconds)
00:31:19 Join mikroflops [0] (~yogurt@h-34-210.a238.priv.bahnhof.se)
00:31:36 Quit lebellium (Quit: ChatZilla 0.9.90 [Firefox 21.0/20130408165307])
00:44:50 Quit bertrik (Remote host closed the connection)
00:47:59 Quit Wardo (Read error: Connection reset by peer)
01:00
01:00:25 Quit Strife89 (Quit: Vamoose!)
01:18:45 Quit mrtux (Ping timeout: 245 seconds)
01:19:54 Join mrtux [0] (~mrtux@unaffiliated/mrtux)
01:20:21 Join ur725 [0] (~ur725@24-180-66-46.dhcp.bycy.mi.charter.com)
01:28:59 Quit bluebrother (Ping timeout: 264 seconds)
01:29:32 Quit fs-bluebot (Ping timeout: 256 seconds)
01:38:17 Join fs-bluebot [0] (~fs-bluebo@f053154205.adsl.alicedsl.de)
01:40:42 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
01:42:17 Join [Saint] [0] (~saint@rockbox/user/saint)
02:00
02:03:20 Quit krabador (Remote host closed the connection)
02:03:43***Saving seen data "./dancer.seen"
02:06:13 Join saratoga [0] (123e1f63@gateway/web/freenode/ip.18.62.31.99)
02:07:09saratogaignoring that the guy in the forums is kind of an idiot, we should probably try to do something about the EQ settings
02:07:53saratogalike I originally said I wanted before accepting the 10 band patch, we should change the number of bands based on CPU
02:09:08 Quit DexterLB (Read error: Connection reset by peer)
02:09:15saratogaI think most users do not realize that the Q settings exist and instead assume it works like a simple non-parametric EQ
02:09:18saratogamost professional equipment uses 3 or 5 bands, so we should probably limit it to 5 bands except on devices where there really is no hit from needlessly enabling extra bands
02:09:45[Saint]I've tried to replicate this on a Nano1G - which is basically the same hardware, and I put this patch through a reasonable amount of punishment before I submitted it, and I never saw this.
02:09:56[Saint]I guess the difference is the disk - dunno.
02:09:58saratogathis will coerce users into thinking about what they are doing and maybe even looking at the manual rather then ruining playback by overloading the system
02:10:26saratogalow resolution display
02:10:34saratogai bet if you battery benched it you'd see a huge difference on the nano 1G
02:10:36[Saint]AH - hum.
02:10:37jhMikeS10 bands is for people who like adjust their EQ rather than enjoying things
02:11:18saratogathere is basically no reason to ever use 10 bands though
02:11:43saratoga8 basically gets you one band per octave over the range of frequencies you'd care about on a portable
02:11:54scorche|sh5 bands ought to be enough for anybody!
02:12:16saratogawell looking online, a lot of albums are apparently mastered with 5 bands
02:12:36saratogaso . . .
02:12:52jhMikeSc'mon, those guys know nothing about this stuff :-)
02:13:06saratogahttp://www.sweetwater.com/store/detail/GML8200
02:13:59saratogahttp://www.avid.com/US/products/EQ-III
02:14:08 Join DexterLB [0] (~dex@95.43.98.183)
02:14:29saratogai assume for that much money they can afford to add as many knobs as they want and there is just no sense i having more
02:14:51[Saint]What we *could* do, is use 5-band presets with 5 zeroed bands, or have both 5 and 10 band presets.
02:15:23saratogado the presets actually use more then 2 or 3 bands??
02:15:38[Saint]than, and yes.
02:16:08saratogaughhh
02:16:25saratogahttp://git.rockbox.org/?p=rockbox.git;a=blob;f=lib/rbcodec/dsp/eqs/Bass.cfg;h=162874ef941f40259dd8b512b92faced870ca333;hb=HEAD
02:16:38saratogathis is fucking stupid
02:16:40[Saint]they've used all 5 bands like...forever. The ones in there at the moment are trash, I have a nicer set sitting in gerrit.
02:20:21saratogaso when I said I was ok with this i assumed we were not going to be stupid and try to build a low pass filter out of 10 (!!!!!!!) band pass filters
02:21:39 Quit [Saint] (Read error: Connection reset by peer)
02:22:33 Join [Saint] [0] (~saint@rockbox/user/saint)
02:23:07[Saint]no idea what happened there...
02:24:22[Saint]but - yeah, I quite dislike the state of the presets at the moment, and I've been waiting for someone to push fire on g#401
02:24:25fs-bluebot3Gerrit review #401 at http://gerrit.rockbox.org/r/401 : 3EQ Presets by Hayden Pearce (changes/01/401/4)
02:26:37saratogais there a plot of what the low shelf filter actually does?
02:28:10[Saint]That's a flat copy of a certain company that makes things starting with i's presets, nothing more, a lot nicer than what is in HEAD presently - but I guess the failing of mine was not testing it on a large display, disk-based PP target.
02:28:12saratogaunless i'm missing something, why can't we just use the shelf filter to implement bass boost with a single filter?
02:28:27[Saint]I have an iPod 4G - and *that* is fine too...
02:28:38saratogagrayscale though
02:28:39[Saint]I guess the display isn't /quite/ large enough to tip it.
02:28:46[Saint]Color. Its a Photo
02:28:46saratogaApple doesn't have parametric EQs though so what use is it to copy them?
02:31:21saratogahttp://bilder.hifi-forum.de/medium/647470/10db-easy-boost_212461.png
02:31:31[Saint]I know little to nothing of Q values, I just copied their frequency stepping, and gain values. I was told not to piss around with Q.
02:31:34saratogathats the bass boost, which uses just one band
02:31:56saratogaour current preset does something similar using 10x the CPU load
02:32:10saratogafor literally no reason
02:32:41[Saint]It was doing so with 5X the CPU load before now. Since...forever.
02:32:43saratoga[Saint]: Q is the width of the filter, if you can't control the width of the filter you need a lot of bands, if you can control the width of the filter you hardly need any bands
02:32:49[Saint]Importance of review I guess.
02:33:46saratogayou can implement bass boost for instance with a lot of narrow bands to fake one wide band (what we do now in the EQ) or equivalently with one very wide band directly (what we more sensibly do now with the bass boost feature)
02:33:56[Saint]If you want to re-do the presets - by all means, go nuts.
02:34:09saratogai'll probably just delete them until someone comes up with sane ones
02:34:53saratogaat very least the bass and treble ones should use the bass and treble feature in rockbox
02:35:21[Saint]I don't think an EQ preset can do that.
02:35:28saratogaaren't they just config files?
02:35:46[Saint]They used to be. The formatting has changed.
02:35:53saratogahmm thats annoying
02:35:58[Saint]I'm not sure it is still a 'kosher' config file.
02:36:14[Saint]I could be entirely wrong, though.
02:36:22saratogathe layout looks like the regular settings file though
02:36:53[Saint]Ohhhh - that's right, its backwards compatible.
02:37:07[Saint]the ones in head weren't edited since saving them will save them in the new format.
02:37:16[Saint]+anyway
02:37:39[Saint]right, yes - so that should work. I forgot it still read the 'old' style too.
02:38:09[Saint]it may not save the bass/treble values in the new format though...humm.
02:39:27 Quit dokan (Ping timeout: 256 seconds)
02:40:08 Join dokan [0] (~minatani@ac250006.ppp.asahi-net.or.jp)
02:41:46*jhMikeS wonders how to make sure libfirmware doesn't get purged of things codeclib needs
02:42:20jhMikeS...before codeclib is linked
02:43:17jhMikeS*rbcodec
02:45:11 Quit ungali_mobile (Ping timeout: 256 seconds)
02:45:28saratogacan i edit a gerrit comment
02:45:39saratogai just realized i left out a "not"
02:45:52[Saint]nup - its annoying.
02:46:02saratogaugh
02:46:07saratogaeverything about git is terrible
02:47:18derfWords to live by.
02:47:48[Saint]could be worse - cvs.
02:49:37[Saint]saratoga: it seems like you're the man to do the presets, so by all means, if you find the time, I would appreciate it.
02:50:18[Saint]go nuts and nuke the current ones, but it only affects new users I guess.
02:50:40jhMikeSit was "worse" (CVS) at one time :)
02:50:51[Saint]we don;t have a way of deleting depricated files from /.rockbox, this annoys me.
02:51:03[Saint](especially as I think it would be trivial to implement)
02:51:43jhMikeSwe
02:51:53saratogai don't even have the development tools installed these days, so ideally someone else would nuke them
02:51:57saratogaplus i can't figure out how to use git
02:52:17jhMikeS[Saint]: we're all busy implementing things that are trivial to implement so don't have time for other trivial things, so feel free
02:52:58[Saint]jhMikeS: Oh, I know I could do it - but I was pretty much told not to bother as it wouldn't get submitted, so, yeah :)
02:53:14[Saint]that wasn't a Mr. Someone request. I'm quite capable. ;)
02:54:28[Saint]saratoga: I could do it - but ideally someone with direct access should, otherwise it'd need to be approved from review anyway.
02:54:44[Saint]probably just easier for someone with direct access to do it initially.
02:54:44*jhMikeS was just being a Mr. Bastard
02:54:45saratogayou don't have git access?
02:54:50[Saint]No.
02:56:03[Saint]I usually just poke at skin stuff, which is fine sitting in its own branch in my tree and being sent to review. Never really needed direct access.
02:56:42saratogayou should probably have access at this point
03:00
03:03:57 Join telliott [0] (~Tim@68.179.137.177)
03:05:41telliottI finally got Rockbox on my iPod Classic. I installed MediaMonkey portable mode to the rpod so I could directly edit tags.
03:06:21telliottBecaose of sorting issues, I need to remove all "the" prefixes
03:07:15JdGordonwhere do you expect "the the" to show up?
03:07:23 Join ungali_mobile [0] (~yaaic@199.119.234.234)
03:07:31JdGordonremoving parts of artist names is dump
03:07:36JdGordondumb*
03:07:39telliottI don't have any of their mjusic
03:08:28telliottSo is having all artists starting with "the" sorted under T
03:10:49telliottI picked the lesser of 2evils
03:23:34 Quit telliott (Quit: Leaving)
03:23:40soapyou could always, you know, remove "the" from file names but not tags.
03:24:18soapoh, gone.
03:33:42*[Saint] does "Cure, The"
03:33:46[Saint]for example.
03:34:02[Saint]Verve, The etc.
03:34:06[Saint]...easier.
03:35:17 Quit saratoga (Quit: Page closed)
03:36:45jhMikeSThe, The
03:37:06[Saint]yup.
03:37:44[Saint]I also have The The, The
04:00
04:03:47***Saving seen data "./dancer.seen"
04:08:21 Quit DexterLB (Read error: Connection reset by peer)
04:13:23 Join DexterLB [0] (~dex@95.43.98.183)
04:15:17 Join dhrasmus [0] (~dhrasmus@173-23-133-55.client.mchsi.com)
04:20:13 Join ungali_mobile2 [0] (~yaaic@184.64.50.234)
04:22:55 Quit ungali_mobile (Ping timeout: 260 seconds)
04:24:07 Nick ungali_mobile2 is now known as ungali (~yaaic@184.64.50.234)
04:27:29 Nick ungali is now known as Haraya_is_awesom (~yaaic@184.64.50.234)
04:28:38 Nick Haraya_is_awesom is now known as ungali_potato (~yaaic@184.64.50.234)
04:43:58 Quit ungali_potato (Quit: Yaaic - Yet another Android IRC client - http://www.yaaic.org)
04:44:13 Join ungali_mobile [0] (~yaaic@184.64.50.234)
04:53:13 Quit amiconn (Disconnected by services)
04:53:13 Quit pixelma (Disconnected by services)
04:53:14 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:53:14 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:53:16 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:53:19 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
05:00
05:00:55 Quit dhrasmus (Quit: Leaving)
05:07:40 Join zz_TheSphinX^ [0] (~briehl@p5DD450F3.dip.t-dialin.net)
05:08:39 Quit ur725 (Quit: Leaving)
05:11:03 Quit TheSphin_ (Ping timeout: 240 seconds)
05:14:28 Quit ungali_mobile (Ping timeout: 252 seconds)
05:23:02 Quit DexterLB (Read error: Connection reset by peer)
05:28:30 Join DexterLB [0] (~dex@95.43.98.183)
05:28:38 Join Rower [0] (husvagn@v-413-alfarv-177.bitnet.nu)
05:46:11 Quit DexterLB (Read error: Connection reset by peer)
05:48:23 Join ungali_mobile [0] (~yaaic@199-7-158-53.eng.wind.ca)
05:48:55 Quit ungali_mobile (Client Quit)
05:49:05 Join ungali_mobile [0] (~yaaic@199-7-158-53.eng.wind.ca)
05:51:33 Join DexterLB [0] (~dex@95.43.98.183)
05:52:06 Quit [7] (Disconnected by services)
05:52:15 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
06:00
06:02:00 Join dfkt [0] (dfkt@unaffiliated/dfkt)
06:03:17 Quit dfkt_ (Ping timeout: 248 seconds)
06:03:49***Saving seen data "./dancer.seen"
06:04:16 Quit froggyman (Ping timeout: 258 seconds)
06:07:34 Quit DexterLB (Read error: Connection reset by peer)
06:12:35 Join DexterLB [0] (~dex@95.43.98.183)
06:15:05 Quit ps-auxw (Ping timeout: 264 seconds)
06:20:58 Join ps-auxw [0] (~arneb@2001:470:c807:0:1532:4e5f:2ad3:4123)
06:21:40 Quit nosa-j (Ping timeout: 245 seconds)
07:00
07:03:45 Join melmothX [0] (~melmoth@unaffiliated/melmothx)
07:08:06 Quit kiwicam (Remote host closed the connection)
07:17:11 Join mortalis [0] (~kvirc@213.33.220.118)
07:22:32 Join akaWolf [0] (~akaWolf@unaffiliated/akawolf)
07:24:00 Join kiwicam [0] (~quassel@101.98.163.139)
07:26:32 Quit DexterLB (Read error: Connection reset by peer)
07:31:43 Join DexterLB [0] (~dex@95.43.98.183)
07:41:45 Join mortalis|2 [0] (~kvirc@213.33.220.118)
07:41:45 Quit mortalis (Read error: Connection reset by peer)
07:47:14 Quit DexterLB (Read error: Connection reset by peer)
07:48:28 Quit jhMikeS (Ping timeout: 246 seconds)
07:48:32kugelare we going to do a devcon this year? I'm surprised this topic didnt come up yet (or at least I didnt see it)
07:50:27 Join bebna [0] (~a.fasold@94.101.33.114)
07:52:16 Join DexterLB [0] (~dex@95.43.98.183)
07:52:16 Quit mortalis|2 (Read error: Connection reset by peer)
07:52:50 Join mortalis|2 [0] (~kvirc@213.33.220.118)
08:00
08:03:50***Saving seen data "./dancer.seen"
08:14:00 Quit DexterLB (Read error: Connection reset by peer)
08:15:28 Join ender` [0] (krneki@foo.eternallybored.org)
08:19:19 Join DexterLB [0] (~dex@95.43.98.183)
08:35:20 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
08:36:28 Quit ungali_mobile (Ping timeout: 256 seconds)
08:36:35 Join kevku [0] (~kevku@2a01:d0:ffff:34a::8:3)
08:36:48bertrikwe don't have USB time sync on the sansa clip zip, or do we? http://www.rockbox.org/tracker/task/12854
08:42:39 Join LinusN [0] (~linus@giant.haxx.se)
08:49:25 Quit DexterLB (Read error: Connection reset by peer)
08:50:11 Join Ward [0] (~Mirandaha@176-120-190-109.dsl.ovh.fr)
08:50:34 Nick Ward is now known as Guest37651 (~Mirandaha@176-120-190-109.dsl.ovh.fr)
08:51:55JdGordonkugel: well its usually june so now is about the time to start discussing
08:52:02*JdGordon votes for sydney :p
08:54:23 Join DexterLB [0] (~dex@95.43.98.183)
08:55:43 Join wodz [0] (~wodz@89-75-41-78.dynamic.chello.pl)
08:56:49wodzkugel: http://www.rockbox.org/irc/log-20130410#22:27:40 :-)
09:00
09:02:57 Join petur [0] (~petur@rockbox/developer/petur)
09:04:17 Join einhirn [0] (~Miranda@bsod.rz.tu-clausthal.de)
09:06:12 Quit shai (Quit: Leaving)
09:11:09mortalis|2wodz: one weird thing I noticed about pca9555 - according to DS i2c address should be in 0x20-0x27 range.
09:11:48mortalis|2why they use 0x40 and how it's possible?
09:11:55 Quit mortalis|2 (Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/)
09:12:04 Join mortalis|2 [0] (~kvirc@213.33.220.118)
09:13:39 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
09:15:08 Nick mortalis|2 is now known as mortalis (~kvirc@213.33.220.118)
09:19:25wodz0x20<<1
09:19:59wodzthe 8th bit sets RW (transfer direction)
09:20:14wodzmortalis: ^
09:21:21 Quit shamus (Read error: Connection reset by peer)
09:21:49 Quit DexterLB (Read error: Connection reset by peer)
09:22:09 Join shamus [0] (~shmaus@ip-206-192-195-49.marylandheights.ip.cablemo.net)
09:22:42 Quit bertrik (Ping timeout: 264 seconds)
09:26:56 Join DexterLB [0] (~dex@95.43.98.183)
09:32:20 Quit Viperfang (Ping timeout: 256 seconds)
09:33:10 Join Viperfang [0] (~Viperfang@x.viperfang.net)
09:39:56 Part swilde ("ERC Version 5.3 (IRC client for Emacs)")
09:42:36 Join nosa-j [0] (~m00k@184.76.254.130)
09:47:35 Quit nosa-j (Ping timeout: 264 seconds)
09:48:11[Saint].
09:48:13[Saint].
09:48:18[Saint]whoops.
09:48:32*[Saint] votes for DevCon Christchurch 2013
09:51:32 Quit DexterLB (Read error: Connection reset by peer)
09:51:59 Join DexterLB [0] (~dex@95.43.98.183)
09:53:23 Quit Viperfang (Ping timeout: 276 seconds)
09:53:53 Join Viperfang [0] (~Viperfang@x.viperfang.net)
09:55:35 Join nosa-j [0] (~m00k@184.76.254.130)
09:58:49 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
10:00
10:03:41 Quit Viperfang (Ping timeout: 264 seconds)
10:03:51***Saving seen data "./dancer.seen"
10:04:19 Join Viperfang [0] (~Viperfang@x.viperfang.net)
10:05:52pamaurywodz: should I send bootloader.bin or bootloader.rkw with rk27load ?
10:06:26mortalisbin
10:08:30 Quit wodz (Quit: Leaving)
10:12:58pamauryany advise on which pins to short ? i'm trying to get CE but no luck so far
10:13:44mortalissamsung flash?
10:14:32pamauryit work \o/ wodz make a great work
10:14:42pamauryI have working screen an ATA -4 error
10:14:51pamaurystandard flash, from hynix
10:15:01pamauryamazing
10:15:05mortaliscool :)
10:18:19pamaurythat's just painful to short pins each time I need to try
10:25:57pamauryI identified the power button (PC1) but no luck for hold and sd detect
10:26:23pamaurydo the GPIO_PxDR read raw value on pin or only if set to input ?
10:39:41pamauryah forgot the famous E and F ports, which are not in the datasheet
10:44:03 Quit pamaury (Remote host closed the connection)
10:48:23 Join pamaury [0] (~quassel@2a01:e35:243f:8460:f4fd:5276:2614:7919)
10:48:24 Quit pamaury (Changing host)
10:48:24 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
10:52:15mortalispamaury: you may need to configure gpio pullup/pulldown. SCU_GPIOUPCON register
11:00
11:04:27 Quit kaputnik_ (Ping timeout: 246 seconds)
11:11:11 Quit DexterLB (Read error: Connection reset by peer)
11:16:37 Join DexterLB [0] (~dex@95.43.98.183)
11:32:15 Quit DexterLB (Read error: Connection reset by peer)
11:37:39 Join DexterLB [0] (~dex@95.43.98.183)
11:43:13 Quit pamaury (Read error: Connection reset by peer)
11:49:05 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
11:57:23 Quit DexterLB (Read error: Connection reset by peer)
12:00
12:00:28 Join mirog [0] (~d971ec2b@www.haxx.se)
12:02:12 Join DexterLB [0] (~dex@95.43.98.183)
12:03:53***Saving seen data "./dancer.seen"
12:05:07 Join krabador [0] (~krabador@unaffiliated/krabador)
12:05:40 Join wodz [0] (~wodz@iwl138.internetdsl.tpnet.pl)
12:05:46wodzpamaury: great!
12:06:03wodzpamaury: shorting together bus lines is the most reliable
12:06:31pamauryI didn't manage to find hold and sd detect, I printed all GPIO_PxDR but only power button changes something
12:06:47pamauryand no luck with gpio pull up/down
12:07:01mirogH10 w/240GB in one partition - first conclusions: DAP is now less responsive - that means there is significant lag in certain situations (mainly when forwarding, reverse-ing, or when moving to nad fro in database (from album artist -> albums -> songs and bacward).
12:07:10pamauryalso there is a weird bug with the screen: the leftmost columns are garbage
12:08:08mirogNow I have 2 H10 - one w/80GB and one w/240GB. The first has no alg at all or the lag is nonobstructive. Database is the same on both.
12:09:53mirogWhat do you think - can it be something with large HDD access? Wouldn't it be better to have 2 partitions (both < 128GB, i.e. within LDA28 address space)?
12:11:00wodzpamaury: what about xpt?
12:11:09pamaurydidn't try it yet
12:12:08 Quit DexterLB (Read error: Connection reset by peer)
12:13:33wodzpamaury: About the screen, is it with lcd_update_rect() or lcd_update()? try to fake lcd_update_rect() by simply calling lcd_update()
12:13:41pamaurylcd_update()
12:16:36wodzpamaury: try adding http://pastie.org/7598121 at the end of lcd_display_init()
12:17:10pamaurywodz: wrong pastie address
12:17:13 Join DexterLB [0] (~dex@95.43.98.183)
12:18:05wodzhttp://pastie.org/7598143
12:18:09pixelmamirog: are all the settings the same, mainly dircache and maybe database in RAM?
12:20:49wodzpamaury: btw. which stage1 bin do you use?
12:20:57pamaurythe one you send me
12:21:13pamauryok i'll try this after lunch
12:21:29wodzCould you try the one from current HEAD?
12:21:53mirogpixelma: 99,9% yes, but can't check now 'cause only one is by the hand - bigger of the two.
12:22:16pamauryno problem, i'll tell you when I try
12:22:43mirogpixelma: And how can I eventually change mentioned settings? And what influence would it have on performance?
12:23:56 Quit mortalis (Read error: Connection reset by peer)
12:24:24 Join mortalis [0] (~kvirc@213.33.220.118)
12:24:38 Quit Guest37651 (Quit: Blarglarg)
12:27:13mirogpixelma: Oooops, my fault - dbasein RAM was off, don't know why, I have copied config from 80GB to 240GB device. I turned it on.
12:27:44mirogNow it is as responsive in database browsing as 80GB one.
12:28:59mirogBut the lag when forwarding to next song or when rewinding to previous one is still considerable (sometimes none, sometimes 1-2 sec.)
12:30:46wodzsounds like spinnup time
12:36:11mirogwodz: rockbox reports 1520ms spinnup time and it should onlyinfluence the first keypress (spindown is 5s)
12:37:47wodz1520ms is ~1.5s so if the file to be played is not buffered you will be hit by such lag.
12:39:25mirogbut if I advance to the next file one by one - there should be no lag
12:39:45soapscorche, the dirty secret there is that if you survey with old-school techniques earth curvature is "calculated" automatically.
12:41:06[Saint]soap: channel?
12:41:08wodzmirog: why?
12:41:12soapSince the spirit bubble levels you use to plump up your equipment are always pointing you towards the center of mass, assuming consistent Earth density (we'll ignore that) you always look tangent to the curve.
12:41:14soapoops
12:41:40[Saint]np - just letting you know.
12:42:47wodzmirog: you start playback, so buffering system grabs as much as it can from current playlist (be it actual playlist, dynamic playlist or something), then you skip. The lag should not happen as long as you jump to the already cached file.
12:51:22mirogBuffer is 26MB so if I start - say 15MB file system should buffer this file and the next one (or part of) - am I right?
12:52:11wodzI guess we buffer whole files but I am not an expert
12:53:52soappartial files must get buffered. I have plenty of tracks larger than the buffer.
12:54:28wodzFor the test disable spindown at all. If that 'fix' your problem then there is nothing you can do actually.
12:54:32wodzsoap: right
12:59:16mirogI can't disable spindown - can make it 3s or max 254s
13:00
13:00:01mirognow it's 10s and it is quite long time imho
13:00:52wodzI am not saying to disable it for regular use but just for test.
13:01:36mirogstrange - i noticed lag in files from one album (15 - 30MB each) bou files from another w/same size (in average) advance w/o lag.
13:01:54mirogmust make more testing
13:02:29mirogwodz: i understand what you mean - but how to disable spindown?
13:05:21wodzset it to 254 and test skipping while disk is still spinning
13:10:09 Quit petur (Quit: *plop*)
13:10:27wodzmirog: do the problematic files have embedded album art?
13:12:12 Quit DexterLB (Read error: Connection reset by peer)
13:12:54mirogall files have
13:14:05wodzWhich size? We had lots of reports about lags which turns to be caused by massive embedded art
13:14:59mirogI just checked - this album has laaarge size embedded arts!!! Bingo!!!
13:15:40wodzproblem solved :P
13:16:22*wodz can't understand the manner of embedding huge art in tiny files
13:17:19 Join DexterLB [0] (~dex@95.43.98.183)
13:17:44mirogiwodz: it was my fault - i simply took wrong picture :)
13:18:10mirogmust fix it and shoul be OK
13:24:53 Join Wardo [0] (~Mirandaha@176-120-190-109.dsl.ovh.fr)
13:29:19mirogwodz: much better now :) it was a coincidence in this issue: big embedded art and large files itself - several of them exceeds buffer significantly. So it seems the problem was complex.
13:30:45 Quit pamaury (Ping timeout: 246 seconds)
13:35:03wodzI am glad I could help
13:42:52 Quit DexterLB (Read error: Connection reset by peer)
13:47:52 Join DexterLB [0] (~dex@95.43.98.183)
13:55:30 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
13:56:15 Nick bebna is now known as aaaaaaaaaaaaaaaa (~a.fasold@94.101.33.114)
13:58:05 Nick aaaaaaaaaaaaaaaa is now known as PonyPonyPonyPony (~a.fasold@94.101.33.114)
14:00
14:01:09 Quit DexterLB (Read error: Connection reset by peer)
14:02:11 Nick PonyPonyPonyPony is now known as PonyPony (~a.fasold@94.101.33.114)
14:02:24 Nick PonyPony is now known as bebna (~a.fasold@94.101.33.114)
14:02:58 Nick bebna is now known as y007glng7 (~a.fasold@94.101.33.114)
14:03:56***Saving seen data "./dancer.seen"
14:04:07 Nick y007glng7 is now known as Gregor (~a.fasold@94.101.33.114)
14:04:15 Nick Gregor is now known as bebna (~a.fasold@94.101.33.114)
14:05:02 Nick bebna is now known as blazamore (~a.fasold@94.101.33.114)
14:05:43 Join stoffel [0] (~quassel@pD9E4318E.dip.t-dialin.net)
14:05:43 Quit mirog (Quit: CGI:IRC (EOF))
14:06:24 Join DexterLB [0] (~dex@95.43.98.183)
14:06:43 Nick blazamore is now known as bebna (~a.fasold@94.101.33.114)
14:27:39 Join amayer [0] (~amayer@mail.weberadvertising.com)
14:52:42 Quit DexterLB (Read error: Connection reset by peer)
14:57:59 Join DexterLB [0] (~dex@95.43.98.183)
15:00
15:00:31 Join kaitsu1 [0] (~bubba@a91-152-170-133.elisa-laajakaista.fi)
15:19:32 Quit wodz (Quit: Leaving)
15:30:54 Quit DexterLB (Read error: Connection reset by peer)
15:36:03 Join DexterLB [0] (~dex@95.43.98.183)
15:41:43 Join froggyman [0] (~me@unaffiliated/froggyman)
15:50:11 Quit mortalis (Remote host closed the connection)
16:00
16:03:12 Nick bebna is now known as PonyPonyPonyPony (~a.fasold@94.101.33.114)
16:04:00***Saving seen data "./dancer.seen"
16:04:09 Quit kevku (Ping timeout: 245 seconds)
16:05:56 Quit Zagor (Quit: Clint excited)
16:21:25 Join jhMikeS [0] (~jethead71@50.4.247.132)
16:21:26 Quit jhMikeS (Changing host)
16:21:26 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
16:31:20 Part LinusN
16:57:32 Join Zagor [0] (~bjst@46.35.227.87.static.tab.siw.siwnet.net)
16:57:32 Quit Zagor (Changing host)
16:57:32 Join Zagor [242] (~bjst@rockbox/developer/Zagor)
16:59:24 Join kevku [0] (~kevku@2001:470:27:773:0:feed:c0f:fee)
17:00
17:00:57 Join Provel_ [0] (~Provel@75-132-18-44.dhcp.stls.mo.charter.com)
17:01:42 Quit Provel (Ping timeout: 256 seconds)
17:22:09 Quit PonyPonyPonyPony (Quit: Leaving.)
17:52:01 Quit DexterLB (Quit: So long and thanks for all the fish)
17:54:35 Quit Zagor (Quit: Clint excited)
17:56:15 Quit stoffel (Remote host closed the connection)
18:00
18:03:59 Join gshrikant [0] (~gshrikant@27.251.145.36)
18:04:04***Saving seen data "./dancer.seen"
18:07:12 Join ur725 [0] (~ur725@97-86-82-70.static.aldl.mi.charter.com)
18:08:12 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier. http://miranda-im.org)
18:15:00 Join mortalis [0] (~mortalis@77.108.98.176)
18:15:26 Join lebellium [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
18:24:05 Quit gshrikant (Quit: Leaving)
18:37:20mortaliswodz: seems your i2c fix wasn't correct. MA9 doesn't work with this fix (frequent freezes, wrong charging state). The same code with not "fixed" i2c driver works (with pops in sound but it at least it usable).
18:37:29 Join Strife89 [0] (~Strife89@2602:306:250d:f4d9:225:d3ff:fed6:15a)
18:40:39 Join dhrasmus [0] (~dhrasmus@71.32.176.214)
18:41:04 Join wodz [0] (~wodz@89-75-41-78.dynamic.chello.pl)
18:43:01wodzmortalis: AFAIK pca9555 i2c transfers are made in key reading routine. This is in tick taks which in turn is called in irq. This is asking for troubles.
18:43:16wodzwe had similar issue on ipod classic
18:47:16 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
18:49:47 Join DexterLB [0] (~dex@90-154-140-88.btc-net.bg)
18:54:59mortaliswodz: what makes you think so. I don't see anything like that in pca9555 ds.
18:56:10wodzits not in pca9555 ds its where i2c transfers are made in our code
18:57:04bertrikone thing that would be nice, would be to have a unified i2c interface
18:57:10bertrik(just thinking out loud)
18:58:25wodzbertrik: I (almost) did it once upon a time but I was unable to test it on all flavors of targets and lost interest
18:59:11wodzmortalis: button_read_device() is called from tick task which is called in irq context
18:59:41 Quit zoktar (Quit: -)
18:59:53bertrikpity, IIRC, we now have things like #ifdef TARGET_X #define I2C_ADDR 0x10 #else #define I2C_ADDR 0x20 #endif, just because the i2c drivers have a different idea about the 7-bit i2c address format
19:00
19:00:41wodzmortalis: if you make (slow) i2c transfers there, chances are that weird things will happen since the transfers can overlap
19:01:13mortalisok, now i get it
19:02:40mortalisi misunderstood meaning of word 'key' in your very first sentence
19:05:22 Quit DexterLB (Ping timeout: 252 seconds)
19:11:27mortaliswodz: so how it was solved on ipod?
19:19:57 Join DexterLB [0] (~dex@90-154-140-88.btc-net.bg)
19:22:19 Join zoktar [0] (~zoktar@unaffiliated/zoktar)
19:29:12mortalisseems it was g194
19:29:15fs-bluebot3Gerrit review #194 at http://gerrit.rockbox.org/r/194 : 3Classic/6G: hold switch detection using GPIO by Cástor Muñoz (changes/94/194/2)
19:48:30 Quit shamus (Read error: Connection reset by peer)
19:48:39 Join shamus [0] (~shmaus@ip-206-192-195-49.marylandheights.ip.cablemo.net)
19:52:57 Join pretty_function [0] (~sigBART@123.252.212.143)
19:53:51 Quit kaitsu1 (Quit: ( www.nnscript.com :: NoNameScript 4.22 :: www.esnation.com ))
20:00
20:04:07***Saving seen data "./dancer.seen"
20:24:50 Join einhirn [0] (~Miranda@p4FC74AA6.dip0.t-ipconnect.de)
20:27:01 Join kaputnik_ [0] (~kaputnik@port-92-206-1-127.dynamic.qsc.de)
20:29:09 Quit einhirn (Ping timeout: 245 seconds)
20:38:48wodzmortalis: yes gpio was found and i2c way was dropped
20:39:27 Join lebellium_ [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
20:39:34 Quit lebellium (Ping timeout: 245 seconds)
20:39:39 Nick lebellium_ is now known as lebellium (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
20:42:13 Quit ur725 (Quit: Leaving)
20:48:17mortaliswodz: pca9555 have INT pin which activated when one of the port pins changes. This pin connected to GPIO. So create gpio interrupt handler could be created which will read pca9555 ports to some variable. button_read_device will just read this variable.
20:48:50mortalis*So gpio interrupt handler could be created
20:49:37bertrikit's generally not nice to do stuff that busy-waits for something in an interrupt handler
20:50:24mortalisbusy waits don't needed here
20:50:54mortalisjust i2c read
20:51:39 Quit kiwicam (Remote host closed the connection)
20:52:24wodzmortalis: it will work if transfer is sufficiently short
20:54:19bertrikbe careful that you can only call a few kernel functions from interrupt context
20:54:57 Join kiwicam [0] (~quassel@101.98.163.139)
20:55:29wodzmortalis: charging_state() also reads pca9555
20:57:24wodzi2c driver uses mutex, don't know if it is irq safe
21:00
21:00:39wodzmortalis: This pretty much summarizes what options you have http://www.rockbox.org/mail/archive/rockbox-dev-archive-2012-03/0042.shtml
21:03:37 Join ur725 [0] (~ur725@97-86-82-70.static.aldl.mi.charter.com)
21:08:13 Quit dhrasmus (Quit: Leaving)
21:13:38 Quit shamus (Ping timeout: 256 seconds)
21:14:06 Join shamus [0] (~shmaus@ip-206-192-195-49.marylandheights.ip.cablemo.net)
21:14:49 Quit melmothX (Quit: #)
21:19:46 Quit wodz (Ping timeout: 276 seconds)
21:29:00 Quit mortalis (Quit: KVIrc 4.3.1 Aria http://www.kvirc.net/)
21:42:51 Quit akaWolf (Ping timeout: 246 seconds)
21:52:43 Quit pretty_function (Remote host closed the connection)
21:59:43 Join wodz [0] (~wodz@89-75-41-78.dynamic.chello.pl)
22:00
22:04:09***Saving seen data "./dancer.seen"
22:20:42 Join liar [0] (~liar@clnet-p09-185.ikbnet.co.at)
22:29:21 Quit kiwicam (Remote host closed the connection)
22:33:31 Join kiwicam [0] (~quassel@101.98.163.139)
22:43:06 Quit ender` (Quit: A SQL query goes into a bar, walks up to two tables and asks, "Can I join you?")
22:47:13 Quit ur725 (Quit: Leaving)
22:48:35 Join theCreator [0] (~theCreato@192.64.114.115)
22:48:38 Join n1s [0] (~n1s@nl118-168-30.student.uu.se)
22:48:38 Quit n1s (Changing host)
22:48:38 Join n1s [0] (~n1s@rockbox/developer/n1s)
22:50:30 Part theCreator
22:52:42 Join lebellium_ [0] (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
22:54:36 Quit lebellium (Ping timeout: 246 seconds)
22:54:37 Nick lebellium_ is now known as lebellium (~chatzilla@lns-c10k-ld-02-m-212-194-176-149.dsl.sta.abo.bbox.fr)
23:00
23:07:19 Quit amayer (Ping timeout: 258 seconds)
23:12:16 Quit y4n (Quit: AMIGAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAHAHAHAHAAAAAAAAAAAAHAHAAA)
23:14:05 Quit n1s (Quit: Ex-Chat)
23:17:21 Quit pamaury (Ping timeout: 246 seconds)
23:20:39 Join kaputnik__ [0] (~kaputnik@port-92-206-88-214.dynamic.qsc.de)
23:22:29 Quit kevku (Ping timeout: 245 seconds)
23:24:20 Quit kaputnik_ (Ping timeout: 272 seconds)
23:52:55 Quit wodz (Quit: Leaving)
23:56:03 Join ur725 [0] (~ur725@24-180-66-46.dhcp.bycy.mi.charter.com)
23:57:48 Join RiD [0] (~RiD@2.83.8.131)
23:58:41 Quit jhMikeS (Ping timeout: 264 seconds)

Previous day | Next day