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 2013-01-06

00:06:12 Quit Robin0800 (Remote host closed the connection)
00:06:26 Quit shamus (Ping timeout: 255 seconds)
00:06:52 Join shamus [0] (
00:13:25 Quit shamus (Read error: Connection reset by peer)
00:13:49 Join shamus [0] (
00:18:08mobenIf you got any other idea how to do this, I'd be quite happy too
00:19:16***Saving seen data "./dancer.seen"
00:22:18 Quit n1s (Quit: Ex-Chat)
00:41:32 Quit stripwax (Quit: Miranda IM! Smaller, Faster, Easier.
00:43:29 Quit Wardo (Read error: Connection reset by peer)
00:50:27 Nick tchan1 is now known as tchan (
00:50:33 Quit tchan (Changing host)
00:50:33 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
00:53:08 Quit SuperBrainAK (Remote host closed the connection)
00:54:11 Quit kevku (Ping timeout: 264 seconds)
01:06:51 Quit pamaury (Ping timeout: 252 seconds)
01:11:48 Quit bluebrother (Disconnected by services)
01:11:53 Join bluebrother [0] (~dom@rockbox/developer/bluebrother)
01:13:31 Quit fs-bluebot (Ping timeout: 272 seconds)
01:15:03 Join fs-bluebot [0] (
01:17:40JdGordon[Saint]: ?
01:18:35[Saint]moben has some form of issue with an .sbs failing but giving no debug output in the sim, and passing checkwps.
01:18:39[Saint]I can't figure it out.
01:20:41JdGordonoh cool, i'll have a look later tonight hopefully if all the info is there
01:20:50JdGordonfonts al available?
01:21:13[Saint]yep, it's all in the logs. It's set up to use the current userfont, nothing special.
01:22:27 Join mirak [0] (
01:22:51 Join SuperBrainAK [0] (
01:30:13 Join krabador [0] (
01:31:40 Join Horscht [0] (~Horscht@xbmc/user/horscht)
01:32:25 Quit ender` (Quit: If Klingons had invented Usenet, killfiles really would...)
01:46:11 Quit Horscht (Quit: Verlassend)
01:48:21 Quit bertrik_ (Ping timeout: 260 seconds)
01:59:51 Quit dfkt (Quit: -= SysReset 2.55=- Sic gorgiamus allos subjectatos nunc.)
02:05:13 Quit lebellium (Quit: ChatZilla 0.9.89 [Firefox 18.0/20121231071231])
02:05:55 Quit bertrik (Ping timeout: 248 seconds)
02:19:19***Saving seen data "./dancer.seen"
02:42:07 Quit mirak (Quit: Ex-Chat)
02:44:47 Quit eckoit (Quit: eckoit)
03:06:16 Part zaphee
03:23:22the-kyleIt's still a crapple apple.
03:24:03the-kyleThe Apple motto: You cannot, You may not, you will not, we won't let you.
03:24:43the-kyleThe readers Digest version of the Apple motto: sorry.
03:26:29[Saint]just checking, you know what channel you're in?
03:27:17the-kyleLooks like I accidentally switched tabs. Sorry.
03:28:48the-kyleThat stuff was not only supposed to go in a different channel, but also a different server. :/
03:29:07the-kyleCan't figure out how it got here, as I didn't switch tabs.
03:30:50[Saint]It happens to the best of us. But, occasionally, people do just blurt out seemingly offtopic material so I thought I'd give you a heads up :)
03:31:12the-kyleThanks hehehehe.
03:41:26 Quit [Saint] (Remote host closed the connection)
03:59:31 Join [Saint] [0] (~saint@rockbox/user/saint)
04:03:42 Quit akaWolf (Ping timeout: 260 seconds)
04:06:19 Join TheSphinX_ [0] (
04:10:01 Quit TheSphinX^ (Ping timeout: 252 seconds)
04:19:23***Saving seen data "./dancer.seen"
04:43:40 Join scorche [0] (~scorche@rockbox/administrator/scorche)
04:45:20 Quit scorche` (Ping timeout: 240 seconds)
04:45:43 Quit pixelma (Disconnected by services)
04:45:43 Join pixelma_ [0] (pixelma@rockbox/staff/pixelma)
04:45:43 Join amiconn_ [0] (amiconn@rockbox/developer/amiconn)
04:45:43 Quit amiconn (Disconnected by services)
04:45:44 Nick pixelma_ is now known as pixelma (pixelma@rockbox/staff/pixelma)
04:45:47 Nick amiconn_ is now known as amiconn (amiconn@rockbox/developer/amiconn)
04:48:03 Quit SuperBrainAK (Quit: pbly going to sleep /_\)
04:48:27 Join SuperBrainAK [0] (
04:53:59 Quit krabador (Remote host closed the connection)
05:40:41 Quit sakax (Remote host closed the connection)
05:50:11 Quit brkfstsqd (Remote host closed the connection)
05:56:04 Quit TheSeven (Disconnected by services)
05:56:13 Join [7] [0] (~quassel@rockbox/developer/TheSeven)
05:57:15 Quit prof_wolfff (Ping timeout: 272 seconds)
05:57:37 Nick amithkk_ is now known as amithkk (uid4289@gateway/web/
05:57:43 Quit amithkk (Changing host)
05:57:43 Join amithkk [0] (uid4289@2buntu/writers/amithkk)
06:10:17 Quit Cinos (Remote host closed the connection)
06:19:26***Saving seen data "./dancer.seen"
06:42:01 Quit shamus (Read error: Connection reset by peer)
06:42:22 Join shamus [0] (
06:44:51zamboniwhat headphones ya guys use
07:04:03 Join brkfstsqd [0] (
08:19:29***Saving seen data "./dancer.seen"
08:25:12 Quit SuperBrainAK (Quit: pbly going to sleep /_\)
08:50:45 Quit zamboni ()
09:15:21 Join pretty_function [0] (~sigBART@
09:20:54 Quit jhMikeS (Read error: Connection reset by peer)
09:23:10 Join kevku [0] (x@2001:470:dcc6:0:69:69:69:69)
09:24:04 Join jhMikeS [0] (~jethead71@
09:24:04 Quit jhMikeS (Changing host)
09:24:04 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
09:32:48 Quit pystar89 (Read error: Connection reset by peer)
10:08:00 Join akaWolf [0] (~akaWolf@unaffiliated/akawolf)
10:08:25 Join ender` [0] (
10:19:32***Saving seen data "./dancer.seen"
10:25:53 Quit jhMikeS (Ping timeout: 260 seconds)
10:38:38 Join stoffel [0] (
10:50:34 Quit pretty_function (Remote host closed the connection)
10:50:40 Join Horschti [0] (~Horscht@xbmc/user/horscht)
10:51:37 Join stripwax [0] (~Miranda@rockbox/developer/stripwax)
11:01:44 Join bertrik [0] (~quassel@rockbox/developer/bertrik)
11:03:53 Quit stripwax (Quit: Miranda IM! Smaller, Faster, Easier.
11:23:29 Quit Horschti (Quit: Verlassend)
11:29:50 Join theli_ua [0] (
11:53:46 Join pretty_function [0] (~sigBART@
12:07:38 Join dfkt [0] (dfkt@unaffiliated/dfkt)
12:16:50 Quit akaWolf (Ping timeout: 252 seconds)
12:17:24 Join y4n [0] (~y4n@unaffiliated/y4ndexx)
12:19:35***Saving seen data "./dancer.seen"
12:20:15 Join akaWolf [0] (~akaWolf@unaffiliated/akawolf)
12:30:39 Join bertrik_ [0] (
12:30:39 Quit bertrik_ (Changing host)
12:30:39 Join bertrik_ [0] (~quassel@rockbox/developer/bertrik)
12:32:31 Quit pretty_function (Remote host closed the connection)
12:36:01 Quit stoffel (Ping timeout: 252 seconds)
12:37:22 Join n1s [0] (
12:37:22 Quit n1s (Changing host)
12:37:22 Join n1s [0] (~n1s@rockbox/developer/n1s)
12:41:16 Join stoffel [0] (
12:59:31makotodoes the ipod mini build of rockbox spin down the microdrive and make use of the flash buffer?
13:01:54n1syes it spins down the drive on every hd based player, some even have the ability to power down the disk, not sure about the mini there
13:02:47makotoi do recall seeing a patch floating about for the compactflash mods
13:07:37 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
13:08:37 Join mrkiko [0] (
13:23:42 Quit shamus (Read error: Connection reset by peer)
13:23:56 Join shamus [0] (
13:26:31bluebrotherstock Rockbox works fine for me on a Mini with CF mod.
13:31:45 Join zaphee [0] (~user@2a01:e35:2e82:4050:f127:7621:eb54:4d15)
13:34:07 Join Wardo [0] (
13:47:18makotojust curious how efficient the 32mb flash buffer is when dealing with large files
13:48:18n1sit reads in as much data as fits in the buffer and then spins down the disk, the size of each file doesn't matter
13:50:22 Join psilonux [0] (~psilonux@2001:470:1f15:10f6:55e8:63a7:8d42:ae4e)
13:54:02gevaertsWhat flash buffer are you talking about?
13:54:16gevaertsThere's storage, which is disk or flash, and there's RAM
13:54:28n1si was assuming the ram buffer :)
13:54:54gevaertsRAM is 32MB on the mini, of which a large part (28MB or so?) is reserved for buffering audio, but that doesn't change depending on type of storage
13:55:46gevaertsAnd yes, flash is expected to be more power-efficient than disk, in part due to the lack of power-hungry spin-up, but not all CF cards are equal
13:59:19 Join breakfastsquid [0] (
13:59:20 Quit brkfstsqd (Read error: Connection reset by peer)
14:15:29 Quit Wardo (Read error: Connection reset by peer)
14:19:37***Saving seen data "./dancer.seen"
14:21:54 Quit stoffel (Ping timeout: 246 seconds)
14:24:48 Join nateloaf [0] (
14:43:59 Quit pamaury (Remote host closed the connection)
14:54:45 Quit nosa-j (Read error: Operation timed out)
14:58:21kugelJdGordon: did you see my question?
15:04:11 Join nosa-j [0] (~m00k@
15:12:26 Join pamaury [0] (~quassel@rockbox/developer/pamaury)
15:38:22 Join cela [0] (
15:39:12celaRK27xx Port. There is an audio bug introduced around b19b250, a popping sound in the background after about 30 seconds of playing an mp3, rolled back the change in system-rk27xx.c, rebuilt and it's ok on my generic rk27xx pmp. Suggest making the change only applicable to HiFiman port, Should I file a bug report?
15:42:31celaOk. I'll give mortalis or wodz chance to pick up on this first, cheers.
15:43:49 Quit cela (Quit: CGI:IRC (EOF))
15:47:00 Join stripwax [0] (~Miranda@rockbox/developer/stripwax)
15:52:32pamaurywe should have asked him to bisect
15:55:42 Quit n1s (Quit: Ex-Chat)
16:00:04 Join Ward [0] (
16:00:29 Nick Ward is now known as Guest31500 (
16:05:53 Quit theli_ua (Ping timeout: 272 seconds)
16:10:24[Saint]makoto: if you watch our gerrit tracker, I'll post a patch that detects the storage type and disables irrelevant options like spindown et al.
16:11:18[Saint]It isn't necessary to do so, setting these settings won't confuse the target if you change the storage type...but it can be confusing to users who don't realize the setting(s) no longer apply.
16:12:20*[Saint] is also reminded he has CE-ATA detection for Classic battery capacity guessing to push
16:12:37 Quit stripwax (Ping timeout: 260 seconds)
16:13:10makotothe only reason im thinking of this, is that my ipod (with the OF at the moment) is quite laggy and crashy
16:13:51[Saint]The OF might get all different types of confused...I don't know what storage type assumptions it makes.
16:14:03[Saint]I strip the OF from my ipods, personally.
16:14:56 Quit psilonux (Quit: Leaving)
16:15:22[Saint]Rockbox can definitely handle the storage type changing, though. The only side effect is slightly confused settings that will no longer apply:
16:15:32[Saint]spindown, antiskip buffer.
16:18:21makotoi'm about to rockbox mine now
16:19:41***Saving seen data "./dancer.seen"
16:22:55[Saint]Ah, congratulations. You'll love it.
16:23:12[Saint]A CF'd Mini is a very fine thing.
16:23:28*[Saint] <3's his
16:33:46makotodone :D
16:34:10makotomy sansa fuze died about 6 months ago, and i really missed rockbox
16:34:55makotoi was very very careless when i redid my battery and had a look round the microdrive
16:35:29makotoi somehow managed to fracture the solder joints for the connector on the disk ribbon :/
16:35:47makotonot great with the soldering iron but i somehow managed to fix it
16:37:56 Join theli_ua [0] (
16:45:45 Join jeoj [0] (
16:46:16makotohrmm, not thought of this now, but, turning the ipod off? there's no sleep like in the OF, right?
16:46:23 Join eckoit [0] (~ryan@
16:48:44jeoji'm a total beginner
16:48:46jeojso bear with me
16:49:20jeojcan someone tell me right off the bat, is there a way to put some kind of a scrobble logger to my sony nwz-s615f?
16:49:53[Saint]If that isn't a supported Rockbox target, ..."no".
16:50:02[Saint]where "no" == "almost certainly not".
16:50:08makotohrmm, no way to change backlight's brightness lavel?
16:50:16jeojyou shouldn't compare strings like that, but ok
16:50:34 Quit eckoit (Client Quit)
16:50:47makotosmells like time to have a play with the source and learn some code.
16:51:06AlexPmakoto: On most targets yes, no idea about the classic
16:51:09 Join Janslike [0] (
16:51:13[Saint]Settings - General - Display - LCD Settings - Brightness iirc
16:51:27[Saint]AlexP: Mini
16:51:38AlexPoh, not sure where I got classic from
16:51:45JanslikeFor a Sandisk Clip +, how long will it typically take to initialize the database for 29gb?
16:51:52[Saint]Probably my comment earlier.
16:51:54JanslikeRoughly 1300 files
16:51:58makotoi can only find contrast
16:52:25[Saint]Ah, right, I forgot about that. Contrast and brightness are basically interchangeable in this context.
16:52:59JanslikeWait scratch that, 4268 files exactly
16:53:03makotocontrast has no effect on the brightness of the LEDs
16:54:18[Saint]Oh dear, I am getting confused. I thought it did fade the backlight as well.
16:54:41[Saint]We can fade the backlight on the Mini I'm pretty sure, it fades in/out when it turns off.
16:55:10makotoyeah- theres fade in/out
16:55:18gevaertsJanslike: hard to say. Not *too* long though
16:55:19makotoand you can control how long you want it to fade in/out for
16:55:28makotoso there must be some kind of control over it
16:55:44makotothis was an idea that came to me last night in bed in a pitch black room.
16:56:26[Saint]looks like the mini doesn't have hardware backlight control.
16:56:32[Saint]...I wonder why I thought it did?
16:57:20[Saint]using software to fake a brightness setting for more than a few seconds at a time would likely fall into the "bad things" category.
16:57:41[Saint]cpu/battery intensive.
16:57:50JanslikeIf it seems to be taking exceptionally long, is there any way to know if it's broken or still doing its thing?
16:58:04JanslikeRight now it says "Building database... 120"
16:58:33[Saint]Janslike: if it hasn't incremented the counter in some minutes, and hasn't finished, it's stuck.
16:58:55[Saint]you can reboot (this is perfectly safe) and enable metadata logging fom the debug menu and try again.
16:59:18 Quit kevku (Quit: KVIrc 4.3.1 Aria
16:59:30[Saint]then, when/if it fails again, there will be a log of where it got stuck written to the root of the disk.
17:00:06makotoahh, done by pwm.
17:00:17makotoi guess i'll grab the source code and do
17:00:18[Saint]The likely scenario is it is choking on some bad metadata tags in a certain file.
17:00:52makotodo "bad things". i only want it on for a few seconds and dimly so it shouldn't be a massive problem if i do it properly.
17:01:07Janslikemaybe that's deerhoof
17:01:37[Saint]the number of files listed in the database count can be confusing.
17:01:48[Saint]It is listing *every* file it finds, not just media.
17:02:12[Saint]that's why the number can seem way off at times.
17:02:15 Part jeoj
17:02:20JanslikeAh well I have foobar set up so it only transfers the music
17:02:32JanslikeGetting stuck on SYL weird have to double check that
17:03:06gevaertsThere are a few hundred files in .rockbox. Those also appear in the count
17:03:25[Saint]Janslike: sorry, did I parse that correctly? You're stripping the metadata from your audio?
17:03:38JanslikeOh right yeah forgot about those
17:03:53JanslikeOh no I thought you meant like, exteraneous folder.jpgs etc
17:05:34[Saint]Ok, damn, then. :) I thought that was going to be an easy answer.
17:05:49Janslikestuck at 1948 this time
17:05:54JanslikeI think it's just being stubborn
17:07:46 Join kevku [0] (x@2001:470:dcc6:0:69:69:69:69)
17:09:23JanslikeStuck at 2112, now 60
17:12:04JanslikeIt reads the files perfectly though
17:12:20*[Saint] would advise checking the filesystem
17:14:43Janslike??? In what way
17:14:51 Quit Rower85 (Quit: Hmmm...)
17:16:24 Quit kevku (Remote host closed the connection)
17:16:42 Join Rower85 [0] (
17:17:00[Saint]Depends on your operating system.
17:17:08 Join kevku [0] (x@2001:470:dcc6:0:69:69:69:69)
17:18:29pamauryour metadata parsers are supposed to withstand all sorts of funny things but it seems that users always come up with new ones :) Bad filesystem seems more likely though
17:19:41JanslikeWindows didn't come up with anything
17:19:47[Saint]I started to think so when it wasn't failing in a repeatable way, yes.
17:20:05[Saint]Hum, well. This is fun.
17:20:25[Saint]What are the files in question? Are they some less-than-common or exotic format?
17:21:20JanslikeAll V0 MP3
17:21:41 Part theli_ua ("Залишаю")
17:21:59Janslike4055 at 28gb
17:22:22[Saint]One thing you can try is to wipe out the existing (if any) .tcd files in /.rockbox
17:22:50[Saint]just to make sure there's no damaged portion of a previous database leftover.
17:26:51Janslikeright, sorry, played a bit of doomf.wad
17:27:03JanslikeNo wonder I couldn't find my way around, completely new map
17:32:31[Saint]You can play /almost/ any .wad files
17:32:51[Saint]Anything that loads in the freedoom project should "just work" there also.
17:35:12JanslikeFroze up when I was browsing
17:36:55JanslikeI'll try a sector scan though
17:37:10JanslikeI mean, I guess it's pointing towards a problem with the microsd
17:37:24JanslikeWhich is depressing if true
17:41:58[Saint]easy to test.
17:42:09[Saint]eject the card.
17:43:16 Join SuperBrainAK [0] (
17:46:21JanslikeDatabased 100 tracks without problem
17:46:33JanslikeI'm gonna leave it and come back in 2-3 min
17:54:16JanslikeI turned on auto-update and load to RAM and that fixed it
17:54:49JanslikeIf it has any further problems playing the files now I know why at least. Hopefully that won't happen. Anyways thanks for the help and have a good day
17:54:59 Quit Janslike (Quit: Hello)
17:55:52[Saint]that doesn't seem to make any sense at all. :-S
17:57:42 Join stripwax [0] (~Miranda@rockbox/developer/stripwax)
17:58:12 Quit stripwax (Client Quit)
18:19:43***Saving seen data "./dancer.seen"
18:22:19 Quit mc2739 (Quit: leaving)
18:32:19 Join eckoit [0] (~ryan@
18:32:28 Join pretty_function [0] (~sigBART@
18:36:48 Quit SuperBrainAK (Ping timeout: 252 seconds)
18:38:30 Quit shamus (Read error: Connection reset by peer)
18:38:47 Join shamus [0] (
18:39:53 Quit Guest31500 (Read error: Connection reset by peer)
18:42:55 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
18:45:30 Join zamboni [0] (~bottledwa@unaffiliated/zamboni)
18:58:14 Join Raptors [0] (
19:06:02 Quit nateloaf (Quit: Leaving.)
19:11:38bluebrotherWriting an image using libpng isn't hard at all. Nice.
19:15:09 Quit Belzebub (Ping timeout: 255 seconds)
19:19:24 Join Belzebub [0] (~torrentow@
19:26:35 Join Guest79668 [0] (
19:27:34Guest79668JdGordon: Hi, i'm having a question about FS #12124 aka list-skinned
19:27:35fs-bluebot draw lists using skin engine (patches, closed)
19:28:20Guest79668does it requires a specific compilation? How does "old school" themes reacts to it?
19:28:49[Saint]It requires no compilation (note, it's closed).
19:28:58[Saint]It's in HEAD currently.
19:29:09[Saint]And, old themes don;t rwact to it, at all.
19:29:16[Saint]*don't react
19:30:05Guest79668So, may i ask if there is a functionnal theme implementing it that could be tested?
19:30:30AlexPGuest79668: A theme contains tags, if it doesn't contain a tag then it won't try to use it
19:30:55[Saint]That is the only known example of skinned lists actually working, afaik.
19:31:06Guest79668Thanks a lot
19:31:09[Saint]and it took quite some degree of hackery iirc.
19:31:17[Saint]gevaerts is quite proud of that one.
19:31:44Guest79668I was working on some grid menu, and then saw that skinned-list.c. That was a way better idea
19:33:13[Saint]afaik there's no way to do a grid type menu without skinned lists.
19:33:13Guest79668Thanks a lot [Saint] :)
19:33:33AlexPI did a grid menu with skinned lists
19:33:35Guest79668i did, but i broke the whole list concept, and was about to write the callback function
19:33:45AlexPWorked quite well on non-touchscreen
19:33:56AlexPI had some problems with scrolling on touchscreen IIRC
19:34:00[Saint]Ah, I meant with the theme language. Editing the core doesn;t count :)
19:34:32Guest79668AlexP : any link out there? :p
19:34:48AlexPI don't remember if I uploaded it, I might have done in the forums
19:34:56AlexPOr given a link at least
19:35:04Guest79668k, gonna search ;)
19:35:16AlexPI can't remember though, sorry
19:35:35Guest79668nop, thanks both of you, that was quite an help :)
19:36:08 Quit nosa-j (Ping timeout: 272 seconds)
19:37:44*gevaerts recommends looking at instead
19:38:03[Saint]Ah, I forgot about that one.
19:38:06[Saint]thanks gevaerts
19:38:12gevaertsWavy is really more of an experiment to see how far I could go, which probably isn't too helpful
19:38:34gevaerts[Saint]: you can do grids
19:38:49gevaertsYou just have to say ",grid" somewhere in one of the tags
19:38:49[Saint]I know.
19:39:06*gevaerts learns to read
19:39:11[Saint]I said no way to do it *without* skinned lists.
19:40:44 Join nosa-j [0] (~m00k@
19:42:56Guest79668and can you navigate into the grid with up and down for lines?
19:43:37gevaertsGrid really works best on touchscreens
19:44:13gevaertsOnly the display changes, not the controls
19:44:21Guest79668yep, though about it
19:44:54Guest79668even by remapping, it would need a bit of extra code
19:49:30 Join prof_wolfff [0] (
19:49:40Guest79668Well, thanks a lot, gonna study those examples ;)
19:49:59 Part Guest79668
19:53:53*[Saint] wonders why plugins and shortcuts don't have a red leading char
19:54:11[Saint]gevaerts: ^
20:03:05 Join krabador [0] (
20:07:46 Join Ward [0] (
20:08:10 Nick Ward is now known as Guest94269 (
20:09:34 Quit pretty_function (Remote host closed the connection)
20:10:46gevaerts[Saint]: You mean the %?cs?
20:11:42[Saint]In the main menu, neither Plugins nor Shortcuts has the red leading char.
20:11:54[Saint]...I can't figure out the significance.
20:11:58gevaertsAh, right :)
20:12:24*gevaerts considers not giving away the secret :)
20:13:02[Saint]Pffffft! not that secret, I could just look. I'm just lazy :)
20:13:07gevaertsIt's because red means the leading letter is different from the previous line
20:13:33[Saint]IS there no way to disable that per-screen?
20:13:41[Saint]Seems like yo ucould've checked for that.
20:13:50[Saint]...unless it's wanted?
20:14:23[Saint]I know why you want it in playlists, and directories etc.
20:17:06[Saint]you could switch it out conditionally with %if and %cs
20:19:45***Saving seen data "./dancer.seen"
20:26:47gevaerts[Saint]: if you want to actually finish my themes, feel free :)
20:27:17[Saint]Instructing others on things that should do is so much more fun :)
20:27:28*[Saint] a word
20:27:36gevaertsYes, that's why I'm doing that!
20:27:56[Saint]...but! Awww.
20:30:49zamboniwhere can i fix my sansa fuze battery
20:30:52zambonithe fuze+ sucks!
20:31:54*pamaury wonders if he is the only one to not dislike the fuze+
20:31:57[Saint]zamboni: constantly coming in here and spouting offtopic stuff is a good way to get people to ignore you.
20:32:45zamboniI'm just not sure where to ask
20:33:48pamauryyou do not ask anything, you write unrelated and vague sentences
20:34:41*[Saint] is quite confident SanDisk forums is the place for SanDisk products
20:35:00[Saint](when unrelated to Rockbox)
20:35:24zambonisandisk forums suck
20:36:44[Saint]So the answer it to pick a channel seemingly at random, and ask unrelated questions.
20:36:51[Saint]Yep, I can see how that workls.
20:40:21zambonia bit
20:40:33zambonisorry, I'm not saint
20:40:35zambonilol get it
20:41:25[Saint]I do. I feel sorry for those that aren't me too.
20:48:01 Join SuperBrainAK [0] (
20:57:47 Join dynion [0] (
20:59:00dynionis it correct that sony and philips mp3 players are not compatible with rockbox?
20:59:15AlexPIf they aren't listed on, then no
20:59:24AlexPBut some philips are to some extent
21:00:25pamaurydynion: I recently explored the possibility of a very generic port to sony nwz's players but it's still very early stages and I know some recent philips players are quite easily rockbox'able with a moderate amount of work
21:01:06[Saint]dynion: *what* phillips player
21:01:17[Saint]the "if it isn't on the main page" thing isn't really true.
21:01:42[Saint]There's a tonne of partially supported players it doesn't list.
21:04:09AlexPI didn't say main page
21:04:13dynioni knew there would be more:)
21:04:20AlexPI said the site
21:04:27AlexPAnd there are links from the main page
21:04:37[Saint]Ah, well...fwiw, some of them barely get a mention there either :)
21:04:42[Saint]also, sorry. I misread.
21:05:00dynionsony nwz-a820
21:05:34AlexPand under
21:05:41dynionor possibly a philips sa28xx
21:06:08[Saint]the go-gear series is all we (partially) support, iirc.
21:06:37 Quit akaWolf (Ping timeout: 260 seconds)
21:07:04pamaurydynion: are you running windows or linux ? If you have a linux at hand I can quickly evaluate the chances that the nwz-a820 be supported or not
21:07:34dynionor android, which basically is linux:P
21:08:04[Saint]for a very broad sense of the term.
21:08:26[Saint]it's a massively crippled linux with most of the interesting/useful bits removed.
21:08:37dynionit works fine though...;P
21:08:42dynionyou did made me laugh
21:08:51pamaurywith linux you could have sent special commands to the device to see if it understands the protocol I reverse engineered. This tool currently doesn't work under Windows but could with a small amount of work. could you open a post in the New Port forum about the nwz-a820 ?
21:09:19dynion15 minutes ago i heard abot rockbox for the first time...:p
21:09:23dynionah well
21:09:26dynioni will just search
21:10:01dynionrockbox developments −−> new ports i suppose?
21:10:08dynionshould i give any information?
21:10:38pamauryThe nwz-a820 that's one of the player I investigated and I wasn't able to brute force the key but if you players responds to the commands, we can potentially get the key. That would be a of great help for our understanding of the nwz series. Just watch the forum thread so I can post to it when the tool is windows ready
21:10:40 Quit shamus (Read error: Connection reset by peer)
21:11:05dynioni just grabbed the box and found out it's an nwz-a828
21:11:09dynionbasically the same
21:11:14dynionbut who knows
21:11:22 Join shamus [0] (
21:11:50pamauryall the nwz-a8xx are the same software-wise :)
21:12:48dynionyeah i thought so already:P
21:14:08dynioni hate the sony for not using flac
21:14:53pamauryI must say the sony software is not that bad except for two things: poor format support and bad usb support under linux
21:15:17dynionso.... rockbox won't enhance the sound?:P
21:15:42 Quit y4n (Quit: PÆNTS ØLF!)
21:16:34 Join jhMikeS [0] (~jethead71@
21:16:34 Quit jhMikeS (Changing host)
21:16:34 Join jhMikeS [0] (~jethead71@rockbox/developer/jhMikeS)
21:17:10pamauryprobably not :)
21:17:38dynionwhat's the pro then?
21:18:20AlexPThat lists a small subsection
21:18:33[Saint]In some players, there is a marked difference in the audio quality.
21:18:37[Saint]In others, notsomuch.
21:19:15[Saint]It really depends exactly how silly and/or crippled the OF/hardware is.
21:19:19dynioni thought rockbox was a different audio driver
21:19:31[Saint]It is a complete firmware replacement.
21:19:33dynionbut it looks like it's more:O cool!
21:19:43AlexPIt replaces everything
21:19:47AlexPIt is a complete OS
21:20:12[Saint](with some exceptions to that rule, but this is true for all the DAPs)
21:20:23dynionOS? so the menus will differ as well...
21:20:32[Saint]*everything* will :)
21:20:32AlexPEverything is different
21:20:38AlexPIs a complete replacement
21:20:42[Saint]or, would, rather.
21:20:59[Saint]I'm talking in a tense that implies your player is supported.
21:21:03dynioni like the fastness of my sony..... is rockbox just as easy?
21:21:14AlexPDepends on you
21:21:16[Saint]moreso in many cases.
21:21:24AlexPSome people would say yes, some no
21:21:41[Saint]we have a lot of options to enable some fairly quick shortcuts to things.
21:21:43AlexPmore powerful for sure, and extra features and possibilities will always increase complexity somewhat
21:22:22dynioni do like a lot of those things
21:22:26[Saint]enabling the shortcuts menu as the quickscreen is probably the fastest way to navigate around often used items.
21:22:46dynioni have no problem with complex
21:22:49*[Saint] realises those terms probably mean very little
21:22:54dynionif it's complex in a logical way
21:23:06dynionfor example: i love excel, as it is complex, but useful
21:23:40[Saint]Some people have a very violent reaction to how "unintuitive" Rockbox seems to them. But this is really a veiled reaction to "OMG, it's different!"
21:23:41dynionwhat player do you guys like most? as in bang for the buck
21:23:58dynionhmm.. i hate ios for NOT being different
21:24:24dynioni'm always trying around with all my stuff... just to look if it works better:)
21:25:04[Saint]I think it's hard, if not impossible, to provide a UI that evry single man woman and child can "just work". So I always laugh when people argue about things not being intuitive.
21:25:20dynioni hear a lot of guys about a sanse clip + with rockbox
21:25:26[Saint]I know what they /mean/, but they often can't articulte what would make it "intuitive".
21:25:41dynionactually that's how i found rockbox at all... but what players would you all suggest?
21:25:42*the-kyle has a clip+ with Rockbox and loves it.
21:26:01[Saint]Depends, flash, or HDD?
21:26:05 Quit mrkiko (Quit: leaving)
21:26:16[Saint]Probably a Clip or a Fuze
21:26:23dynionit has to be rock solid to survive me... sony's are solid...somehow
21:26:27[Saint]cheap, accessible, expandable storage.
21:26:33[Saint]...and supported :)
21:26:45dynioni had a sansa before
21:27:06dynion4 to be exact... the 3,5mm in just broke down time after time
21:27:14the-kyleI have a Clip+ now, but the next supported player I purchase will likely be the ClipZip.
21:27:16dynioneven though i was
21:27:28dynionwell i would even say subtle
21:27:52[Saint]If money wasn't an option. iPod Classic, hands down.
21:27:56dynionit just broke down
21:28:07[Saint]Well made, awesome hardware, and it is a beast in terms of power.
21:28:19dynionthere's an apple logo on it... right?
21:28:23dyniondidn't that mean it sucks?
21:28:29[Saint]Not at all :)
21:28:44[Saint]Apple make great hardware, they just suck at life and business.
21:28:47dynioni listened to a few apple's and the all made me puke
21:29:02dynionit really is the worsest experience i ever had
21:29:07the-kyleThe Classic only sucks if Rockbox isn't on it.
21:29:19[Saint]Rockbox would change that opinion, I almost guarantee that.
21:29:29dynionoh wait
21:29:33dynionthat's not the old nano
21:29:43dynionits the hdd thingy
21:29:49[Saint](which we also support)
21:30:08dynionthe old nano is the worsest thing i ever found
21:30:19dynionthe classic sounded like crappy drivers and good hardwar
21:30:39[Saint]I actually think that is the Rockbox player with the best sound quality.
21:30:43dynionbut that thing is 200 euro... a hifiman hm-601 is preferred in such a case, i suppose
21:30:53[Saint]It's really just a little flash powered ipod video with a tiny screen.
21:32:12dynionthe ipod classic has an hdd...
21:32:15dynionand it breaks down a lot
21:33:28dynionand i like cycling on a racing cycle through the woods and mud at 3:00 at night
21:33:43dynionand sometimes water if i'm not careful:/
21:34:01the-kyleVery little will survive that very long without a protective case of some sort.
21:34:16dynionthe water only happened just once
21:34:26dynionthe sony is keeping up with me for about 2 years now
21:34:56dynionwith only 2 scratches
21:35:33dynionbut my sennheiser CX880 died last week... so i'm searching for new earbuds and when i'm busy, i might upgrade my mp3 player as well
21:35:50[Saint]A Clip(V2/+) seems like a very good option for you.
21:36:03[Saint]they are basically indestructible, although they /feel/ cheap.
21:36:15[Saint]Several people here have had them survive a washing machine.
21:36:20the-kyleI've had my Clip+ for over a year now. I'm hard on it, but I can't say I abuse mine that much hehehehe.
21:37:13[Saint]If you require a color screen, all Clip(s) are out.
21:37:18dynionhmmm everyone is positive about this clip... is there any chance i misjudged sansa?
21:37:32the-kyleBest thing is I didn't pay very much for it, so it's also quite easily replaceable in case something bad does happen.
21:37:42dynioni don't give a shit about colored screens... i have a nexus 7 if i want videos of some sort...:p
21:37:55the-kyleThe ClipZip isn't color?
21:37:58*[Saint] will say uncategorically that he HATES the Clip(s) hardware.
21:38:10[Saint]It feels like a McDonalds Happy Meal toy.
21:38:21dynioni agree on that poind
21:38:27[Saint]...but the sound quality is awesome, and they are cheap, powerful, and virtually indestructible.
21:38:41dynionmy english really sucks... haven't used it in a while
21:38:51[Saint]you're doing very well.
21:39:11dynioni know 3 clips, the normal one, the clip+, the clipzip
21:39:18the-kyleI decided not to judge the book by its cover, and I ended up being right. It feels cheap on the outside, but it's much better inside.
21:39:32dynionthe clipzip is smaller en lighter
21:39:35the-kyleClipZip is newest.
21:39:36dynionbut i dont care about weight
21:39:40[Saint]"the normal one" is really two main variants with at least two variants of a variant. :)
21:39:47[Saint]It can be confusing.
21:39:50dynioni caer about audi quality
21:39:56dynion"the normal one" is really two main variants with at least two variants of a variant. :)
21:40:02dynionand the next thing you say
21:40:05dynionit can be confusing
21:40:08dynionoh really?:P
21:40:39the-kyleMost of that is taken care of in software, so you don't notice the differences using the Rockbox Utility to install.
21:40:45dynioni found out when you used the word variant 4 times in one sentence, and referred to it once
21:40:54[Saint]there are V1 and V2 Clips, and the hardware can differ slightly, even though the model number doesn't.
21:41:05[Saint]as far as I know it is just the FM chip that can differ.
21:41:06dynionah, that's readable
21:41:09[Saint]...possibly the LCD.
21:41:16dynioni don't use radio anyway:)
21:41:26dynionwhat about the clip+?
21:41:36dynionit's widely sold in the netherlands
21:41:48the-kyleClip+ has 2 FM chips also I think. Not sure though.
21:42:30dynionfm radio chips... but no better audio quality?
21:42:41dynionsansa says it has
21:42:58[Saint]It's a nice player, the Clip+ is ever so slightly smaller, and has a nicer feeling directional pad.
21:43:06the-kyleI can't compare with/without Rockbox, since I'm visually impaired, and so can't use my Clip+ without Rockbox.
21:43:11funman[Saint]: the CPU is different between v1 & v2, also the amount of RAM
21:43:16funmanclipv2 can run doom, clipv1 can't
21:43:28dynionthat does differ quite a lot than:P
21:43:37dynionthe lcd, the hardware and the radio
21:43:37bluebrotherred leading letter?
21:43:39funman8MB versus 2MB
21:43:50funmanand armv5 versus armv4
21:43:50dynionthat's almost another player
21:43:54[Saint]They realised, I think, that they included a rather excessive amount of RAm in the V1 Clip.
21:43:58funmanalso clipv2 battery life is 3 times longer
21:44:00[Saint](thanks funman I forgot that)
21:44:02dynioni hate armv4 :/
21:44:07dynionit doesn't run anything
21:44:11funman[Saint]: actually it's the reverse (8mb for clipv2)
21:44:27[Saint]bah, mishit a key, yes, thanks.
21:44:36[Saint]Wait, no I didn't.
21:44:44[Saint]Nope, I was outright wrong :)
21:44:53funmandynion: that's rather uneducated, armv4 runs rockbox
21:45:02 Join Cinos [0] (
21:45:15dynionarmv4 doesn't run anything with android
21:45:22[Saint]armv4 is plenty for a DAP.
21:45:28dynioni know for rockbox about half an hour:p
21:45:35dynionnormally i'm in another line of work
21:45:53dynionsmartphones, tablets, computers... more advanced in processing power, less in audio quality:)
21:46:29dynionbut as my in-ears died, i can also learn a lot about this market... it can not hurt:)
21:47:51dynionclip + vs clip zip in audio quality and build quality?
21:48:08dyniondesign is not an issue as it's in my pocket almost always
21:48:49[Saint]Other people won't agree with me, I'm sure, but if you have the money and like well made hardware...I recommend an iPod Classic 6/6.5/7G. If you want a cheap, basically disposable player, that is still awesome...and available, a Clip/Fuze.
21:49:20dynion <−− this was the stupid, love to breakdown sansa i had
21:49:35[Saint]I don't think any player can beat an iPod on build quality, personally. But that's me.
21:49:52dynion3 reasons for not buying the apple
21:49:56dynionit's expansive
21:50:03dynionall my friends know i hate apple
21:50:22dynionit actually would be fun to misuse an apple and say it's good, but most will laugh at me... i don't want that
21:50:27bertrik_dynion: what model is that?
21:50:37dynionand i breaks down way faster with my kind of behaving
21:51:07*[Saint] has never seen that player wither.
21:51:24dynionbertrik_ sansa express
21:51:26[Saint]reminds me of a C200, kinda.
21:51:44dynioni needed a mp3 player just before the first clip came out
21:51:54dynionbut it was for my birthday, so i couldn''t wait.
21:52:00[Saint]It's a "C200 Shuffle" :P
21:52:02dynionthen i bought this one
21:52:16bertrik_[Saint]: kind of a bastard combination of a clip and a C200 and an M400
21:52:25*[Saint] nods
21:52:44dynionand i really, really hated it... it's the most hated mp3 player i had, and i only hate the ipod shuffle and the ipod nano of older generations more
21:53:07pamaurybluebrother: I just realized that sg3_util has a windows port !
21:53:26dynionit broke down 4 times in two weeks, and each time i got a new one, which broke down as well
21:53:28*[Saint] restrains himself from ranting about his much beloved iPod Nano1G
21:53:59dynionthen i bought my sony which somehow can't die
21:54:28dynionsaint, what earbuds do you use?
21:54:31dynionor headphones
21:56:03*[Saint] points dynion to #rockbox-community for that
21:56:19dynionok, willdo
21:56:50dynionbut still, clip+ vs clipzip
21:57:32 Join einhirn [0] (
21:57:55dynionsansa says the clip+ has better sound
21:58:13dynionclip zip looks less solid
21:58:18dynionbut i might be wrong on both
21:58:54 Quit einhirn (Client Quit)
21:59:15[Saint]What sansa says about the sound is kinda irrelevant, as they can't produce a true pitch with their firmware anyway nless they patched it out.
21:59:47funmanaudio quality for clip zip and + should be identical as they are the same device with a different screen
22:00:22[Saint]funman: did they ever fix the pitch issues in the OF?
22:00:22[Saint]I know they /tried/, did they succeed?
22:00:50funmanOF is off topic here
22:01:05dynionthanks funman:)
22:01:16dynionand build quality?
22:01:33 Join einhirn [0] (
22:01:39dynionmostly smaller, lighter things are stronger
22:01:55dynionas they can't crash itself with their own weight
22:06:11dynionwhat about the fiio x3?
22:06:57 Quit krabador (Ping timeout: 255 seconds)
22:19:47***Saving seen data "./dancer.seen"
22:27:24*[Saint] wonders if anyone has ever considered dynamic scaling with more than just boosted/unboosted
22:27:47[Saint]the Classic boosting to ~200MHz just to run the UI is a bit...overkill.
22:28:23[Saint]considering it never, ever boosts during playback (mp3@320 CBR) until it needs to rebuffer.
22:28:28bluebrotherpamaury: I know. However, I really don't want yet-another-dependency
22:28:33bluebrotherregardless of the OS
22:28:56bluebrotherespecially since we have some kind of custom SCSI support in ipodpatcher already
22:29:10bluebrotherhowever, it needs to get adjusted. Which is simply work someone needs to do :)
22:29:17[Saint]Would it be possible to add intermediate frequencies, instead of just boosted/unboosted?
22:29:35bluebrother(and the OS X implementation is unfortunately a bit limited. And doing custom SCSI on OS X is a real pain)
22:30:30gevaerts[Saint]: possible, yes. Deciding when to use which level, well...
22:30:37gevaertsNot the simplest of issues :)
22:31:13[Saint]There's many fairly simple models for governors out there.
22:31:17pamaurywell, since it's already a dependency on linux I don't see why it's a problem on windows. fortunately it's much simpler to do directly on windows so that's not a real problem I guess
22:31:22[Saint]..but I guess you're talking integration.
22:32:06[Saint]the most simple model, is likely ondemand, iirc.
22:32:26bluebrotherno, it's not a dependency on Linux
22:32:36[Saint]"is the CPU in high demand? boost a step. Is it still? Yes - boost again. No? check if we can step down. etc.
22:32:38bluebrotheronly time-sync uses sg3_utils
22:32:46bluebrotheripodpatcher (and thus Rockbox Utility) does not
22:33:07preglowso, how bout them commit mails? still no go?
22:33:14bluebrotherpreglow: no
22:33:21 Join Dex__ [0] (
22:33:26[Saint]preglow: still no go, yes.
22:33:33pamaurybluebrother: in ipodpatcher I can see #include <scsi/sg.h> in ipod-posix.c !
22:34:00preglowweird, why not? one would think some simple scripting would make that happen
22:34:12[Saint]be our guest :)
22:34:17*bluebrother checks
22:34:23preglow[Saint]: sure, just wondering if there's a better reason first
22:35:04bluebrotherah, yes. But that's not part of sg3_utils
22:35:24bluebrotheron my machine that's part of glibc-common
22:37:29pamauryah so it does raw ioctl too
22:38:13JdGordonkugel: what question? about buffering_reset()? I just wonder if it is ever safe
22:38:29JdGordonit now nukes every handle currently allocated
22:38:33JdGordonso everything is going to break
22:39:25prof_wolfffSaint, i was really busy last months but still had time to do some testing and analizing how OF behaves, actually i am running my classic at 36MHz while playing mp3@320, at this freq lcd is slow, OF probably goes as low as 18MHz., i think we can try to reach 13.5Mhz while playing and boost it while decoding
22:40:34[Saint]prof_wolfff: the idea of intermediate scaling frequencies intrigues me.
22:40:46[Saint]...but, I don't think I have the codig ability to get there.
22:41:16Dex__has anyone heard the new Dr.Zilog album
22:41:33pamaurybluebrother: ok, then I think it would be worthwhile to have a clean implementation of scsi in our repository and to only depend on it to avoid extra dependencies. another task for Mr. Someone
22:41:36[Saint]Dex__: off-topic
22:42:31bluebrotherpamaury: yes :)
22:42:59*bluebrother knows Zilog only as chip manufacturer *g*
22:43:37pamaury[Saint]: intermediate scaling could probably be very useful to targets like imx233 which can range from 24MHz to 455MHz !
22:43:40prof_wolfffi think we can do it, got very nice info about how clkcon section can be controlled, actually i am running some test on my ipod, i want to send some patches ASAP, but need to rearange all my notes and clean the code
22:43:58Dex__sorry for derailing guys
22:44:01Dex__and gals
22:44:29*[Saint] suddenly remembers who prof_wolfff is
22:44:40[Saint]sorry man, didn't recognize the nick for a while
22:44:57[Saint]"iPod Classic freq scaling guy" :)
22:46:07saratogafor decoding audio with the screen off on ARM9, 24 MHz is actually pretty close to where you want to be
22:46:13prof_wolfff:), i am very interested on powersaving and battery charger, also i got my ipod to record from microphone but need to be boosted, probably it could be done at lower freqs but need to test it
22:46:20[Saint]pamaury: I also thought so for the Classic, and perhaps the Nano2G and Beast
22:47:05[Saint]...everything really. I'm not sure it could hurt.
22:47:14pamauryI don't know how hard it would be to have the core move from the boost/unboost model to something more scalable. However the low-level set_cpu_frequency interface stays the same so it's mainly a target independent issue at least
22:47:48saratogai thought about allowing codecs to signal to the core how much CPU time they're likely to need
22:48:27saratogalike allowing flac to send BOOST_LOW and ask for the lowest possible clock since it needs to little power, then if a driver can support multiple boost speeds it knows to use the lowest one
22:48:32pamaurydynion: still there ?
22:48:54[Saint]Classic needs 0.08MHz for flac8 realtime :)
22:49:33[Saint]It can play all audio formats I use without boosting at all.
22:49:36saratogayou could generalize that to other parts of the core too, for instance allow the WPS code to request a higher minimum boost frequency if complex tags are in use
22:49:57prof_wolfffi think really we can do freq scaling on CPU, AHB, APB using diferent divisor on each clock
22:50:44saratogayeah, on AMS the memory divider actually has a big impact on both performance and power use
22:50:49[Saint]baby steps :)
22:51:36dynionhere i am, pamaury
22:52:16[Saint]saratoga: JdGordon will correct me I'm sure, but I _think_ the only time the skin engine needs to run boosted is when drawing peak meter(s).
22:52:27[Saint]I think everything else is reasonaly low demand.
22:52:31saratogaah ok
22:52:42saratogaprobably easier to just boost whenever the screen is on anyway
22:52:55[Saint]And, iirc, it already boosts when peak meters are in display.
22:52:56saratogaso that there is no studdering when the clock changes
22:53:10pamaurydynion: I built our tool for windows and sent you the link a private message, tell me when you have it so I can guide you
22:53:25JdGordon[Saint]: pretty much... skin engine isnt very cpu intensive
22:53:50JdGordonwell, it hasnt been benchmarked, but it shold not need boosting (I'd call it a bug if it did)
22:54:23[Saint]am I right that peak meters in view trigger boosting?
22:54:37[Saint]can't find the source relevant to it.
22:54:44JdGordoni dont even think peak meters boosts
22:54:50prof_wolfffone aproach is each task informing about the task it is running, then each target depending on what is running at each time decide the freq for CPU/bus/...
22:54:51JdGordonit ups the refresh rate to 25Hz though
22:54:53[Saint]Hum. Ok, awesome.
22:56:18[Saint]Intermediate scaling levels has been in the back of my mind for about 4 years now :)
22:56:37[Saint]No idea why it took me so long to ask. I guess I assumed it would be there, were it possible.
22:56:39[Saint]Bad mistake.
22:57:03JdGordonisnt the normal way to do scaling, having an idle task which entirely controls the scaling if it gets starved cpu time?
22:57:24[Saint]that's pretty much ondemand, yeah.
22:58:10[Saint]if it gets starved, trigger up a step and chek if it is still staving or not and downscale/upscale accordingly.
23:00:50saratogaprof_wolfff: probably you just need to test the performance (using test_codec) for a couple different settings, and maybe also do a battery bench at a few settings
23:01:15saratogai did this on the AMS devices, and from a performance/battery standpoint, there are optimal settings you can use
23:01:52 Quit Rower85 (Quit: Hmmm...)
23:02:09prof_wolfffplaying already decodec music required no CPU cycles, DMA does it alone
23:06:27*[Saint] wonders if this batter bench will beat 40 hours
23:06:37[Saint]("Thick" Classic)
23:06:50[Saint]*battery too
23:06:57prof_wolfffactually Classic can decode mp3@320 at 36MHz, but i think there are other codecs that requieres more cycles, the idea is to decode at full speed (5-10 seconds of music) and then do playback at a very low speed (13.5MHz or less???)
23:07:04[Saint]38 hours and counting, with ~18% battery left
23:07:33bertrik_JdGordon: I think I suggested that once, but basically got told it's been tried and rejected
23:08:41[Saint]prof_wolfff: those numbers match my findings, yes.
23:09:14[Saint]about 36MHz to decode CBR mp3@320, about 14MHz for realtime
23:09:35prof_wolfffi have access to a 160 thin Classic, did a test on it, SVN version: 36h, running at 36MHz: 42.5h
23:10:07[Saint]Do you have a patch somewhere to lower min_freq?
23:10:15prof_wolfffcontinuous playback -35dB, 32ohm headphones inserted
23:10:17[Saint]iirc, it's ususally 54MHz is it not?
23:13:08saratogaprof_wolfff: if you have time, you could run the test files we have on the website through test_codec and update this page with Ipod6G results:
23:13:27saratogayou can download the entire test file collection here:
23:13:32*[Saint] did a test_codec run he needs to upload.
23:13:40[Saint]but I need to strip out the failed results.
23:14:00[Saint]"wmapro == 4000000% realtime" :-S
23:14:24[Saint]whilst producing white noise that physically hurts.
23:14:29[Saint]even with volume at mute.
23:15:25[Saint]at least one of the vorbis files failed too.
23:15:38[Saint]need to wait for the So to get home before I can poke at that again.
23:15:39prof_wolfffyes, SVN version runs 54/54 CPU/AHB, my patched version runs @ 36/18, voltage is also reduced from 1050->1000 mV
23:16:15[Saint]prof_wolfff: on gerrit?
23:16:16saratogaah nice
23:16:32saratogahow low can voltage safely go when at low clock speed?
23:16:51[Saint]some Classics don't like the undervolting at all.
23:16:55prof_wolfffi want to send a patch to gerrit for revision/testing but need to recode the thing, to reach this settings it should be done on various steps, actually my code is a ugly patch
23:20:23saratogasounds good
23:20:26pamaurymost devices don't really like being undervolted w.r.t clock speed. On imx233 this immediately locks the chip.
23:20:39prof_wolfffprobably we can decrease voltage as we decrease freq., i run my ipod80 @ 36/18/1000 for 3-4 months with no problems, normal use, no extensive testing
23:21:22[Saint]I recall emCORE did some slight undervolting, and some users got into unfortunate situations
23:21:54[Saint]Some Classics handled it, others didn't. (from memory)
23:22:09 Quit alexbobp (Read error: Operation timed out)
23:22:48prof_wolfffactually 54/54/1050mV seems to go ok for everyone, no issues til now
23:24:12prof_wolfffsaratoga: will try to do the test_codec and send the results
23:24:42 Join alexbobp [0] (
23:24:52saratogaprof_wolfff: i suggest not including the slower APE files
23:25:00saratogawill go MUCH faster and no one cares about them since they won't be real time
23:27:47prof_wolfffi have my classic highly patched, my idea is to create a bunch of patches and send it to gerrit if the are "dangerous" modifications or proposed stuff (freq scaling, dma, recording), and directly commit to git the more inocuous code
23:28:21 Quit einhirn (Quit: Miranda IM! Smaller, Faster, Easier.
23:29:00[Saint]You enabled recording?
23:29:00[Saint]You little ledgend.
23:29:12[Saint]Get the sh*t on gerrit, now! :)
23:29:45*[Saint] would also like to see the DMA patch
23:30:01prof_wolfffi can record at full speed, but thing it could be done unboosted
23:30:30kugelJdGordon: yes, but there shouldnt be any handles
23:30:44kugelsince playback is stopped
23:31:13kugelhm, but one thing could be problematic
23:31:35kugelthe fms calls audio_get_buffer() before, without also grabbing the voice buffer
23:31:57prof_wolfffi rewrite an old dma queue driver to work on RB, actually it works with playback and lcd
23:32:12[Saint]gerrit, now! :)
23:32:15kugelI think buffering_reset() allocs a n ew voice buffer
23:33:12 Quit shamus (Ping timeout: 255 seconds)
23:33:13prof_wolfffneed to do lots of cleanup and testing on some areas but i promise i will send all stuff soon
23:33:36 Join shamus [0] (
23:35:29prof_wolfffi also have very interesting info about how ipod HW behaves, i will be nice is this info is published on the web or somewhere, maybe it can help other people who wants to develop on classic
23:37:31 Quit Guest94269 (Read error: Connection reset by peer)
23:37:59kugelno, nevermind
23:41:50kugelJdGordon: you could try just remoing the buffering_reset call(), it's probably unecessary anyway
23:43:37JdGordonkugel: buffering_Reset() nukes every buflib handle no?
23:43:41JdGordonnot just audio
23:43:59kugelno, just the audio buffer
23:44:18saratogaprof_wolfff: boosting while recording is probably ok
23:44:41saratogai would commit it with boosting and then if you want fix it to work without boosting later
23:45:00saratogaprobably though someone recording won't care too much about battery life anyway
23:48:52saratogaBagder, Zagor: could one of you upload the contents of this folder to the test_files folder on the website:
23:49:07kugelJdGordon: you could also try changing the first paramter of audio_get_buffer() to true
23:49:49kugelit means to also grab the talk buffer, so if the bug has anything to do with talk it should make a difference
23:49:56mobenJdGordon: got some time? this passes checkwps and sim gives no error, but it doesn't show anything:
23:50:17prof_wolfffsaratoga: actually RB unboost while recording so we need to disable it or allow recording unboosted, i think later is possible, another issue is that on ipod160 the micro on headphones jack should be activated throught the "mikey" controller (apple peripherical id chip), i managed to do it but need todo some more testing (i have access to a ipod160 but not all the time)
23:50:26JdGordonmoben: not really :/
23:51:01saratogaprof_wolfff: worst case you can put an #ifdef IPOD_6G in the recording code or in the driver
23:51:07saratogai believe some other targets do this
23:51:11kugel(grabbing the talk buffer obviously means talk support is disabled)
23:51:11mobenokay, I'm in no hurry
23:51:20JdGordonI'll try to look tonight
23:51:45saratogaprof_wolfff: by the way, usually its a good idea to keep uploading things to gerrit as you do them so that other people know what your'e working on :)
23:51:46prof_wolfffwe can even read the remote button events using sennheiser px100ii-i (includes micro + remote)
23:52:00saratogacool, some people were asking about that the other day
23:53:16prof_wolfffknow that but these couldnt be considered patches, maybe test code, many things are not functional at all and need to be finished
23:55:35saratogaprof_wolfff: things don't need to be finished to put them on gerrit, i've uploaded lots of things that don't even compile just so that I don't lose them :)
23:55:52[Saint]prof_wolfff: just so you know, as long as my partner hasn't stolen it for work, I have access to a 160Gb "thick" Classic
23:57:53prof_wolfffthats true!, i was really busy, lots of travelling, but had time to look at OF and latter try to test some discoverings, actually my notes/code is arround my main computer, a couple of laptops and half a dozen of pendrives and external hd, need to recompile all

Previous day | Next day