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 2014-09-28

00:05:58saratogaUSB patch looks good, testing it i just did a series of USB insert and pulls and it did not have any trouble, so shutting down the PHY looks safe enough
00:06:39lebelliumyou should try it with my theme to make sure :P
00:07:00lebelliumif it works with my theme, you're sure it will work with any other
00:09:17ulmutulDoes the clip zip has an OLED display? This would explain the big theme-related power savings: the more black, the less power.
00:10:03lebelliumyes it has, a [shitty] oled display :)
00:10:25Catelite'The Clip Zip includes all the features of the Clip+ and is based on the same processor as the Clip+ and Fuze v2 but includes a larger 1.1-inch full-color OLED[9] display with support for album art (but called "LCD" in a press release[10]),'
00:10:27Cateliteand lol yes :D
00:10:33Cateliteit does this trippy thing if you shake it up and down
00:10:40Cateliteit looks like the display is warping and twisting
00:12:10Catelite..I still haven't figured out how to lock the clip zip while running rockbox, I am guessing it's a button combination or button hold I haven't tried yet
00:12:38lebelliumIt's written in the user manual ;)
00:13:40 Join Cultist [0] (
00:13:58CateliteI didn't see one for the clip zip, I guess I can use one for another machine? found it anyway!
00:14:32gevaertsit's there...
00:15:56 Join bp0 [0] (~bp@unaffiliated/bp0)
00:16:10gevaertsI wonder where you looked if you didn't see it
00:18:05 Join xorly [0] (
00:18:43ulmutulApropos LCD: there are two settings: Backlight and sleep. Some targets have a transflective LCD display. What's the best way: power off backlight, but keep the LCD enabled after backlight timeout, and switch off after LCD-sleep-time?
00:22:20 Quit xorly (Ping timeout: 250 seconds)
00:36:31[Franklin]Would you mind commiting a mega-patch of rockblox?
00:36:46[Franklin](whitespace, ghost pieces, hold, keymap changes all in one)
00:36:57[Franklin]or prefer split patches?
00:40:05ulmutulI'd prefer to split
00:40:41*[Franklin] will probably lump multiple things together
00:40:45[Franklin](but not one patch)
00:40:49[Franklin]probably 3 or so
00:45:19[Franklin]Wow... that was easier than I'd expected
00:45:28[Franklin]the code is surprisingly flexible
00:45:58[Franklin]hmm... but the colors are weird
00:46:16[Franklin](as in they look just like "real" pieces)
00:47:35[Franklin]... and yes, I'll test on GS and monochrome :P
00:49:23 Quit ender1 (Quit: #define sizeof(x) ((rand() % 100 == 42) ? sizeof(x)-1 : sizeof(x)))
00:49:49*[Franklin] wonders what color(s) the ghost pieces ought to be
00:49:59*[Franklin] summons foolsh for suggestions
01:05:54 Join ulmutul_ [0] (
01:08:37 Quit ulmutul (Ping timeout: 246 seconds)
01:08:50 Nick ulmutul_ is now known as ulmutul (
01:13:00 Quit foolsh (Quit: foolsh)
01:14:36 Quit ulmutul (Quit: ChatZilla [Firefox 32.0.3/20140923175406])
01:15:57*[Franklin] can't seem to find some good colors for the ghost pieces
01:16:28ZincAlloyhow about a semiopaque white?
01:17:39[Franklin]yeah, that looks too much like the T tetronimo
01:17:55*[Franklin] considers changing the whole color scheme
01:18:53[Franklin] <−− scroll down
01:19:34ZincAlloyor make them the same color as the tetronimos, but translucent?
01:19:41[Franklin]yeah, I'm trying that
01:20:09[Franklin]perhaps 1/4 the RGB values?
01:20:59*[Franklin] likes that idea
01:21:08[Franklin]powers of two=SUPER EASY division
01:23:40[Franklin]hmm... perhaps a bit too dark?
01:23:40*[Franklin] will try 1/2 then
01:29:29[Franklin]or... I could calculate the values at start time, or even hard-code them
01:29:36[Franklin](so I could do 1/3, etc)
01:30:09CateliteO a, kisy
01:30:14CateliteWrong hand position :o
01:30:37***Saving seen data "./dancer.seen"
01:31:55CateliteI have just discovered how to do replay gain o_o
01:32:00CateliteBut I have no idea what to use for fall-back gain
01:32:24Cateliteguessing it doesn't matter if I calculate it for all of my music files anyway
01:35:40[Franklin]hmm... now to optimize this
01:38:07 Quit Aldem2 (Quit: Leaving)
01:44:41saratogai didn't think we had a fall-back gain setting
01:50:20*[Franklin] wonders if ghost pieces should be disabled on low-depth targets
01:50:25[Franklin](or use greylib)
01:50:39[Franklin]because it's impossible to distiguish from "real" pieces
02:03:26[Franklin]anyone with a 1-bit LCD device, please test g990 to see if it's playable
02:03:29fs-bluebotGerrit review #990 at : New Rockblox features: by Franklin Wei
02:03:42[Franklin]seems a bit unplayable in the sim
02:03:53[Franklin](with ghost piece looking exactly like every other piece)
02:08:33 Quit lebellium (Quit: ChatZilla [Firefox 33.0/20140923222114])
02:09:59 Quit varogami (Ping timeout: 244 seconds)
02:22:25 Quit ZincAlloy (Quit: Leaving.)
02:23:16 Quit pamaury_ (Ping timeout: 272 seconds)
02:26:17 Join Catelite` [0] (
02:28:11 Quit Catelite (Ping timeout: 245 seconds)
03:00:01 Quit AlexP (Read error: Connection reset by peer)
03:00:37 Quit stickyb1t (Ping timeout: 246 seconds)
03:01:38 Quit Provel (Ping timeout: 260 seconds)
03:02:13 Join Provel [0] (
03:03:07 Join stickyb1t [0] (
03:10:58 Quit bertrik (Quit: Lost terminal)
03:21:28 Quit steffengy (Disconnected by services)
03:21:28 Join steffengy1 [0] (~quassel@
03:28:18 Join ygrek [0] (~user@
03:30:38***Saving seen data "./dancer.seen"
03:33:54 Join kev2014 [0] (
03:34:34 Quit stickyb1t (Quit: Konversation terminated!)
03:35:13 Quit kev2014 (Quit: Nettalk6 -
04:04:52 Quit [Franklin] (Quit: Lost terminal)
04:16:56 Join CaptainKewl [0] (
04:27:29 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:27:29 Quit amiconn (Disconnected by services)
04:27:29 Quit pixelma (Disconnected by services)
04:27:30 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:27:33 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:27:33 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:42:31Catelite`...actually I am curious suddenly
04:42:42Catelite`if anyone's been able to figure whether OGG takes more batter power to decode than MP3
04:43:21Catelite`I could never tell a difference on my sansa clip zip, the machine seemed to die after the same period of time regardless of playing 160 VBR mp3 or 96 VBR ogg
04:43:23fs-bluebotGerrit review #160 at : Fuze+: All games plugins keymaps by Jean-Louis Biasini
04:46:16 Quit ygrek (Ping timeout: 272 seconds)
04:54:02 Join stickyb1t [0] (
05:02:59 Join Aldem [0] (~Aldem@unaffiliated/aldem)
05:20:35saratogaCatelite`: depends on the device
05:20:55Catelite`additionally, trying to figure out replay gain :o which is awesome
05:21:19Catelite`most everything sounds great besides a few oddball songs, like a floppy drive orchestra which ended up with +69 db for some reason
05:21:33Catelite`which is beyond loud to the point of being nothing but clipping
05:22:06Catelite`and most of my chiptunes, which are getting positive values like +10db in spite of already being quite loud
05:22:35saratogawhat software did you process them with?
05:22:37Catelite`..not sure if this is winamp being derp
05:22:40saratogasounds like its buggy
05:22:42Catelite`winamp \o
05:22:57saratogaon most devices Vorbis and MP3 are quite close
05:23:04saratogai'd try foobar, its the gold standard
05:23:31Catelite`doing that then
05:23:33saratogacould make vorbis quite a bit faster still, but its fast enough that i'm not motivated to
05:23:37Catelite`winamp is showing its age too much
05:24:01 Quit [7] (Disconnected by services)
05:24:19 Join TheSeven [0] (~quassel@rockbox/developer/TheSeven)
05:24:25Catelite`yeah I mostly jumped on OGG because it fits into a much smaller space and is, relatively, entirely transparent to MP3 of much higher bitrate
05:30:39***Saving seen data "./dancer.seen"
05:37:44 Quit stickyb1t (Quit: Konversation terminated!)
05:57:45Catelite`foobar, multi-threaded replay gain scanning <3
06:05:11Catelite`saratoga: Where did your IRC nick come from o.o
06:05:24saratogai don't remember
06:05:29saratogaprobably the city
06:05:30Catelite`there's a ship with that name in Achron
06:10:05Catelite`ahhh yeah
06:10:17Catelite`foobar's replay gain, immediately more competent-seeming
06:10:24Catelite`didn't assign any obviously errored values to gain
06:43:59Strife89Is there a quick way to check which cross-compilers are ready?
06:45:32Strife89I'm trying to build all five (using, but it has aborted on me twice - the first time due to a missing dependency, and the second time due to being unable to create a directory (because it already exists).
06:48:00 Quit burgobianco (Remote host closed the connection)
06:51:12 Join burgobianco [0] (
06:57:20Strife89Also, I'm seeing some strange datestamps for code changes.
07:01:50 Join dfkt_ [0] (dfkt@unaffiliated/dfkt)
07:02:11 Quit dfkt (Disconnected by services)
07:02:14 Nick dfkt_ is now known as dfkt (dfkt@unaffiliated/dfkt)
07:03:58 Join davidfg4 [0] (david@pdpc/supporter/active/davidfg4)
07:13:34[Saint]Strife89: its using the date of submission.
07:14:06[Saint]The gerrit tasks kinda warp the recent dates a bit.
07:15:09[Saint]Also...honestly...just don't bother building for arm-app (or whatever the ypR* toolchain is called - its fundamentally fucking broken)
07:15:40[Saint]Let a port maintainer or someone that even remotely cares about the target sort it out.
07:16:56[Saint]If a 64 bit toolchain archive is of any use to you, the logs here will likely reveal me distributing said archive to pamaury some time prior.
07:30:43***Saving seen data "./dancer.seen"
07:54:27 Nick SuperBrainAK is now known as DormantBrain (~andy@2001:470:8:a61::5f92:59a1)
08:01:04 Join AlexP [0] (~alex@rockbox/staff/AlexP)
08:42:09 Join ygrek [0] (~user@
09:03:36 Quit Catelite` (Ping timeout: 245 seconds)
09:30:46***Saving seen data "./dancer.seen"
09:34:06 Join nk2032 [0] (
09:45:29 Quit CaptainKewl (Quit: ( :: NoNameScript 4.22 :: ))
10:00:15 Join ender` [0] (
10:37:31 Join xorly [0] (
10:38:20 Join petur [0] (~petur@rockbox/developer/petur)
10:47:16 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
11:30:50***Saving seen data "./dancer.seen"
11:45:33 Quit xorly (Ping timeout: 272 seconds)
11:48:01 Join lebellium [0] (
11:55:33 Join kugel [0] (~kugel@
11:55:33 Quit kugel (Changing host)
11:55:33 Join kugel [0] (~kugel@rockbox/developer/kugel)
12:03:11 Join rela [0] (~x@pdpc/supporter/active/rela)
12:06:30lebelliumkugel: ping
12:15:11 Quit ygrek (Ping timeout: 260 seconds)
12:20:39 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
12:21:02 Quit pamaury (Remote host closed the connection)
12:26:06 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
12:27:56 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
12:51:03 Join ZincAlloy [0] (
12:52:40kugellebellium: pong
12:53:40lebelliumkugel: r#566 broke YP-R0 build
12:53:48lebellium g#566
12:53:51fs-bluebotGerrit review #566 at : Rewrite filesystem code (WIP) by Michael Sevakis
12:54:02lebelliumIt's been quite a long time now :S
12:54:16kugelin what way?
12:54:23kugelremember I don't have a yp-r0 anymore
12:55:04lebelliumwell in a way depending on the bootloader, either it just crashes when loading rockbox or at best it tells you "no rockbox directory"
12:57:13lebelliumI don't know who is supposed to fix that. I would say the patch's author, even if he doesn't have a R0?
12:59:01 Quit megal0maniac (Ping timeout: 245 seconds)
13:01:48 Join megal0maniac [0] (~megal0man@unaffiliated/megal0maniac)
13:06:30kugellebellium: could you add some printfs into file firmware/target/hosted/filesystem-app.c
13:06:42kugeland then start rockbox with redirecting stdout to a file
13:07:22 Join varogami [0] (
13:08:29kugelwith the official bootloader you do this by creating a user.rc and setting MAINFILE_REDIRECT='> /mnt/media0/rb-stdout.txt'
13:09:01kugelprintfs so that it prints the paths of the files and dirs that are tried to be opened
13:11:36lebelliumwell I'm not sure I use the official RB bootloader on any of my R0
13:11:47lebelliumI'll ask Lorenzo if he can do that
13:18:52 Quit rela (Read error: Connection reset by peer)
13:25:11 Join ikeboy [0] (
13:30:54***Saving seen data "./dancer.seen"
13:31:22 Quit ikeboy (Quit: Leaving)
13:38:45 Quit megal0maniac (Killed ( (Nickname regained by services)))
13:38:48 Join megal0maniac [0] (~megal0man@
14:03:26 Join lorenzo92 [0] (
14:03:26 Quit megal0maniac (Read error: Connection reset by peer)
14:04:16 Join megal0maniac [0] (~megal0man@unaffiliated/megal0maniac)
14:05:49 Quit megal0maniac (Read error: Connection reset by peer)
14:09:17 Join megal0maniac [0] (~megal0man@
14:09:17 Quit megal0maniac (Changing host)
14:09:17 Join megal0maniac [0] (~megal0man@unaffiliated/megal0maniac)
14:35:14lorenzo92kugel: app_open: /dev/fb0
14:35:15lorenzo92 handle_special_dirs: /mnt/mmc/dev/fb0
14:35:29lorenzo92that's why opening device file fails....
14:41:01lorenzo92there is definitely a bug in path substitution, let me see
14:45:16lorenzo92moreover the patch has a shortcoming: i don't see the possibility of defining multiple drives? i.e. {ext_drive_1,path1 ; ext_drive2,path2...}
14:52:12kugellorenzo92: the virtual symlink stuff for the microsd is still in
14:53:48kugeli guess it's prepending ROCKBOX_DIR to all paths which is wrong
14:54:14lorenzo92kugel: i don't get this. Anyway, the bug is perhaps that get_volume_name returns a void string in our case, so strstr tells us that empty stuff is part of the string :/
14:54:53lorenzo92kugel: are you the author of the patch? because I might be wrong :)
14:55:14kugelwhich patch? the one that broke the ypr0 is by jhMikeS
14:55:32lorenzo92yes the filesystem-related one, yes
14:57:26lorenzo92my 2 cents: the patch can be slightly reworked so that a LIST of "virtual" symlinks can be defined. in the instance of ypr0, we have indeed 2 drives that are mounted on separate paths + we also need "raw" access to the whole filesystem
14:58:46lorenzo92jhMikeS: ^^
14:59:29lorenzo92in any case, the problem is that get_volume_name is returning an empty string, which then makes the whole thing useless
15:06:42Strife89[Saint]: I'm just doing them all for a build client. :)
15:09:19lorenzo92jhMikeS: actually, we have 4 partitions on ypr0...and 3 on ypr1
15:13:26 Join xorly [0] (
15:28:05 Join foolsh [0] (
15:30:57***Saving seen data "./dancer.seen"
15:35:57kugellorenzo92: I dont undertand
15:36:27lorenzo92kugel: i think i don't know what the patch fixed/changed actually
15:37:01kugelit just reworked low-level storage code, it mostly concerns native targets
15:37:24lorenzo92I'm just sure the code doesn't work, and at some point a function call is completely useless, always returning empty string
15:37:45lorenzo92which makes in turn the substituition at every call, which is erratic
15:38:07kugelfor the ypr0 it doesnt do much, and even less so because it doesnt have dircache
15:38:29lorenzo92indeed, it harms :)
16:16:56 Quit Aldem (Quit: Leaving)
16:27:55 Join varogami1 [0] (
16:30:41 Join Aldem [0] (~Aldem@unaffiliated/aldem)
16:30:54 Quit varogami (Ping timeout: 272 seconds)
17:10:57 Quit dfkt (Remote host closed the connection)
17:11:15 Quit foolsh (Remote host closed the connection)
17:27:54 Quit lorenzo92 (Ping timeout: 272 seconds)
17:31:01***Saving seen data "./dancer.seen"
17:44:22 Join jhMikeS [0] (
17:44:22 Quit jhMikeS (Changing host)
17:44:22 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
17:47:53 Quit pystar89 (Ping timeout: 246 seconds)
17:48:41 Join pystar89 [0] (
17:51:28 Quit cmhobbs (Disconnected by services)
17:56:39 Join lorenzo92 [0] (
17:57:19 Quit lorenzo92 (Client Quit)
18:06:25 Join ygrek [0] (~user@
18:47:37 Quit megal0maniac (Killed ( (Nickname regained by services)))
18:47:41 Join megal0maniac [0] (
18:58:14 Quit ygrek (Ping timeout: 272 seconds)
19:01:53 Join stickyb1t [0] (
19:09:41 Quit stickyb1t (Remote host closed the connection)
19:10:02 Join kev2014 [0] (
19:10:12 Join stickyb1t [0] (
19:14:36 Join ygrek [0] (~user@
19:15:48 Join flexigon [0] (63e9b3e8@gateway/web/freenode/ip.
19:16:27flexigonwhere do i get the original firmware for sansa fuze v2?
19:17:17 Quit flexigon (Client Quit)
19:19:24 Quit pamaury (Ping timeout: 250 seconds)
19:19:59 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
19:20:48 Part kev2014
19:31:05***Saving seen data "./dancer.seen"
19:36:03saratogathe USB and backlight patches ( g#988 g#989) improve battery life on my Zip by 32%
19:36:08fs-bluebotGerrit review #988 at : Make sure the USB PHY is disabled after use. Patch by Mihail Zenkov who has by Michael Giacomelli
19:36:08fs-bluebotGerrit review #989 at : Don't enable the current sink for the Clip Zip backlight until its actually needed. by Michael Giacomelli
19:37:04saratogai'll test the voltage changes next
19:37:29lebelliumsaratoga: a feedback here
19:37:42lebelliumrather negative :)
19:37:59bertriksaratoga: nice!
19:38:02saratogayeah with the clock that low most things are not going to decode
19:38:05saratogai don't like the clock changes
19:38:16saratogabut the results are interesting to know
19:38:27saratogabertrik: do you remember much about frequency scaling on the AMS devices?
19:38:45bertrikno, not really
19:39:07saratogaif that guy on the forums has really found a set of low clock speed settings that uses a lot less power, that argues that frequency scaling would be worthwhile
19:39:08bertrikincreasing just the CPU clock didn't help much IIRC
19:39:20saratogaas i recall we implemented it and found it actually reduced battery life
19:39:39saratogaperhaps a mistake was made or the settings were not correctly chosen
19:39:52kugeli can recall that
19:39:55 Quit pamaury (Ping timeout: 260 seconds)
19:40:12 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
19:40:20kugeli can't*
19:41:30saratogakugel: by the way, i have some questions about the pitch screen, do you know much about that code?
19:42:06bertriksaratoga: if g#988 g#989 works for you, I'd set get it in
19:42:11fs-bluebotGerrit review #988 at : Make sure the USB PHY is disabled after use. Patch by Mihail Zenkov who has by Michael Giacomelli
19:42:11fs-bluebotGerrit review #989 at : Don't enable the current sink for the Clip Zip backlight until its actually needed. by Michael Giacomelli
19:42:13kugelBack when it was being disabled in svn someone messured 3 or 4 hours reduced runtime (i.e. dynamic boosting gave more runtime)."
19:42:47saratogawith v2 or v1? I seem to remember that for V1, but i thought v2 wasn't in svn?
19:45:06saratogaif i set a git author, do i need an email address for that person or can i just use their name?
19:47:18saratogaby the way, the original Zip backlight patch changes the drive current value to be the same as the other AMS player (previously it was off by 1)
19:47:28saratogai'm assuming this was an error and they should all use the same value?
19:47:36saratogaprobably some tiny difference doesn't matter anyway
19:50:08saratogashort of deleting my branch and starting over, how do i ammend a previous commit message?
19:51:07gevaertssaratoga: ideally, yes, but for the svn to git migration we've used "fake" email addresses, so those should work too
19:51:19gevaertsAnd git commit −−amend
19:51:21bertrikgit commit −−amend IIRC
19:52:31saratogathat will only amend the most recent commit though right?
19:52:35saratogahow do i amend several in a row?
19:53:28 Join cmhobbs [0] (~cmhobbs@fsf/member/cmhobbs)
19:55:29bertrikI guess git commit −−amend HEAD^
19:56:06bertrikhm sorry, doesn't work
20:00:39 Quit ygrek (Quit: ygrek)
20:00:41saratogai guess i can just create a new branch, cherry pick (??????) the gerrit task, ammend it and commit
20:01:44saratogaha no i can't because i accidentally made those changes in master
20:01:52saratogaok i have no idea how to do this
20:04:20 Quit pamaury (Ping timeout: 272 seconds)
20:04:25 Join pamaury_ [0] (~quassel@rockbox/developer/pamaury)
20:05:22gevaertssaratoga: might help
20:05:35 Quit kugel (Ping timeout: 245 seconds)
20:06:05 Join kugel [0] (~kugel@rockbox/developer/kugel)
20:06:19saratogai'm trying to rebase now
20:06:22saratogalet me see if this works
20:07:12saratogaslightly screwy on gerrit but i think ok now
20:09:40saratogai'm going to change the backlight drive setting to be the same as the other players
20:09:46saratogai doubt it matters, but its bothering me
20:11:35saratogaok the final patches are up
20:11:40 Quit bluebrother (Disconnected by services)
20:11:41saratogaanyone want to look at them before I push?
20:11:45 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
20:14:36kugelsaratoga: with git rebase −−interactive you can amend every commit
20:14:51 Quit fs-bluebot (Ping timeout: 272 seconds)
20:15:56saratogakugel: did you see the logs about the pitchscreen?
20:16:09kugelah yea, what's your question?
20:18:04 Join [Franklin] [0] (
20:18:15saratogai'm a little confused how that screen is supposed to work
20:18:38saratogaare the separate screens for pitch and pitch+timestretch intentional?
20:18:57saratogaseems odd to show the first one, I think its redundant with the second? or is there something i miss
20:20:23saratogabertrik: do you know much about the as3543 voltages? grepping the source I don't even see where we set CVDD1
20:20:25 Join fs-bluebot [0] (
20:20:59saratogaMihail says reducing it saves several milliamps, which makes me wonder how it even gets its current value (hardware deefault?)
20:23:35saratoga" <bertrik> I didn't see any explicit setting of the CVDD1 voltage (which I think is the CPU voltage), the CVDD2 voltage is explicity set on the zip"
20:23:41saratogawell ok then
20:24:50 Quit stickyb1t (Quit: Konversation terminated!)
20:25:39saratogathe debug screen doesn't print the value either
20:25:49saratogai wonder what we actually run it at?
20:34:39saratoga <bertrik> in the logs you mentioned reading out voltages, and looking at the AS3543 datasheet there is a way to read out the CPU voltage but its not used in the debug screen
20:34:44saratogadid you ever try that ?
20:35:55saratogai'm looking at the debug screen now
20:36:18saratogaone of the entries that is marked NULL in that code is actually a mux that can be switched between various supply voltages
20:36:50saratogatrying to understand how that code works now
20:41:12*[Franklin] actually tested G#990 on color, gray, and mono :)
20:41:15fs-bluebotGerrit review #990 at : New Rockblox features by Franklin Wei
20:49:35saratogabertrik: are you still around?
20:52:34pamaury_saratoga: the system code of as3525 sets CVDD1 and CVDD2 explicitely
20:52:52pamaury_oh no CVDD2 only sorry
20:53:02saratogai think i understand how to read it
20:53:03 Nick pamaury_ is now known as pamaury (~quassel@rockbox/developer/pamaury)
20:53:06saratogagoing to check what value it has
20:53:38kugelsaratoga: separate in what way?
20:54:00saratogakugel: right now theres two screens, one where you can only change pitch, and one where you can change pitch and speed
20:54:08saratogaseemed confusing to me
20:54:25kugelit's the same screen, just different modes
20:54:39pamaurysaratoga: ascodec_read_pmu(0x17, 1) ?
20:55:03saratogasorry, i should have said mode
20:55:11saratogadoes the first mode need to exist? it seems redundant
20:55:37kugelno, why?
20:55:48saratogacouldn't the second mode do everything?
20:56:01kugelchanging only pitch valid use case
20:56:16saratogayeah but couldn't you do that just as easily from the second screen?
20:56:30kugeli dont think the second mode allows that
20:56:30saratogapamaury: yes, just double checking that its safe to write to that register since it controls bunch of things
20:56:51saratogawait isn't it 0x1c?
20:57:42pamaurydepends on what you want to read I guess
20:57:51saratogai want to access the mux under AS3543_PMU_ENABLE
20:58:07saratogathen read 0x5: CVDD1
20:58:27pamauryah ok, I thought you wanted to read the CVDD1 voltage
20:58:36pamauryanyway reading should be safe
20:58:37saratogathats what i'm trying to do
20:58:40saratogais there an easier way?
20:59:36saratogaon AMSv2 it looks like you have to configure a mux then read a specific channel to access it
21:01:42 Quit y4n (Quit: MOTHER EUROPA CALLING ME!)
21:04:10saratogawoot works
21:04:18saratoga1.190 v
21:04:24saratogain good agreement with the expect 1.2
21:04:56 Join krabador [0] (~krabador_@unaffiliated/krabador)
21:04:59 Join CaptainKewl [0] (
21:07:46fs-bluebotBuild Server message: New build round started. Revision 02414bf, 255 builds, 33 clients.
21:11:46 Quit krabador (Ping timeout: 260 seconds)
21:15:05 Join Nemayogok [0] (~nemayogok@
21:15:33 Quit Zambezi (Ping timeout: 272 seconds)
21:16:12 Nick DormantBrain is now known as SuperBrainAK (~andy@2001:470:8:a61::5f92:59a1)
21:20:02fs-bluebotBuild Server message: Build round completed after 333 seconds.
21:21:13NemayogokHi... anybody knows what is "AIA error: -2" ???
21:22:07bertrikprobably "ATA error: -2", an error related to the flash or had disk
21:22:21 Quit xorly (Read error: Connection reset by peer)
21:22:45Nemayogokbertrik, thanks
21:23:42 Join xorly [0] (
21:31:08***Saving seen data "./dancer.seen"
21:31:23pamaurysaratoga: ah, I thought you want to read the programmed value of CVDD, and not actually measure it
21:31:32saratogabertrik, pamaury: can one of you double check g#991 before I commit it?
21:31:35fs-bluebotGerrit review #991 at : Add various additional AS3525v2 SOC voltages to the debug screen. by Michael Giacomelli
21:33:25pamaurysaratoga: it looks ok
21:34:03saratogai just noticed it kind of glitches the backlight sometimes when reading the extended voltage registers
21:34:07 Join Zambezi [0] (
21:34:25saratogai guess thats ok though since its only a debug screen
21:35:14 Join Strife1989 [0] (~Strife89@2600:1005:b020:18e5:a871:695f:bb2f:ad71)
21:36:00saratogai guess the codec read can preempt the codec write to enable backlight
21:36:03saratogano big deal though
21:36:18fs-bluebotBuild Server message: New build round started. Revision f014a76, 255 builds, 33 clients.
21:37:51*Strife1989 wonders who else has noticed the red on the H120 HTML manual
21:38:19saratogait was recently added to the build table to encourage someone to figure out why it stopped building
21:38:28saratogai looked at it briefly but did not understand enough latex to fix it
21:38:46Strife1989Me neither
21:39:08gevaertsI had a look, and as far as I can figure out it's a bug in one of the things we use
21:39:18gevaertsI don't know enough about it to easily work around that
21:42:30fs-bluebotBuild Server message: Build round completed after 371 seconds.
21:42:35[Franklin]Nemayogok: what target?
21:43:02[Franklin]recently that ATA patch broke stuff, for me at least
21:44:02Nemayogoksdc4/32gb kingston
21:44:15[Franklin](I was referring to ipod6g)
21:44:19 Quit xorly (Read error: Connection reset by peer)
21:44:50 Quit Provel (Ping timeout: 245 seconds)
21:45:34 Join xorly [0] (
21:45:36*[Franklin] hopes saratoga has noticed the red on c/e/m200 and fuze
21:45:43 Join Provel [0] (Provel@
21:46:09Nemayogokmmm hifiman hm-601
21:46:52[Franklin]then no
21:47:29Strife1989saratoga, gevaerts: I only noticed it while checking build stats and was more curious than anything. Sorry.
21:47:59gevaertsNemayogok: which rockbox version are you running?
21:48:31[Franklin]well, worth a try
21:49:58*[Franklin] hates that the build targets aren't searchable
21:50:45[Franklin]you can't run CTRL+F and search
21:50:59gevaertssearch for what?
21:51:03[Franklin]target names
21:51:12*[Franklin] just flips his monitor
21:52:06Nemayogokgevaerts, Version: 515a3c5-140924
21:53:02[Franklin]3 commits ago
21:53:13[Franklin](which shouldn't affect hm-601 at all)
21:56:42 Quit petur (Remote host closed the connection)
21:59:51[Franklin]Nemayogok: so does it panic after the ATA error?
22:00:09Nemayogokshut down
22:00:33Nemayogoknow it says ATA error: -4
22:00:35[Franklin]ahh... so it's unusable?
22:01:04Nemayogoka lot....
22:02:07[Franklin]can you still boot OF?
22:02:39Nemayogokyes but connected to usb
22:03:03[Franklin]try an older RB version?
22:03:11[Franklin]perhaps last stable?
22:03:39Nemayogokyes that's my next step
22:03:40Nemayogokthanks :D
22:06:47 Join Strife89|netbook [0] (~Strife89@2602:306:bce1:8c20:c1c9:7160:8c9e:2740)
22:06:57 Quit Strife1989 (Ping timeout: 272 seconds)
22:10:48 Quit Mir (Quit: Konversation terminated!)
22:13:54 Join foolsh [0] (
22:14:36[Franklin]foolsh: I got ghost pieces working! G#990
22:14:39fs-bluebotGerrit review #990 at : New Rockblox features by Franklin Wei
22:15:47foolshCool, :-) I'll compile later and check it out to be sure, I've been doing a lot of actual "work" lately
22:16:01[Franklin]"work"... hehe
22:16:30*[Franklin] doesn't have the exact colors nailed down yet, but the rest is done
22:17:32[Franklin]for now, just 1/4 the RGB values of the "real" pieces
22:17:37[Franklin](and for gray, 1/2)
22:17:50[Franklin]otherwise, 1/4 is just black
22:21:23*[Franklin] kind of wants a "hold" feature, but most targets don't have enough buttons
22:22:23foolshI don't really know, but would a flashing 1bit ghost piece work for those targets?
22:23:01[Franklin]foolsh: the way it is now, it looks just like a normal piece
22:24:03[Franklin]perhaps a long press center would be hold?
22:25:29Strife89|netbookThat sounds like a good idea.
22:25:45foolshI'm unfamiliar with the hold feature, what is it?
22:25:50Strife89|netbookAs long as there aren't any delays in rotating the piece.
22:26:12 Nick megal0maniac is now known as Guest46550 (
22:26:12 Quit Guest46550 (Killed ( (Nickname regained by services)))
22:26:15 Join megal0maniac [0] (~megal0man@
22:26:34[Franklin]I mean, how do I detect that it's a long press?
22:27:01[Franklin](this doesn't use PLA, so that's like 100 keymaps)
22:27:17[Franklin](and also, I'd need to modify ALL the backdrops to include a space for that
22:28:07foolshHow about press and hold "buttondata=SELECT" holdeds, and releasing the button "buttondata=SELECT_REL" slams the piece down
22:28:19 Quit K1773R (Excess Flood)
22:28:41foolshspelling errors aside, I'm doing two things at once
22:30:20foolshin other words pressing and holding select does the "hold" your talking about...
22:30:23 Join K1773R [0] (~K1773R@unaffiliated/k1773r)
22:30:36[Franklin]so if I wanted to hard drop, it would hold first
22:30:55*[Franklin] is pwning some people in tetris
22:30:58foolshA quik press and release would hard drop
22:31:15foolshA press and hold would "hold"
22:32:22foolshwhat ever it is
22:32:39[Franklin]no... it'd still have that problem
22:41:25 Join krabador [0] (~krabador@unaffiliated/krabador)
22:42:39 Quit Strife89|netbook (Ping timeout: 272 seconds)
22:53:35[Franklin]foolsh: perhaps incrementing a counter each time it's BUTTON_SELECT is recieved?
22:53:50[Franklin]and on BUTTON_SEL_REL , if counter>50, hold, otherwise drop?
22:55:12*[Franklin] likes that idea
23:01:40[Franklin]but that's 11 backdrops
23:08:59[Franklin]which all need modifying to add that hold space
23:09:15[Franklin](which I'm not doing)
23:09:34[Franklin]so, unless someone does it for me (adding the textures), I don't have the time
23:10:44*[Franklin] waits for some modifyed textures to pop up on Gerrit :)
23:14:37saratogai need to do a clip+ battery test
23:15:13 Quit krabador (Quit: Sto andando via)
23:16:30[Franklin]does greylib use a substansial amount of power?
23:16:50 Join krabador [0] (~krabador@unaffiliated/krabador)
23:18:06[Saint]Edit: Testing the USB and backlight changes increased my Zip runtime by 32%.
23:18:08[Saint]...fuck me.
23:19:16 Quit Jinx (Quit: reboot)
23:20:05[Franklin]all AMS devices, too, right?
23:20:31[Saint]re: USB, yes.
23:20:35[Saint]backlight, no.
23:21:03[Saint]Looking at the patch kinda makes me crings.
23:21:13[Saint]And reminds me why I stay the fuck away from those areas.
23:21:43*[Franklin] too
23:21:51[Saint]leaving USB powered regardless if its actually being used or not after the first time its used seems a massive oversight.
23:22:01[Saint]kinda points to the infancy of the port.
23:22:19*[Franklin] calls that a "duh" but
23:22:22 Quit krabador (Quit: Sto andando via)
23:22:31[Franklin]there's "ahh" bugs and "duh" bugs
23:23:09[Saint]I do suspect that someone had full intention of going back and revisiting this at a later stage.
23:23:17[Franklin]and then forgot it
23:24:32*[Franklin] wonders when Rockbox will run on those generic MP3 players that they sell in Wal-Mart and other places for ~$20
23:24:34 Join krabador [0] (~krabador@unaffiliated/krabador)
23:24:47[Saint]And this guy having decent results with lower clocks is...weird.
23:25:03*[Franklin] would buy one if he knew a port was feasible, and then (hopefully) work on the port
23:25:09ZincAlloy[Franklin], that would be super awesome
23:25:12[Saint]Back when development was consistent and I actually owned this device, when we were trying to do dynamic scaling it made practically no difference.
23:25:24[Saint]And in some cases actually made runtime drastically worse.
23:25:43*[Saint] shrugs
23:25:47[Franklin]just generic, no-name mp3 players
23:26:03saratogai think the problem with scaling before was that we were running the PCLK too high even when scaling
23:26:14saratogaso that was limiting the gain
23:26:14[Franklin]is this an ipod? −−>
23:26:21saratogaalso some of the voltages may have been higher than needed
23:26:36[Saint]aha, that's right, we found out we got massive codec performance boosts with a high clock and it wasn't worth scaling, right?
23:26:40[Franklin]has the click wheel...
23:26:46*[Saint] vaguely recalls this
23:26:53saratogai'm not sure
23:26:58copperfroggyman: no it's not
23:27:04saratogai just remember testing late and finding that a low PCLK saved a lot of power
23:27:08saratogaeven if it made the CPU slower
23:27:12froggymancopper, huh?
23:27:21[Saint]wrong f-guy
23:27:23saratogafunman: do you remember much about frequency scaling on AMSv2?
23:27:34[Saint]y'all need some last-spoken-order-sort.
23:27:36copperfroggyman: wrong nick, sorry
23:27:46copper[Franklin]: it's obviously a cheap knock-off
23:27:59*[Franklin] thought so, but the click wheel got me
23:28:29[Saint]hold gives it away.
23:28:37[Saint]not that this is even remotely ontopic.
23:29:01[Franklin]yes it is! maybe a port would be possible
23:29:25copperwhy would you want to run Rockbox on that junk
23:29:34*[Saint] looks at the rk ports
23:29:38[Franklin]why run it on anything?
23:29:51copper[Franklin]: because "anything" is not junk
23:29:58[Saint]Some of those reference design rockchip boards are less than steller.
23:30:01[Franklin]problem is, they barely give any details on the HW
23:30:09[Saint]Why do it? Because someone could, and wanted to.
23:30:14[Franklin]so you gotta buy it to find out
23:30:19foolshBy everyone her one and we'll see
23:30:27[Saint][Franklin]: that's what spudgers are for
23:30:51[Franklin]but you need to buy it
23:30:55[Saint] - for finding out what's in them.
23:30:58[Franklin](the mp3 player)
23:31:09***Saving seen data "./dancer.seen"
23:31:12[Saint]Its like $15, buy it, break it down, return the fucker.
23:31:17[Franklin]and by taking it apart, you can't return it
23:31:24[Saint]hahaha...right. ;)
23:31:32[Saint]In a perfect world, sure.
23:31:37[Saint]This one isn't.
23:31:46 Join ikeboy [0] (
23:31:56*[Franklin] wonders if he could get away with that
23:32:08*[Franklin] looks for the *cheapest* mp3 player he can find
23:32:16ZincAlloythese are being sold here cheaply:
23:32:16 Join ikeboy_ [0] (
23:33:10[Franklin]correction: I'm looking for the cheapest one WITH A SCREEN
23:33:43[Franklin]Brand name: generic
23:33:55ZincAlloy$7 is really cheap
23:34:19*[Franklin] doubts it will be powerful enough to run RB
23:34:40foolsh"Video Rider" Video format: MTV ???
23:34:55foolshwtf is MTV
23:35:14[Saint]The 1990s all over again.
23:35:27ZincAlloytheir other player can handle .avi
23:35:35foolsh Apparently its a thing
23:35:42[Franklin]this looks promising:
23:35:45 Quit nk2032 ()
23:35:57[Franklin]mp3, ogg, and wma
23:36:17[Franklin]and screws for easy disassembly
23:36:22 Quit ikeboy (Ping timeout: 250 seconds)
23:36:39*[Franklin] wonders if that's a scroll wheel on the front
23:37:16foolshdollars to donuts it just a Dpad
23:37:31[Franklin]yeah, me too
23:38:03foolshIs that a lcd too?
23:38:25foolshAs in monochrome?
23:40:03ZincAlloywith microsd capabilities the intensos could be fun targets..
23:40:21[Saint]can't be monochrome.
23:40:27[Saint]its clearly displaying 3 colors
23:40:47 Nick ikeboy_ is now known as ikeboy (
23:41:07foolshI swear it looks like a dot matrix on a two tone back drop to me
23:41:52Nemayogoki have a question, on the settings menu i can choose the mah of the battery on HM-601, the larger is 600mah, but my battery's capacity is 2800 mah, is there a way to fix this?
23:42:12[Saint]Nemayogok: its irrelevant.
23:42:21[Saint]It almost certainly doesn't do what you think it does.
23:42:39[Saint]The only thing that's used for is the estimated runtime statistics.
23:42:45Nemayogok isee
23:42:51[Saint]Its not used for battery estimation at all.
23:42:59[Saint]that's voltage based.
23:44:01[Saint]If you /really/ want to see the time remaining in the battery debug screen (which I think is the only place we present this info), you'd need to compile you own binary to get the value to go that high.
23:44:20[Saint]Its a simple one-line change, but its not really worth the effort IMO.
23:45:03[Saint]The estimated runtime almost certainly isn't particularly accurate as it is.
23:45:32[Saint]It assumes the battery is in _perfect_ 100% efficient working order. Always. And, as we know, that's not true.
23:46:20 Quit pamaury (Ping timeout: 250 seconds)
23:50:25 Quit [Franklin] (Ping timeout: 260 seconds)
23:57:21 Join [Franklin] [0] (
23:57:58foolsh Fucking touch pad

Previous day | Next day