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

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

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

#rockbox log for 2010-04-18

00:00:10 Quit lpereira (Quit: Leaving.)
00:03:47 Join esperegu [0] (~quassel@
00:09:21 Nick guymann_ is now known as guymann (
00:13:45 Join funman [0] (
00:13:49 Quit funman (Changing host)
00:13:50 Join funman [0] (~fun@rockbox/developer/funman)
00:16:24 Join TillW [0] (
00:21:25 Quit bertrik (Quit: De groeten)
00:28:10 Part Llorean
00:35:34 Quit tchan (Quit: WeeChat
00:40:09 Join nicknamedave [0] (
00:41:08nicknamedaveI just read through the manual and there are still several things I don't understand. Starting simply: Which program should I download for compatibility? I noticed several, and I'm not sure exactly which one is, well, best.
00:41:18 Quit dantje (Quit: Ex-Chat)
00:43:11 Quit merbanan (Ping timeout: 246 seconds)
00:44:33funmantry them all and decide for yourself?
00:45:18nicknamedaveSounds like fun
00:48:47 Quit nicknamedave (Quit: CGI:IRC (EOF))
00:54:57soapqtscrobbler is the best. Moving on... ;)
00:57:26 Quit liar (Read error: Connection reset by peer)
01:07:48 Quit dirtyfrank (Quit: CGI:IRC)
01:09:20***Saving seen data "./dancer.seen"
01:13:23 Quit stripwax (Read error: Connection reset by peer)
01:34:09 Quit S_a_i_n_t_ ()
01:41:08 Join evilnick_BS [0] (~48e51a1b@rockbox/staff/evilnick)
01:42:45 Join S_a_i_n_t [0] (S_a_i_n_t@
01:42:53 Quit evilnick_BS (Client Quit)
01:45:55 Quit flydutch (Quit: /* empty */)
01:46:47 Quit dfkt (Read error: Connection reset by peer)
01:58:04 Join simNIX [0] (
01:58:53 Quit TillW (Ping timeout: 252 seconds)
01:59:52 Quit esperegu (Remote host closed the connection)
02:00:27 Quit S_a_i_n_t (Ping timeout: 276 seconds)
02:01:04 Join S_a_i_n_t [0] (S_a_i_n_t@
02:04:26 Quit mapi (Ping timeout: 264 seconds)
02:06:54 Quit Gump (Ping timeout: 268 seconds)
02:12:08 Join Gump [0] (~josh@
02:17:38 Quit Unhelpful (Read error: No route to host)
02:18:47 Join TillW [0] (
02:19:13 Quit Gump (Ping timeout: 252 seconds)
02:21:18 Join mapi [0] (
02:21:48 Quit funman (Quit: free(random());)
02:24:09 Quit planetbeing (Ping timeout: 268 seconds)
02:24:41 Join planetbeing [0] (~planetbei@
02:26:12 Quit petur (Quit: Zzzzz)
02:27:34 Join Unhelpful [0] (
02:27:34 Quit Unhelpful (Changing host)
02:27:34 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
02:32:48 Quit Unhelpful (Remote host closed the connection)
02:33:17 Quit kramer3d (Ping timeout: 245 seconds)
02:34:35 Quit efyx (Remote host closed the connection)
02:36:32 Quit anewuser (Ping timeout: 246 seconds)
02:43:48 Quit n17ikh (Ping timeout: 265 seconds)
02:46:18 Join detaos [0] (
02:46:50 Quit detaos (Client Quit)
02:47:08 Join detaos [0] (
02:49:00 Join n17ikh [0] (
02:49:02 Join Gump [0] (~josh@
02:53:31 Quit detaos (Quit: - Chat comfortably. Anywhere.)
03:01:38 Join hamish_ [0] (
03:02:16 Part hamish_
03:02:38 Join hamish_ [0] (
03:07:00 Quit Schmogel (Quit: Miranda IM! Smaller, Faster, Easier.
03:09:23***Saving seen data "./dancer.seen"
03:12:44 Join detaos [0] (
03:20:53 Quit hamish_ (Quit: Ex-Chat)
03:21:16 Join aim [0] (
03:29:23 Join Unhelpful [0] (
03:29:23 Quit Unhelpful (Changing host)
03:29:23 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
03:30:41 Part aim ("Ex-Chat")
03:30:52 Join aim [0] (
03:31:11 Quit Unhelpful (Remote host closed the connection)
03:33:08 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
03:36:14 Quit Unhelpful (Remote host closed the connection)
03:38:15 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
03:51:00 Quit Unhelpful (Ping timeout: 260 seconds)
03:57:21 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
04:05:01 Join dys` [0] (
04:07:19 Quit dys (Ping timeout: 276 seconds)
04:11:02 Join Barahir_ [0] (
04:13:26 Join anewuser [0] (anewuser@unaffiliated/anewuser)
04:14:36 Quit Barahir (Ping timeout: 258 seconds)
04:19:04 Join Rob2223 [0] (
04:22:35 Quit Rob2222 (Ping timeout: 245 seconds)
04:22:41 Quit adnyxo (Ping timeout: 276 seconds)
04:26:31 Quit ollebe (Quit: Leaving)
04:33:50saratogahuh the new clip+ firmware from sandisk has an "mp4decoder" block added
04:40:43 Quit shaggy-h (Ping timeout: 240 seconds)
04:49:58saratogaif this is really an AAC decoder, its kind of interesting, tons of fixed point multiplies but they're all 32 bit x 32 = 32 bit, no long multiplies
04:50:09 Quit Rondom (Disconnected by services)
04:50:24 Join Rondom_ [0] (
04:50:38saratogai guess AAC is big enough that you could go through all that effort to figure out how to do it all with low precision and still given ok sounding results
04:54:46 Quit Llorean (Ping timeout: 276 seconds)
05:09:04 Quit krazykit (Ping timeout: 260 seconds)
05:09:25***Saving seen data "./dancer.seen"
05:09:59 Quit GeekShad0w (Read error: Connection reset by peer)
05:13:43 Quit saratoga (Quit: Page closed)
05:51:00 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
05:56:33 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
06:07:48 Quit Unhelpful (Remote host closed the connection)
06:10:27 Join ansuz [0] (
06:11:00 Quit anewuser (Quit: What do you know...THE WORLD'S first NTRQ (that's for NES/FAMICOM) tracking compo. Have powerpak? Try it out! Otherwise ROM IMAGE.)
06:13:17 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
06:23:52 Join JdGordon_ [0] (~jd@
06:23:58 Quit JdGordon_ (Changing host)
06:23:58 Join JdGordon_ [0] (~jd@rockbox/developer/JdGordon)
06:32:09 Quit JdGordon_ (Ping timeout: 248 seconds)
06:52:20 Quit Unhelpful (Remote host closed the connection)
06:57:51 Join Unhelpful [0] (
06:57:51 Quit Unhelpful (Changing host)
06:57:51 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
06:57:59 Quit panni_ (Quit: ( :: NoNameScript 3.81 :: ))
07:09:26***Saving seen data "./dancer.seen"
07:13:30 Quit Horschti (Ping timeout: 268 seconds)
07:16:43 Quit saratogalab (Quit: Page closed)
07:18:46 Join Horschti [0] (~Horscht2@xbmc/user/horscht)
07:43:19 Quit antil33t ()
07:55:27 Quit mc2739 (Ping timeout: 276 seconds)
07:55:42 Join krazykit [0] (
08:14:52 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
08:32:55 Join esperegu [0] (~quassel@
08:33:18 Join mikroflops [0] (
08:34:04S_a_i_n_t its closed now, but S_a_i_n_t wonders what this guy ( was on about...
08:34:25S_a_i_n_tSelect+Play to shut down a Nano using the current build...huh?
08:34:56 Quit _arbingordon (Quit: `)
08:35:08S_a_i_n_the calls it "the typical button combination"
08:36:39 Join stoffel [0] (
08:37:42 Quit mikroflops_ (Ping timeout: 276 seconds)
08:38:33 Join petur [0] (~petur@rockbox/developer/petur)
08:39:39 Quit bmbl (Read error: Connection reset by peer)
08:40:07 Quit Rob2223 (Ping timeout: 265 seconds)
08:43:12 Join Rob2222 [0] (
08:43:43 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
08:46:27 Quit amiconn (Disconnected by services)
08:46:28 Join amiconn_ [0] (quassel@rockbox/developer/amiconn)
08:46:50 Nick amiconn_ is now known as amiconn (quassel@rockbox/developer/amiconn)
08:47:56 Quit pixelma (Disconnected by services)
08:47:58 Join pixelma_ [0] (quassel@rockbox/staff/pixelma)
08:48:21 Nick pixelma_ is now known as pixelma (quassel@rockbox/staff/pixelma)
08:53:54JdGordonS_a_i_n_t: preemptive ping?
08:54:04JdGordonprob gonna have a patch needing testing in maybe 15min
08:55:13S_a_i_n_tyep, sure
08:56:12JdGordonI'm working on a clean way to make sure only the needed bars are drawn
08:57:12S_a_i_n_tThat *should* fix up %?pv and %bl in one fell swoop.
08:58:59S_a_i_n_twell, I imagine (in theory) that you'd be able to use one bar for -0dB and another for +0dB, then with a mute and a line-level image...the masses are happy ;)
09:00:21S_a_i_n_tbut *hopefully* %pv|blah.bmp|-|-|-|-| will still draw the full range from min/max as it does in SVN now so I won't jave to change my code ;D
09:05:07JdGordonyou shouldnt need any code changes after applying that
09:05:11JdGordonif you've done it right :)
09:05:28JdGordonthat will also fix %pv| inside conditionals
09:07:55S_a_i_n_tso, %pv|blah.bmp|-|-|-|-| (without conditions) will still 'just work'. and now I can use a bar inside coditionals and it knows its for X segment?
09:08:07S_a_i_n_tshould be fun to play with...
09:08:31JdGordonI mean the thing pixelma was winging abnout last night, where %pv was inside a conditional but not a long form bar
09:09:27***Saving seen data "./dancer.seen"
09:10:50S_a_i_n_tso, does the syntax change at
09:11:30S_a_i_n_tI'm not quite sure whats changed, I think I had issues with something different altogether.
09:13:22JdGordon2 fixes, pixelma's thing, and it always drawing the full bar even when you want the bar for the first 10%
09:13:35JdGordon(we need to come up with a better answer for that though :p ) it did fix the thing I had a cry about also.
09:17:02JdGordonI dont have a complicated wps to test it on
09:17:39S_a_i_n_tneither do I...I have to make one ;) I stupidly binned that code that didn;t work last night *facepalm*
09:18:23 Quit CaptainKewl (Remote host closed the connection)
09:24:11 Quit pjm0616 (Read error: Operation timed out)
09:24:15 Quit Battousai (Read error: Operation timed out)
09:24:18 Join Battousai [0] (~bryan@gentoo/developer/battousai)
09:24:29 Join pjm0616 [0] (~user@
09:24:32 Quit shai (Quit: Leaving)
09:31:40S_a_i_n_tJdGordon: If its OK, and there's no immediate rush I may test this patch over the course of the night and get back to you in the morning?
09:32:21S_a_i_n_tI'm halfway there, just kinda busy right now. The wee small AM hours are geberally my time to shine ;)
09:33:36 Join shai [0] (
09:37:01JdGordonno worries
09:37:02JdGordonI might even get round to testing it myself :p
09:38:14 Join stripwax [0] (
09:40:43 Join ender` [0] (
09:41:32 Nick fxb__ is now known as fxb (
09:44:18 Quit simNIX (Quit: Ik ga weg)
09:46:23 Join flydutch [0] (
09:53:17 Join shai_ [0] (
09:54:05 Join Xerion_ [0] (~xerion@
09:54:51 Join jordan`` [0] (
09:55:59 Quit ender` (Quit: Lisp hackers have to be bound (to-do 'it) ...)
09:56:35 Quit shai (*.net *.split)
09:56:36 Quit CGL (*.net *.split)
09:56:36 Quit jordan` (*.net *.split)
09:56:36 Quit skx` (*.net *.split)
09:56:36 Quit Xerion (*.net *.split)
09:56:36 Quit sinthetek (*.net *.split)
09:56:37 Nick Xerion_ is now known as Xerion (~xerion@
09:56:43 Join skx` [0] (
09:57:11 Join xman099 [0] (
09:57:22 Join ender` [0] (
09:57:34 Join CGL [0] (~CGL@
09:59:34pixelmapointing out a bug - that a recent change broke a perfectly valid WPS is "winging"? Whatever that is... :\
10:02:42 Join mt-phone [0] (~mtee@
10:04:50 Quit xman099 (Quit: CGI:IRC (EOF))
10:10:15 Nick shai_ is now known as shai (
10:10:47 Quit aim (Read error: Connection reset by peer)
10:13:16 Join sinthetek [0] (
10:28:56 Quit avn (Ping timeout: 248 seconds)
10:30:42 Join avn [0] (~avn@
10:35:53 Join TheSeven [0] (~theseven@rockbox/developer/TheSeven)
10:44:52 Join DataGhost [0] (
10:44:52 Quit DataGhost (Changing host)
10:44:52 Join DataGhost [0] (~dataghost@unaffiliated/dataghost)
10:53:54 Join emrecelikten [0] (~58f1a5e0@gateway/web/freenode/x-tsgsuenhbbzyqrzz)
10:57:19 Nick dys` is now known as dys (
10:57:46 Join The_Seven [0] (
10:58:22 Join esperegu_ [0] (~quassel@
10:58:22 Quit esperegu_ (Remote host closed the connection)
10:59:17 Quit The_Seven (Remote host closed the connection)
10:59:21 Quit esperegu (Ping timeout: 258 seconds)
10:59:23 Join The_Seven [0] (
10:59:48 Quit The_Seven (Remote host closed the connection)
10:59:53 Join The_Seven [0] (
10:59:58 Quit n17ikh (Read error: Connection reset by peer)
11:00:37 Join n17ikh [0] (
11:01:20 Join esperegu [0] (~quassel@
11:01:44 Quit The_Seven (Remote host closed the connection)
11:04:21 Quit TheSeven (Disconnected by services)
11:04:24 Join The_Seven [0] (~TheSeven@rockbox/developer/TheSeven)
11:04:47 Quit The_Seven (Disconnected by services)
11:04:58 Join TheSeven [0] (~theseven@rockbox/developer/TheSeven)
11:05:16 Quit TheSeven (Client Quit)
11:05:22 Join TheSeven [0] (~TheSeven@rockbox/developer/TheSeven)
11:09:29***Saving seen data "./dancer.seen"
11:12:45 Join Lixun [0] (
11:13:28 Part Lixun
11:17:07 Join mirak [0] (
11:18:05 Quit mirak (Max SendQ exceeded)
11:19:26 Join mirak [0] (
11:20:38 Quit mirak (Max SendQ exceeded)
11:26:19 Quit pjm0616 (Ping timeout: 260 seconds)
11:26:34 Join pjm0616 [0] (~user@
11:28:06 Quit avn (Ping timeout: 258 seconds)
11:29:59 Join avn [0] (~avn@
11:31:24*S_a_i_n_t atemps to go down the "progressbar route" to test this new patch, as opposed to working out a bunch of timy bars for the battery...
11:31:32S_a_i_n_tBattery will be "step two" ;)
11:35:17 Join DerPapst [0] (
11:36:01JdGordonyou could use the same images for both...
11:38:52S_a_i_n_tprogressbar will be easier to see if it works...I can use existing code then.
11:39:06S_a_i_n_tWorking out viewport coords is what pisses me off ;)
11:39:39S_a_i_n_tI'm not far off a test-case, finishing it up now
11:39:53pixelmacan't you use your pastebinned code from yesterfay, or is there too much missing of the rest of the WPS?
11:40:03pixelmayesterday too
11:42:15S_a_i_n_tBlargh...I didn't think of that, ahh..I'm over the worst of it now ;)
11:42:15*JdGordon looks at screenshots in the forums and thinks it would be awesome to be able to use a bmp for the line selector (and/or a bmp for just the outer edge)
11:42:34S_a_i_n_tdidn't I already ask for that?
11:42:57JdGordondid you?
11:43:05S_a_i_n_tMy "other" want is .bmp scrollbar
11:43:09pixelmawasn't the problem with that 2 line heigh selector bars in some screens?
11:43:13S_a_i_n_tand yeah, pretty sure I have.
11:43:32JdGordon2 line is a problem, but meh, almost no screens use that
11:44:03CIA-5New commit by jethead71 (r25666): Threads of PRIORITY_REALTIME and above are no longer subject to being switched away based upon aging of lower priority threads. Now, recommend a ...
11:44:09S_a_i_n_tI've personally *never* seen a 2 line selector bar...huh, live & learn
11:45:33pixelmathe ID3 info screen is an example
11:50:25emreceliktenDoes this make any changes to gcc that is installed at my system, or does it make a separate installation?
11:57:25 Join Luca_s [0] (
12:02:56Tornei hate the font cache :(
12:03:15Tornei can't conceive of a way to fix it to load fast that isn't a huge pain in the ass
12:03:28CIA-5New commit by jethead71 (r25667): Fix red from r25666 (forgot to use macro to not use core variable on single core), update a comment and remove core check for setting THREAD_SWITCH ...
12:03:50JdGordonTorne: use the cache file as an on-disk copy..
12:04:56JdGordonactually, is the order of gylphs specified? it might just make sense to reorder the glyphs in the standard bdf to put the usual ones together
12:05:02JdGordon(that may not make any sense)
12:05:11Tornethe order is least-recently-used
12:05:31Tornewhich is *excceptionally* well randomised, from a loading POV :)
12:06:13Tornedumping the cache to disk as a file means you need one for each font when doing multifont, and you need to fix up some pointers and other foo. it's not impossible.
12:06:34Tornethe way multifont works wrt. the cache is rather unfortunate
12:06:48Tornethe cache is created based on the LRU order of the UI font
12:06:58JdGordonno, skin fonts dont load using the cache
12:07:04Torneyes they do
12:07:11Torneotherwise they wouldnt work
12:07:21Torneyou wouldn't be able to load a font larger than 10KB :)
12:07:32Tornewhich is most of hem.
12:07:36JdGordonI mean they dont check the cache file for which to load
12:07:40Torneyes they do
12:07:48JdGordonok /me shuts up and heads out
12:07:56Tornewhicch means it will overwrite the glyphs a bunch, as a bonus :(
12:08:19Tornesince it will start with the least recently used and work forward
12:08:36Torneyou'll end up with as many of the recently used glyphs as will fit but if the 60000 byte main cache was full.. you've now loaded 6 times as many glyphs as actually fit, rather wastefully
12:08:54Torneso for the skin fonts it's *even worse* :)
12:09:27S_a_i_n_tJdGordon: that patch seems ok...I'm *realy* tired, so I didn't really put it through its paces, but the basic functionality of it seems to work.
12:09:32Torne(same amount of work, but upp to 5/6ths of tit is wasted)
12:09:45 Join pamaury [0] (~c2c7a50a@rockbox/developer/pamaury)
12:10:24S_a_i_n_t...and, it doesn't break using just "%pv|blah.bmp|-|-|-|" either. So, it would *seem* that the masses are happy?
12:11:00S_a_i_n_tI'll put it through its paces some more once I figure out WTF my sims aren't building... >:[
12:11:15S_a_i_n_t(nothing to do with the patch I'm sure...)
12:14:36 Join liar [0] (
12:15:05 Nick liar is now known as Guest52688 (
12:19:22 Join merbanan [0] (
12:20:15S_a_i_n_t If something like multifont can apparently make *that* much difference to HDD target load times...than, I'm *way* happier with flash.
12:20:48S_a_i_n_tSome of my themes are calling 6/7 fonts, with 0 niticable load time increase.
12:22:37Torneit's not multifont specifically
12:23:01Torne*every* font load that doesn't fit loads 5-30 times slower, depending on target :0
12:23:30Tornethe maximum slowdown only appears to happen on ipod video with large drives, though
12:23:35Torneer, large sector
12:24:01*S_a_i_n_t hugs his army of flash targets ;)
12:24:14Torneit's probably 5 times slower on your flash targets, but the otriginal time is fast enough that five times slower is still quite fast
12:24:29Torneit might be, say, 50ms or so per font though
12:27:50 Nick Guest52688 is now known as liar_ (
12:32:39 Join shaggy-h [0] (
12:40:01 Quit stoffel (Ping timeout: 246 seconds)
12:45:32 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
12:47:26 Join |DaMaGeD| [0] (~|DaMaGeD|@
12:53:58 Quit |DaMaGeD| (Ping timeout: 258 seconds)
12:59:19 Quit mt-phone (Ping timeout: 248 seconds)
13:08:25 Join emrecelikten_ [0] (~anubis@
13:09:30***Saving seen data "./dancer.seen"
13:16:18 Join stoffel [0] (
13:16:18 Quit stoffel (Remote host closed the connection)
13:16:38 Join stoffel [0] (
13:28:36 Join mirak [0] (
13:30:00 Nick liar_ is now known as liar (
13:36:41 Quit stoffel (Ping timeout: 245 seconds)
13:37:54 Quit mirak (Quit: Ex-Chat)
13:39:36 Quit n17ikh (Read error: Connection reset by peer)
13:39:57 Join n17ikh [0] (
13:45:51 Join kugel [0] (~kugel@rockbox/developer/kugel)
13:46:07 Join mt-phone [0] (~mtee@
13:52:30 Quit bertrik (Ping timeout: 246 seconds)
13:54:34 Join bertrik [0] (~bertrik@rockbox/developer/bertrik)
13:57:22 Join Kitr88 [0] (
14:00:26 Quit Kitar|st (Ping timeout: 245 seconds)
14:02:25 Quit Kitr88 (Ping timeout: 276 seconds)
14:02:37S_a_i_n_tCan someone please revert 25665? It seemed like a good idea at the time...but it only makes things worse.
14:05:02gevaertspreferably someone who understands the issues
14:05:06*gevaerts will not touch that
14:05:56 Join Kitar|st [0] (
14:08:34 Join MethoS- [0] (~clemens@
14:12:50S_a_i_n_tit was a keybind which was supposed to be "better", but instaed, it just made it worse.
14:13:33pixelmaS_a_i_n_t: how does it make things worse? It makes the combo work when you press play first then the other key
14:13:43pixelma(as you said you would prefer)
14:14:00S_a_i_n_tthe idea sounded good, but, it just doesn't work...
14:14:43S_a_i_n_tit's kinda hard to describe, but only the play part gets registered, you never get a chance to actually hit select.
14:15:23S_a_i_n_tas, hitting play in the context menu drops straight back to the WPS
14:15:28S_a_i_n_tso, it doesn;t work.
14:15:38S_a_i_n_tI should have thought of that at the time.
14:17:38S_a_i_n_tit works a lot better to *use* the hotkey, but it's not impossibly to set one.
14:17:49S_a_i_n_tnot? *now even
14:17:55pixelmathat's probably the "cancel" action in the linked standard context that is hit then
14:17:57S_a_i_n_tand *impossible
14:19:25pixelmayou could try if it gets better if you change the BUTTON_NONE on line 54 in the keymap file to BUTTON_PLAY too
14:20:00pixelmanot sure at the moment though
14:20:18 Quit MethoS- (Remote host closed the connection)
14:20:57pixelmait could also hit the "tree stop" action then
14:21:54S_a_i_n_tit doesn't seem to work to set the hotkey in either the file tree/wps context least with the old keymap, it was just hard to set (yet worked), in this keymap it is easy to use, yet *impossible* to set. :/
14:22:12 Join MethoS- [0] (~clemens@
14:27:45kugelS_a_i_n_t: maybe it should be "press select first, then play"?
14:27:52 Join Lear [0] (chatzilla@rockbox/developer/lear)
14:27:52 Join stoffel [0] (
14:28:09kugel(a combo which works well on e200/fuze
14:29:56S_a_i_n_tkugel: yes, that should wouk actually...long select doesn't seem to trigger anything.
14:30:07S_a_i_n_t*in the context menus
14:30:20S_a_i_n_tand work even
14:31:07S_a_i_n_tthe initial problem was having to press two buttons (still select and play) at *exactly* the same time.
14:31:39pixelmaisn't the hotkey also (and primarily) used from the WPS? In which case long select triggers the context menu, if I see correctly
14:32:09S_a_i_n_tahhhh...good point.
14:32:18S_a_i_n_twhere were you last night pixelma ;)
14:32:29S_a_i_n_tso, it guess it *has* to be the way it is...
14:32:38pixelmaso changing this, probably needs some "guarding" too
14:32:39 Quit MethoS- (Remote host closed the connection)
14:32:42S_a_i_n_terr...the way it *was*
14:33:02pixelmaI guess it's possible but needs a bit more thinking and trying ;)
14:33:06S_a_i_n_tpixelma: guarding?
14:33:32kugellong select+up works good on e200/fuze even in the wps
14:33:57pixelmain the keymap file - making sure that no other action is triggered before the one you want
14:34:38S_a_i_n_twell, I could only but try it (long menu+play) I mean
14:34:51S_a_i_n_tif it doesn;t work, it doesn't work I guess...
14:35:12pixelmakugel: don't forget that the Ipods have a few buttons less though :)
14:35:19 Join MethoS- [0] (~clemens@
14:35:28 Quit pamaury (Quit: Page closed)
14:36:14S_a_i_n_tthat's the problem initailly...the fact that the ipod keymaps are so cramped up.
14:36:20S_a_i_n_t(because they have to be)
14:40:08 Join emrecelikten1 [0] (~anubis@
14:41:40 Quit emrecelikten_ (Ping timeout: 276 seconds)
14:48:48pixelmathe problems due to "overlapping" actions and contexts can even be seen in a simulator, so it would be nice if people checked before such a commit
14:49:19 Quit stoffel (Remote host closed the connection)
14:49:29S_a_i_n_tnow, since when do people do that?
14:49:43 Join anewuser [0] (anewuser@unaffiliated/anewuser)
14:49:56pixelmaseems to be a dying art :(
14:50:01 Join efyx [0] (
14:50:13gevaertsI'd assume that people who suggest a keymap change have tested it
14:50:19S_a_i_n_tcommit now, fix later seems to be the style of the time ;)
14:50:57kugelgevaerts: me too
14:51:12S_a_i_n_twe know what assumption gets then, don't we
14:51:20*pixelma is "old school" and usually tests, especially things like that
14:52:21gevaertsS_a_i_n_t: yes, it gets us to seriously consider just ignoring certain people
14:53:10S_a_i_n_tsaying "yes, that should work" is hardly saying "I've looked at the keymap and considered it carefully, please run off and commit this"
14:53:24S_a_i_n_tbesides, it was Blue_Dudes suggestion, not mine.
14:53:39 Quit shaggy-h (Ping timeout: 240 seconds)
14:54:01S_a_i_n_tHim asking, can you test this patch (or not asking in fact) is now my fault?
14:56:19 Quit bmbl (Quit: Bye!)
14:56:24pixelmamy complaint was more directed at Blue_Dude (and some other that show this attitude)
14:56:37 Join EK2 [0] (
14:57:05pixelma"commit and let the users be the testers2
14:58:56EK2Hi Guys, I am new to rockbox and I am trying to compile it for my sansa clip+
15:00:52EK2When I call make I get he following error "/bin/sh: /home/eylonk/rockbox/tools/ Permission denied" followed by other errors.
15:01:34EK2I see that all the .pl file are not with the executable permission, should I change them all?
15:02:21bertrikhm, I would have expected that this file already had executable permission when checked out from svn
15:02:37gevaertsEK2: that file should be exectutable already. How did you check out?
15:03:01bertrikproperty svn:executable is set on that file
15:03:22EK2I have checkedout at home using svn and ziped it and extracted at work.
15:04:07gevaertsyour zip tool doesn't preserve executable bits then...
15:04:49gevaertspossibly svn revert might fix it
15:05:04EK2If it is only few files I can do it manually...
15:06:27gevaertsyou can try
15:06:38*gevaerts doesn't know how many files are involved
15:07:08 Join shaggy-h [0] (
15:07:23EK2Thanks guys
15:07:59 Quit EK2 (Quit: CGI:IRC (EOF))
15:08:13 Join Strife89 [0] (
15:09:32***Saving seen data "./dancer.seen"
15:18:05 Quit Lear (Quit: ChatZilla 0.9.86 [Firefox 3.6.4/20100413172113])
15:20:52 Quit Strife89 (Quit: Leaving)
15:21:16 Join Strife89 [0] (
15:22:45CIA-5New commit by Blue_Dude (r25668): Revert r25665 - ipod keymap. More trouble than its worth
15:23:22 Join Blue_Dude [0] (~chatzilla@rockbox/developer/Blue-Dude)
15:23:33Blue_DudeIpod keymap patch at FS #11208.
15:23:36 Quit Blue_Dude (Client Quit)
15:25:47 Quit Strife89 (Ping timeout: 252 seconds)
15:27:37kugelthat patch doesn't appear to change anything
15:32:30S_a_i_n_tno, it doesn't...does it.
15:45:03 Join phanboy4 [0] (
15:46:11 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
15:46:45 Quit Luca_s (Quit: CGI:IRC (EOF))
15:47:53 Nick YPSY is now known as Ypsy (
15:51:48bertrikWasn't there a general rule that stuff in firmware/ should not refer to stuff in apps/ ?
15:52:51 Quit rasher (Read error: Connection reset by peer)
15:54:37 Join panni_ [0] (
15:56:40 Join rasher [0] (
15:56:40 Quit rasher (Changing host)
15:56:40 Join rasher [0] (~rasher@rockbox/developer/rasher)
15:59:36kugelbertrik: did someone break that rule?
15:59:42bertrikI'm getting a bit confused by apps/playback.c, some functions from there are declared in firmware/export/audio.h and others in apps/playback.h
16:00:25bertrikso *firmware*/export/audio.h refers to functions implemented in *apps*/playback.c
16:00:50kugelI noticed that too, I think they should be moved to playback.h
16:01:29 Quit mc2739 (Ping timeout: 264 seconds)
16:01:52 Join adnyxo [0] (
16:08:08 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
16:22:54 Quit mc2739 (Ping timeout: 268 seconds)
16:23:30 Join GeekShadow [0] (~Antoine@reactos/tester/GeekShadow)
16:24:34 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
16:28:39 Join mirak [0] (
16:29:21 Quit liar (Quit: Verlassend)
16:29:42 Nick fxb is now known as fxb__ (
16:35:06 Quit phanboy4 (Quit: Leaving)
16:36:39 Quit Unhelpful (Remote host closed the connection)
16:43:43 Join Unhelpful [0] (
16:43:43 Quit Unhelpful (Changing host)
16:43:43 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
16:49:27 Quit mc2739 (Ping timeout: 276 seconds)
16:50:25 Quit xiainx (Ping timeout: 245 seconds)
16:50:52 Quit mirak (Ping timeout: 260 seconds)
16:51:06 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
16:52:27 Join Lixun [0] (
16:54:42 Join saratoga [0] (~463f90ed@rockbox/developer/saratoga)
16:55:17Lixunhi, here's Lixun. I am ready for GSOC interview.
16:55:48 Quit Unhelpful (Read error: Connection reset by peer)
16:56:47gevaertsLixun: hello! We'll be ready in a few minutes
16:56:56Lixunok. no problem
16:57:59 Quit mc2739 (Ping timeout: 240 seconds)
16:58:05 Join |DaMaGeD| [0] (~|DaMaGeD|@
16:58:27 Join Kitr88 [0] (~Kitar_st@
16:59:12CIA-5New commit by alle (r25669): Add the guard where it should be used. The net effect is zero now but it might be not in the future if the definitions are changed.
16:59:45CIA-5New commit by bertrik (r25670): Move audio_peek_track declaration from skin_display.c to audio.h
17:00:08 Quit Kitar|st (Ping timeout: 240 seconds)
17:00:40 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
17:01:11 Join Unhelpful [0] (
17:01:12 Quit Unhelpful (Changing host)
17:01:12 Join Unhelpful [0] (~quassel@rockbox/developer/Unhelpful)
17:02:38 Quit Kitr88 (Ping timeout: 240 seconds)
17:03:33 Quit Lixun ()
17:03:38 Join Lixun [0] (
17:06:00 Join xiainx [0] (~iain@wpa106056.Wireless.McGill.CA)
17:07:11 Join mirak [0] (
17:07:40 Join Kitar|st [0] (
17:08:32 Quit mc2739 (Ping timeout: 268 seconds)
17:09:33***Saving seen data "./dancer.seen"
17:11:35 Nick |DaMaGeD| is now known as antimatter (~|DaMaGeD|@
17:11:58 Quit kugel (Ping timeout: 246 seconds)
17:12:53 Quit mirak (Max SendQ exceeded)
17:15:36 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
17:19:19 Join liar [0] (
17:19:47 Nick liar is now known as Guest62623 (
17:21:22 Quit Guest62623 (Client Quit)
17:21:42 Join liar_ [0] (
17:23:54 Quit mc2739 (Remote host closed the connection)
17:24:09 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
17:24:42CIA-5New commit by bertrik (r25671): Use boolean instead of int for keeping track of mutex signalled state and wakeup signalled state
17:25:38 Quit GeekShadow (Quit: The cake is a lie !)
17:25:56 Join kugel [0] (~kugel@rockbox/developer/kugel)
17:27:15 Quit xiainx (Ping timeout: 258 seconds)
17:32:11 Quit mc2739 (Ping timeout: 252 seconds)
17:34:03 Join mc2739 [0] (~mc2739@rockbox/developer/mc2739)
17:35:23 Join xiainx [0] (~iain@wpa106056.Wireless.McGill.CA)
17:35:36 Quit avn (Ping timeout: 276 seconds)
17:37:14 Join avn [0] (~avn@
17:38:13 Quit mc2739 (Ping timeout: 246 seconds)
17:39:26 Join |DaMaGeD| [0] (~|DaMaGeD|@
17:40:51 Quit antimatter (Read error: Connection reset by peer)
17:41:57emrecelikten1Emre here, ready for the interview.
17:42:45gevaertsemrecelikten1: hello! We're still busy, I'll contact you when we're ready, probably in about 20 minutes
17:47:40CIA-5New commit by bertrik (r25672): Clean up IAP / ipod remote tuner a bit.
17:50:04 Join mikroflops_ [0] (
17:52:53CIA-5New commit by bertrik (r25673): Correct svn:keywords on firmware/driver/tuner/ipod_remote_tuner.c
17:54:26 Quit mikroflops (Ping timeout: 265 seconds)
17:56:59 Quit adnyxo (Quit: Leaving)
17:57:12 Join adnyxo [0] (
17:57:31 Join robin0800 [0] (
17:58:06saratogaLixun: while we can't give you training exactly, you are welcome to spend time in this channel and talk to us about audio programming
17:58:19 Part Lixun
18:01:40 Quit jordan`` (Read error: No route to host)
18:02:01 Join mirak [0] (
18:02:41 Quit mirak (Max SendQ exceeded)
18:03:29 Join mirak [0] (
18:05:12 Quit xiainx (Ping timeout: 258 seconds)
18:05:33 Quit mirak (Max SendQ exceeded)
18:06:16 Join mirak [0] (
18:07:11 Quit mirak (Max SendQ exceeded)
18:08:01 Join mirak [0] (
18:12:53 Quit mirak (Quit: Ex-Chat)
18:14:01 Join jordan` [0] (
18:17:17 Join xiainx [0] (~iain@wpa071038.Wireless.McGill.CA)
18:23:10 Join Strife1989 [0] (
18:23:54 Quit mt-phone (Ping timeout: 265 seconds)
18:24:35 Join M3DLG [0] (
18:25:05 Part |DaMaGeD| ("Reality is that which, when you stop believing in it, doesn't go away")
18:25:18 Join |DaMaGeD| [0] (~|DaMaGeD|@
18:26:08 Quit shaggy-h (Ping timeout: 240 seconds)
18:27:15 Nick Strife1989 is now known as Strife89 (
18:30:39 Join shaggy-h [0] (
18:30:53 Join JohannesSM64 [0] (
18:33:08 Quit Strife89 (Ping timeout: 240 seconds)
18:34:01 Join Strife89 [0] (
18:36:19 Join mc2739_ [0] (~mc2739@rockbox/developer/mc2739)
18:36:48 Nick mc2739_ is now known as mc2739 (~mc2739@rockbox/developer/mc2739)
18:36:54 Quit TillW (Quit: this concludes our broadcast day)
18:44:29 Quit mc2739 (Ping timeout: 276 seconds)
18:54:33 Join kugel_ [0] (
18:54:49 Quit kugel (Disconnected by services)
18:54:53 Nick kugel_ is now known as kugel (
18:54:57 Quit kugel (Changing host)
18:54:57 Join kugel [0] (~kugel@rockbox/developer/kugel)
18:58:35CIA-5New commit by bertrik (r25674): Sansa as3525: Add missing lcd-clip.h header file to sansa clip lcd drivers. Set svn:keywords on scrollwheel-as3525.c.
18:59:32 Nick liar_ is now known as liar (
19:01:40 Join BHSPitMonkey [0] (~stephen@unaffiliated/bhspitmonkey)
19:07:10 Quit saratoga (Ping timeout: 248 seconds)
19:09:19 Join emrecelikten2 [0] (~anubis@
19:09:35***Saving seen data "./dancer.seen"
19:09:50 Quit emrecelikten (Ping timeout: 248 seconds)
19:10:50 Nick emrecelikten2 is now known as emrecelikten (~anubis@
19:10:51 Quit emrecelikten1 (Ping timeout: 260 seconds)
19:13:12 Quit |DaMaGeD| (Quit: Errection killed by beer.)
19:19:17 Quit CGL (Quit: Saliendo)
19:21:48CIA-5New commit by bertrik (r25675): AS3525: set more svn properties
19:32:40 Join CGL [0] (~CGL@
19:34:01 Join Blue_Dude [0] (~chatzilla@rockbox/developer/Blue-Dude)
19:35:12Blue_Dudebertrik: The problem with moving the audio.h declarations to playback.c is that those functions are supposed to work for both swcodec and hwcodec targets. Sometimes the audio.h functions point to playback.c and sometimes to mpeg.c. So maybe move parts of audio.h to apps.
19:37:59 Join Schmogel [0] (
19:42:55 Quit MethoS- (Remote host closed the connection)
19:43:33Blue_Dudekugel, S_a_i_n_t: The purpose of the patch was to have a place to fool around with a keymap for ipods that would work. The first version was in error, since it wasn't synced to the reverted keymap. The next version could work better than the one committed yesterday since it moves the hotkey map further up and could prevent keypress conflicts. There's no way to know until you try it.
19:44:21kugelyou can certainly determine behavior without trying
19:45:02Blue_DudeNot with a sim. Keypresses in the sim don't necessarily mirror the same ergonomics as on the device.
19:45:24Blue_DudeIt can be very hard to get combos pressed the right way on a keyboard.
19:45:36kugelbut the sim shows whether there's a fall through or not
19:46:15Blue_DudeOnly if you can press the combo. And only if the sim reads keys the same way as on target. I don't know if that's true.
19:46:57Blue_DudeSo only way to be sure is for an iPod user to buildit and try it.
19:47:02kugelhave you tried it now or was that change a guess?
19:47:11Blue_DudeI don't have an iPod.
19:47:16kugelin the sim
19:47:32Blue_DudeNo. Pressing that combo in the sim is a bear.
19:47:50kugelhow can you know?
19:48:09kugelI think you should at least give it a try before saying stuff like this
19:48:31Blue_DudeThe sim doesn't seem to clear the keypress when moving to a different screen. So if you don't release the keys immediately it repeats the keypress for another round.
19:48:51Blue_Dudekugel: I didn't want to enable hotkeys for other targets to begin with.
19:49:00Blue_DudeNot until they were tested.
19:49:39kugelthis is currently about one (group of) target
19:49:58Blue_DudeSo under pressure I made a best guess attempt based on existing keypresses and it turns out the stolen keypresses aren't always appropriate. What else can I do at that point than guess?
19:50:36Blue_DudeI can turn off ipods for hotkeys until they are fully tested. Which is what I wanted to do in the first place.
19:50:59kugelI'm not complaining about what you did initially, but I do find it strange that you're uploading that patch for a single target without giving it a try in the sim
19:52:08kugelalso sims are nice for keymap things, I don't understand why you think it's the opposite
19:52:42Blue_DudeI'm not convinced the sim reads keys the same way as the target does.
19:53:11 Quit Strife89 (Quit: Changing OS.)
19:53:11kugelbut you're not even trying
19:53:44pixelmaaction conflict can be seen in the sim, electrical or mechanical restrictions can not
19:53:47Blue_DudeI tried under the current keymap several times. And it's just very hard to time keys correctly.
19:54:07kugelyou seem to be denying the usefulness of sims based upon assumptions, and take that as an excuse for not trying your patches
19:54:54kugelgevaerts: my interview is starting soon?
19:55:08Blue_DudeThe patch was an off the cuff "fix". I even said at the time that keymaps are easy to change, but very difficult to know what to change them to.
19:55:09gevaertsyes, don't worry, we're not forgetting you :)
19:55:12 Quit xiainx (Ping timeout: 258 seconds)
19:55:42Blue_DudeSo should we turn off all non-tested targets now? That probably would be smart.
19:55:58 Join CaptainKewl [0] (
19:56:33gevaertsBlue_Dude: I don't think that's needed. Just don't commit fixes for them if those fixes haven't been tested :)
19:57:42Blue_DudeYou have a problem with me making quick commits. OK, guilty as charged. It won't happen again. But please therefore don't pressure me into committing something much bigger that hasn't been fully tested. Yesterday's "fix" was a much smaller sin in my opinion than switching on nearly every target untested.
19:58:40kugelit seemsyou misunderstood me
19:59:06pixelmaBlue_Dude: I *believe* by the way that it can be disabled for the Iaudio M3 as it suffers the same restrictions as the other Iaudios (that you can't have button combos basically), but would like amiconn as the M3 porter to have a quick look
19:59:20pixelma(something you can't see in the sim ;) )
19:59:46Blue_DudeAnyway, I don't have time to hash it out now. I have to leave for work. I'll check the logs later.
19:59:50 Quit Blue_Dude (Quit: ChatZilla 0.9.86 [Firefox 3.6.3/20100401080539])
19:59:56kugelI don't have a problem with any of your commits, I just find it strange that your patch is completely untested because the amount of work needed to check the combo for a single target sim is very manageable
20:02:11 Join mt-phone [0] (~mtee@
20:02:23 Quit mt-phone (Client Quit)
20:04:20 Join Strife89 [0] (
20:07:07 Join guymann_ [0] (
20:07:15 Quit guymann (Read error: Connection reset by peer)
20:12:34CIA-5New commit by bertrik (r25676): Clean up libmusepack: move declarations to internal.h and set svn properties.
20:16:16 Join Luca_S [0] (
20:25:17 Quit robin0800 (Remote host closed the connection)
20:28:07 Quit adnyxo (Ping timeout: 265 seconds)
20:29:15 Join adnyxo [0] (
20:32:44 Quit adnyxo (Client Quit)
20:32:50 Quit flydutch (Quit: /* empty */)
20:32:54 Join adnyxo [0] (
20:34:09 Join stoffel [0] (
20:37:48 Join funman [0] (~fun@rockbox/developer/funman)
20:50:56 Join xiainx [0] (
20:51:52 Nick Ypsy is now known as YPSY (
20:58:00 Join mirak [0] (
20:58:53 Quit mirak (Max SendQ exceeded)
21:01:16 Join mirak [0] (
21:01:52 Nick YPSY is now known as Ypsy (
21:02:13 Join bmbl [0] (~Miranda@unaffiliated/bmbl)
21:02:39 Quit mirak (Max SendQ exceeded)
21:05:19 Quit CGL (Quit: Saliendo)
21:08:02 Join amr [0] (
21:09:37***Saving seen data "./dancer.seen"
21:10:48 Quit amr (Client Quit)
21:13:22 Join Buschel [0] (
21:38:19 Join Llorean [0] (~DarkkOne@rockbox/user/Llorean)
21:42:22 Nick fxb__ is now known as fxb (
21:42:48 Quit Horschti (Ping timeout: 246 seconds)
21:42:52 Join Horscht [0] (~Horscht2@xbmc/user/horscht)
21:46:21 Quit Schmogel (Quit: Miranda IM! Smaller, Faster, Easier.
21:49:08 Join mirak [0] (
21:52:28 Quit Buschel ()
21:52:55 Join phanboy4 [0] (
21:53:30 Quit bmbl (Quit: Bye!)
21:53:45 Quit JohannesSM64 (Quit: WeeChat 0.3.2-dev)
21:56:13 Join Barahir [0] (
21:56:31 Join ollebe_ [0] (
21:57:37 Quit ollebe_ (Quit: Leaving)
21:59:23 Quit Barahir_ (Ping timeout: 264 seconds)
21:59:24 Join ollebe_ [0] (
21:59:24 Nick ollebe_ is now known as Guest78620 (
21:59:24 Quit BHSPitMonkey (Ping timeout: 265 seconds)
21:59:44 Nick Guest78620 is now known as ollebe__ (
22:00:11 Quit ollebe__ (Client Quit)
22:00:22 Join ollebe__ [0] (
22:00:40 Join Rideless [0] (
22:07:54 Quit stoffel (Remote host closed the connection)
22:10:23 Join robin0800 [0] (
22:12:05 Join wodz [0] (
22:12:42 Join p3tur [0] (~petur@rockbox/developer/petur)
22:13:05 Join plauclair [0] (
22:15:01plauclairdoes Rockbox do m4a/aac on the Fuze ?
22:15:05plauclairor in general
22:15:39Lloreanplauclair: We do generally ask that you make at least a cursory reading of the manual and/or search of the site before asking questions in here.
22:16:20plauclairI did but couldn't find that specific info
22:16:47plauclairfor the fuze, I might have looked a bit fast, sorry about that
22:16:49LloreanFunny, the first thing I find when I search for AAC is a listing of the supported formats.
22:17:32plauclairok, shame on me :)
22:18:13LloreanGenerally, all formats are supported on every device (except some of the old Archoses, or when devices happen to be slower than usual for one reason or another)
22:19:14 Quit mirak (Quit: Ex-Chat)
22:19:36 Join mirak [0] (
22:20:45 Quit mirak (Max SendQ exceeded)
22:21:45 Join mirak [0] (
22:22:57 Quit mirak (Max SendQ exceeded)
22:23:09plauclairI used to have an old D2, I really liked Rockbox on it.. I'm really looking forward to play with the new versions and hopefully maybe make some contributions to the great project :)
22:24:01 Join mirak [0] (
22:26:21 Join TillW [0] (
22:30:27 Join halmi [0] (~netbook@
22:30:45amiconnRegarding audio.h - it's completely okay for apps code to include firmware headers. The reverse is what's forbidden
22:31:11 Quit tchan (Quit: WeeChat
22:31:20*amiconn doesn't like what he's seen so far in the Elio ROM :\\
22:32:24stripwaxamiconn - wanna spill the beans? gpl'd code? names addresses + telephone numbers? :)
22:32:43amiconnA colour target using the PP's mono lcd bridge
22:33:01amiconnThe c200 (v1) also does that - but it's 132x80 while the Elio is 220x176
22:33:14halmigevaerts: I'm available now.
22:33:19amiconnThat means sloooow lcd updates even with optimised asm
22:34:41stripwaxeven slower than ipodvideo? / per pixel? (is ipodvideo slow because of the BCM in the way?)
22:34:55amiconnIpod video is rather fast per pixel
22:35:14amiconnIt's just *a bit* slow overall due to the huge display
22:35:18stripwaxoh! haha, I just assumed it wasn't because of all the complaints. it's just that it has a lot of pixels.
22:35:30amiconnCoarse estimation for the Elio: <10fps
22:36:13amiconnWell, maybe there are some trick yet unknown to me
22:37:03stripwaxwaaaay back when, I recall it being discussed that we're still using effectively 'a debug mode' to update the ipodvideo lcd - what did that mean exactly? (just that we're pumping raw data to the BCM, rather than DMA or something BCM-specific?)
22:38:09stripwax(which I think originally came from the fact that our lcd driver came from the ipod's debug/test app, rather than the full OF)
22:38:35amiconnWell, we're using the bcm in whatever mode it has been set up by the loader
22:38:57amiconnWe cannot write our own code because it's a custom, undocumented architecture
22:39:51amiconnThe lcd update used to be slower than it is now - since I've introduced asynchronous updates using a fixed update window, it's pretty okay
22:41:19amiconnIn fact ~15MPixels/s (that's ~30MByte/s) is one of the fastest update speeds among the PP targets
22:41:26amiconn100fps at 80MHz
22:41:59 Quit Barahir (Ping timeout: 260 seconds)
22:42:01funmani remember seeing much higher numbers for e200v1
22:42:32amiconnHence, "one of"
22:42:44amiconne200v1 is approx. twice as fast, but then it uses dma
22:42:52 Quit esperegu (Remote host closed the connection)
22:42:57stripwaxis BCM documentation $$ or NDA?
22:43:00amiconn(which we don't set up ourselves because we don't know how - we just use it)
22:43:13stripwaxoh - that's interesting
22:43:43stripwaxso we've probably got ipod lcd DMA available, in the OF, but not spent the time on that yet since the loader gives us something workable
22:43:54stripwax^ an unsubstaniated guess, of course
22:44:12amiconnActually my estimation for the Elio was a bit pessimistic, because I remembered c200v1 speed wrong
22:44:37amiconnA better estimation says ~37fps at 80MHz
22:45:36wodzamiconn: Do You know what are interrupt vectors beyound 0x7c in coldfire 5249/50? I readed manual dozen of times and can't figure that out
22:45:39amiconnYes - if we knew how to set up PP dma, iPod Video could probably be sped up further, but only for "ordinary" updates
22:45:52stripwax'ordinary' = ?
22:46:10amiconnordinary == framebuffer -> lcd
22:46:19amiconnI.e. YUV blitting cannot be sped up that way
22:46:40stripwaxcould do YUV->framebuffer -> lcd ?
22:46:42 Quit efyx (Read error: Connection reset by peer)
22:46:52amiconnwodz: Many modules can set their interrupt vector. See the 5249/5250 manuals
22:47:08amiconnThat would be slower than writing the yuv directly to the lcd
22:47:13 Join efyx [0] (
22:47:36stripwaxreally? i thought framebuffer was in iram though
22:48:07amiconn(1) Framebuffer of the Video can't be in iram. Ever calculated the necessary size?
22:48:11*stripwax see YUV blit framerate on ipod 5.5g −− wow, why is that no better than c250?
22:48:31stripwaxamiconn - hah. good point.
22:48:38amiconn(2) Even if it were - writing to IRAM and writing to the bcm are essentially the same speed. See?
22:49:09amiconnstripwax: Calculate c200 framerate * (132*80) / (320*240)
22:49:13amiconnWhat do you get?
22:49:36stripwaxfair enough. i assumed they weren't, because of the lcd update code (that I'll be the first to admit I don't know nearly enough about) waiting on BCM flags
22:49:57amiconnDone? Right, 13.7 fps. You see the Video *is* faster - per pixel
22:50:04wodzamiconn: Thats true but generaly You have primary interrupt controller (compatibile with 5206) and secondary interrupt controller with aditional 64 interrupts.
22:50:39stripwaxamiconn - no, I mean: c250 main lcd is 138.5fps, YUV is 100fps. 5.5g is 101fps, YUV is 30fps.
22:50:56amiconnYes, because on the c200 the controller is the limiting factor
22:51:13stripwaxoh. yeah, that makes sense now.
22:51:18 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
22:51:21amiconnThis means YUV conversion is almost for free, since it happens mostly during the time we have to wait for the controller anyway
22:51:43stripwaxyep. I just realised. and BCM is fast enough that we don't really have any waiting time.
22:51:48amiconnOn the Video it's the opposite - apart from setup and completion we don't have to wait at all
22:52:13amiconn(and completion is done asynchronously, monitored via an isr)
22:52:17 Quit tchan (Client Quit)
22:56:47 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
22:57:36amiconnwodz: Yes, and? The secondary interrupt controller's interrupt are all autovectored - you can just set the base. But the primary interrupt controller's interrupts are *not* all autovectored - the respective module defines the vector
22:59:05amiconnThe vectors can be set to an arbitrary 8 bit value
22:59:55wodzok comming back to manual
23:01:13 Quit kugel (Remote host closed the connection)
23:03:43 Quit S_a_i_n_t (Ping timeout: 268 seconds)
23:05:18 Join S_a_i_n_t [0] (S_a_i_n_t@
23:07:40wodzaimconn: Thanks now I understand.
23:07:51 Quit tchan (Quit: WeeChat
23:09:40 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
23:09:41***Saving seen data "./dancer.seen"
23:12:06 Quit tchan (Client Quit)
23:12:33wodzamiconn: Regarding bad MPIO performance I tried to find out what's going on. The strange thing I found is that ui is pretty slow (I mean selecting items in main menu produce a few seconds delay). If I force breake in debugger during that delay I always end up in stop #0x2000 instruction comming from sleep_core()
23:14:16 Join tchan [0] (~tchan@lunar-linux/developer/tchan)
23:14:30 Quit DerPapst (Quit: Leaving.)
23:14:54wodzI don't understand this behaviour since manual breake is random so I should hit some function from time to time and not always stop.
23:18:05amiconnThe probability depends on how frequently a certain code snippet is executed
23:18:47amiconnAnd 'stop' will put the cpu core to sleep until the next interrupt, so if there isn't much to do, it will sleep a lot
23:19:12amiconnIt *should* not sleep often, if at all, when boosted though
23:19:20 Nick Ypsy is now known as YPSY (
23:19:28wodzthats the point. Why there is not 'a lot to do
23:19:37amiconnIf it does, there is a problem somewhere in the code
23:19:39wodzif there is ui activity
23:20:53 Quit plauclair (Ping timeout: 264 seconds)
23:21:21 Join esperegu [0] (~quassel@
23:27:40 Join chamunks [0] (
23:28:31chamunksWould anyone be able to link me a more detailed explination as to why rockbox wont run on my 3rd gen classic 160gb slim?
23:31:33linuxstbchamunks: Simply because it's new hardware (compared to the ipods Rockbox does run on), and no-one has done the work to port Rockbox to that hardware. will give you an idea of the work involved.
23:32:09 Quit merbanan (Ping timeout: 258 seconds)
23:32:13chamunkslinuxstb, thank you very much I appreciate it. :)
23:32:56 Quit robin0800 (Remote host closed the connection)
23:34:00chamunksI wonder if there are any scans done of this hardware as I could supply some. I'm not a coder just would love to help any way that I can.
23:35:17linuxstbCoding is what's needed at this stage.
23:36:50chamunksGuess I'll have to wait on switching to flac till another day. I heard that the amp on the ipod's arent really good enough for flac anyways.
23:41:44 Quit M3DLG (Ping timeout: 258 seconds)
23:44:20sinthetekwhat kind of coding?
23:44:35sinthetekassign me something
23:49:00 Quit Luca_S (Quit: CGI:IRC (EOF))
23:49:07chamunksShhh dont get my hopes up :P I'd love nothing more than to be able to switch away from mp3v0's
23:53:32bertrikoh, I thought it was hard encryption that blocked a rockbox port
23:54:30linuxstbbertrik: The DFU mode has been exploited, so we can execute code on it. Also, TheSeven has ported his "iBugger" debugger, so code can then be uploaded (and RAM contents downloaded) via USB.
23:55:16gevaertsNo known way to have rockbox boot without DFU I think though
23:55:52linuxstbIIUC, the boot ROM can be dumped via iBugger (it's memory-mapped), but the NOR is accessed over SPI, so a driver for that needs to be written, and the NOR dumped (and probably decrypted). So it's still at the very early reverse-engineering stages.
23:56:52 Join merbanan [0] (
23:58:43 Quit anewuser (Ping timeout: 245 seconds)

Previous day | Next day