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 2009-12-12

00:01:48kkurbjunso my colors menu was reset to rockbox blue for the backdrop, I changed the color to black and then went back to the plugins and it still showed the blue. I exited the sim and then restarted it and the plugins started showing the backdrop color that is set
00:02:08kkurbjunbut if I change it again it doesn't take effect in the plugin menus till a restart
00:02:56 Join grndslm [0] (
00:02:56 Join jon-kha [0] (
00:02:56 Join ThomasAH [0] (
00:02:56 Join Slasheri [0] (i=miipekk@rockbox/developer/Slasheri)
00:03:23 Nick Guest43936 is now known as gevaerts (n=fg@rockbox/developer/gevaerts)
00:03:31 Part froggyman
00:03:31 Quit fml ("CGI:IRC")
00:03:39 Quit domonoky (Read error: 104 (Connection reset by peer))
00:03:43***Saving seen data "./dancer.seen"
00:04:35 Nick ThomasAH is now known as Guest14302 (
00:12:55 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
00:22:07 Quit tomers (Read error: 113 (No route to host))
00:23:15 Join kugel [0] (n=kugel@rockbox/developer/kugel)
00:28:18 Quit Strife89 ("Leaving.")
00:33:39 Quit Tomis (Success)
00:38:35JdGordonkkurbjun: ok, can you flyspray it? I wont have a chance to even look at it till tomorow night
00:53:43 Quit flydutch ("/* empty */")
00:57:15 Quit Omlet ("( :: NoNameScript 4.22 :: )")
01:15:44 Quit kugel (Read error: 113 (No route to host))
01:16:11Unhelpfuln1s: i'll take a look on e200, but really it seems a no-brainer that reducing 3 lookups to 2 is a win. shame we've no arm targets with rbit.
01:16:51 Join lontra [0] (n=lontra@
01:17:36lontrahi i have a 1st generation ipod and i can't install rockbox on it ... it is telling me 'can't load rockbox.ipod: bad checksum hold MENU+SELECT to reboot then SELECT+PLAY'
01:19:02 Quit BlakeJohnson86 ("Leaving.")
01:20:12lontraalso it's telling me my system is readonly
01:21:23lontraalso it's telling me my mount point is not writable
01:22:11archivatorlontra: when you unzipped, did it give any errors?
01:22:30archivatorIn fact, how did you install rockbox?
01:22:39lontraarchivator: no it didn't
01:22:46lontrai had to do it the manual way
01:22:54mc2739lontra: That usually means your filesystem has errors.
01:23:34lontramc2739: how can i fix that?
01:23:43archivatorEnter disk mode, do a chkdsk run, extract again, reboot the player (don't forget to safely remove/unmount first!)
01:24:15archivatorAlso, don't forget the /f parameter to chkdsk (the one that actually fixes things :) )
01:24:24lontrai don't have chkdsk
01:24:38lontrai am using linux
01:25:09mc2739lontra: use dosfsck or fsck.vfat
01:25:58archivatorThough, bear in mind that there are some errors that fsck.vfat can't fix..
01:26:13 Quit LambdaCalculus37 ("Leaving")
01:26:15 Quit petur ("Zzzzz")
01:26:21lontrashould i just revert it to factory defaults w/ itunes first?
01:26:26lontraand then start again?
01:26:54archivatornah, just enter disk mode. Menu+select, select+play
01:29:32lontraand unzip as a regular user right?
01:31:23archivatorOn a side note, could someone enlighten me how to remove a thread I started (RockboxKernel says that a thread should not return)?
01:33:22 Join FOAD_ [0] (
01:33:40 Join Stephen_ [0] (
01:35:43archivatorit worked?
01:35:45lontrait's been a while since i've used rockbox
01:35:49Stephen_can themes released into the public domain be updated and added tot the theme site ?
01:35:55lontraabout a year so it looks pretty nicde
01:36:27archivatorlontra: so you know about the manual, right? Just a friendly reminder :)
01:37:17lontrayup ;)
01:37:56archivatorgood :) Btw, for future reference, 1st generation ipod != 1st generation nano :)
01:38:18lontrai know i was being sloppy
01:42:36lontraarchivator: thanks for the help
01:46:37 Join palaoa [0] (
01:46:50 Quit lontra ("Lost terminal")
01:46:58 Quit FOAD (Read error: 110 (Connection timed out))
01:46:59 Nick FOAD_ is now known as FOAD (
01:48:03 Join BlakeJohnson86 [0] (
01:48:48 Quit palaoa (Client Quit)
01:53:33 Quit PaulJam_ (Read error: 60 (Operation timed out))
01:56:18 Quit phanboy4 (Read error: 104 (Connection reset by peer))
01:57:03 Join MethoS- [0] (n=clemens@
02:03:46***Saving seen data "./dancer.seen"
02:18:08 Join fdinel [0] (
02:20:42 Quit JdGordon (Read error: 110 (Connection timed out))
02:27:14 Quit togetic ("WeeChat 0.3.0")
02:28:28 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
02:29:18 Quit w1ll14m (" HydraIRC -> <- The alternative IRC client")
02:29:47 Join w1ll14m [0] (
02:34:41 Quit Rondom (Nick collision from services.)
02:34:44 Join Rondom [0] (
02:43:25 Join saratoga [0] (i=9803c6dd@gateway/web/freenode/x-ymutajvkimthgwlx)
02:43:38saratogadid we get rid of the source tarball or just move it? I want to update the wiki
02:45:51 Quit archivator ("Leaving")
02:53:48 Join Darkknight512 [0] (
02:55:28Darkknight512I'm thinking of running a battery benchmark on my Nano 2g because I noticed the battery guage is not calibrated. I looked at the results page and notice it only says Nano, do I just submit it under Nano or make one called Nano 2g
02:56:36mc2739Darkknight512: You should make a new entry for nana 2g
02:58:00Darkknight512so I just make a new heading?
03:00:12saratogathat seems sensible
03:00:31Darkknight512thanks, ill post it up when its done, just installed rockbox 2 days ago must say I am impressed
03:14:45 Join CaptainKewl [0] (
03:31:28 Quit jesusaurus ("Reconnecting")
03:31:30 Join jesusaurus [0] (
03:33:34 Join hiptobecubic [0] (n=john@unaffiliated/hiptobecubic)
03:34:08hiptobecubicHowdy all, I'm wondering if there is a reasonable way to access the rockbox database or is it some proprietary format?
03:35:48hiptobecubicrsync takes me too long to update so i've written a crummy standalone database manager to keep track of what needs to be updated, but if I could use rockbox's database features it would be perfect
03:38:09 Quit Thundercloud (Remote closed the connection)
03:40:37Stephen_this might help ?
03:41:05Stephen_i think its old tho
03:41:31hiptobecubichmm interesting. thanks
03:42:29 Quit jesusaurus ("leaving")
03:43:18hiptobecubicOh snap there is already a python script for this. What a waste of time
03:43:19hiptobecubicStephen_, thanks.
03:43:34Stephen_no problem.
03:44:36hiptobecubicWell, i guess i didn't waste THAT much time. And I learned some sqlite, so all is not lost
03:47:05 Join n17ikh [0] (
03:54:04 Join kugel [0] (n=kugel@rockbox/developer/kugel)
04:00:08 Join loyx [0] (
04:03:49***Saving seen data "./dancer.seen"
04:08:22 Quit Stephen_ ("Leaving")
04:12:15 Quit n17ikh (Read error: 104 (Connection reset by peer))
04:15:43 Quit GeekShadow ("The cake is a lie !")
04:19:40 Quit kugel (Read error: 60 (Operation timed out))
04:24:42 Join n17ikh [0] (
04:40:32 Quit BlakeJohnson86 ("Leaving.")
04:40:55 Join BlakeJohnson86 [0] (
05:00:45 Quit Sajber^ (Read error: 104 (Connection reset by peer))
05:09:16Darkknight512Batterybenchmark has been running for 2 hours, lalala
05:19:09 Quit BlakeJohnson86 ("Leaving.")
05:19:11 Quit hd (Read error: 104 (Connection reset by peer))
05:19:25 Join HellDragon [0] (
05:47:12 Quit Darkknight512 ("CGI:IRC (EOF)")
05:50:04 Join csh [0] (n=kvirc@
05:51:15 Quit csh (Client Quit)
05:54:40 Quit Guest44305 (Read error: 104 (Connection reset by peer))
05:57:35 Join BlakeJohnson86 [0] (
06:03:51***Saving seen data "./dancer.seen"
06:06:54 Quit BlakeJohnson86 ("Leaving.")
06:10:36 Quit MethoS- (Read error: 104 (Connection reset by peer))
06:23:01 Quit fyre^OS ("Ur skills' fireproof like a wooden panel -- U got feds talking leet on your IRC channel!")
06:37:57CIA-6New commit by tomers (r23936): Cowon D2: Revert some of r23927 (keymap changes), and restore Radio keymappings ...
06:40:45 Join BlakeJohnson86 [0] (
06:46:27 Join tomers [0] (
06:46:38DogBoywhat happens when you have files with bit depth of 24
06:48:05LloreanDogBoy: They should be sampled down to 16 for playback.
06:49:07DogBoywould it happen crudely in rockbox or not at all
06:49:27LloreanIt will happen in Rockbox, but it's not the highest quality.
06:49:36DogBoynot sure how to do that with high quality
06:49:48DogBoyonly thing that comes to mind is audacity
06:50:31 Part hiptobecubic ("For a holy stint, a moth of the cloth gave up his woolens for lint.")
06:53:27UnhelpfulDogBoy: or just convert to lossy for rockbox, as the encoders tend to work with floats, anyway
06:54:04DogBoyif I convert it to ogg it's still gonna be 24 bit ogg right
06:54:18UnhelpfulDogBoy: there is no such thing as 24-bit ogg. or 16-bit ogg.
06:54:30DogBoywell that clears that up
06:54:46DogBoybut is the encoder going to do the best job
06:56:00Unhelpfullibvorbis uses floats internally. there should not be a problem.
06:57:37 Quit panni_ ("( :: NoNameScript 3.81 :: )")
06:59:57 Join Tristan [0] (
06:59:58 Quit HBK (Read error: 104 (Connection reset by peer))
07:00:22 Nick Tristan is now known as Guest51958 (
07:01:07 Join HBK [0] (
07:05:05DogBoythanks Llorean, Unhelpful
07:07:12 Quit bluebrother (Nick collision from services.)
07:07:15 Join bluebroth3r [0] (n=dom@rockbox/developer/bluebrother)
07:16:57 Quit CaptainKewl (Read error: 60 (Operation timed out))
07:18:03 Join PaulJam_ [0] (
07:25:00 Join Horschti [0] (n=Horscht2@xbmc/user/horscht)
07:43:32 Quit Horscht (Read error: 110 (Connection timed out))
07:51:14 Join fyrestorm [0] (
08:00:31 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
08:02:22 Quit HBK (Read error: 104 (Connection reset by peer))
08:03:53***Saving seen data "./dancer.seen"
08:11:54 Join HBK [0] (
08:21:43CIA-6New commit by tomers (r23937): Cowon: Fix wrong pluginlib action
08:24:36CIA-6New commit by tomers (r23938): CodeBuster: Replace mapping defines with direct use of PLA_* macros. ...
08:26:15Unhelpfulhrm, so, i now have a working arm-optimized pow2_int, and it's faster than the old one that was using a long mutiply to multiply by a single shifted bit. however... it does not produce the same result as the existing version.
08:27:38CIA-6New commit by tomers (r23939): Use positive logic in #ifdef hell
08:28:17Unhelpfulthe existing pow2_int returns 1 if the fixed-point input is precisely equal to 0, 0 if the input is any other value <1.0, and otherwise returns 2^x rounded to the nearest integer value. looking at the callsites in faad2's repo, the float version of the codec just uses pow(2.0, x), and the code is otherwise similar.
08:28:41CIA-6New commit by tomers (r23940): Fix Hebrew translation
08:29:05 Join n1s [0] (
08:29:17Unhelpfulthe existing behavior of pow2_int, in other words, appears to be a bug rather than something that the decoder expects instead of what the name suggests. thoughts on replacing it?
08:41:37n1sif it's a bug, fixing it would be good, no? :)
08:43:40 Quit PaulJam_ (Read error: 110 (Connection timed out))
08:44:40Unhelpfuln1s: *if* it's a bug... perhaps it's intended to be pow2_int_unless_x_is_less_than_one_but_not_exactly_equal_to_0, and they just shortened the name for convenience? ;)
08:45:23n1sUnhelpful: does it give any difference i the output?
08:48:10 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
08:51:10Unhelpfuln1s: definitely.
08:52:41n1sand here is where my knowledge of lossy codecs end but shouldn't there be a way to determine if the changes lead to better or worse accuracy?
08:53:36Unhelpfulpow2_fix has the same behavior, too... again, this seems like a bug or bad naming, pow2(-1) when the output format supports values to 2^-14 seems like it should return .5 rather than 0
08:55:23Unhelpfuln1s: perceived quality is not readily measured. if the algorithm is designed around these weird pow2 semantics, then changing them could be breaking it. but, as i said, it sure looks like the float version of libfaad just uses pow(2.0,x) without testing to see if 1>x>0 or x<0 and forcing 0 utput in those cases
08:55:29amiconnThat looks like bugs to me
08:56:12Unhelpfulamiconn: and yet it's unchanged in modern libfaad :/
08:57:13*amiconn wonders how an implementation similar to log_s16p16() and exp_s16p16() in grey_core.c would perform in comparision to that
08:58:16 Join angelwolf71885 [0] (
08:58:23amiconnNo table lookup and no multiplications (the exp() implementation uses one to improve precision, but it's not strictly needed)
08:58:30 Quit angelwolf71885 (Client Quit)
09:00:14Unhelpfulamiconn: i have pow2 using only one short mul now, in its linear interpolation stage. also that's doing a *lot* more manipulation.
09:06:38amiconnFor pow2(), handling of the integer part could be simplified vs. those, of course
09:12:21Unhelpfullibfaad's handles the fractional part directly via an LUT + linear interpolation, and then does a long multiply to shift by the whole part. i've removed the long multiply, and i can also improve the interpolation accuracy by left-shifting the base value rather than right-shifting the offset... it'll just take a larger shift to descale the final value.
09:13:40*pixelma wonders a bit about tomers' PLA change in codebuster but then finds all PLA things suspicious :\\
09:15:30 Join flydutch [0] (
09:15:59Unhelpfulalso, i don't really understand the algorithm used here too much... although it looks like a variation of one that i saw, with the loops unrolled and the LUT values replaced by literals. it looks like if (x > 3/4) { x = 4*x/3; r += log2(3/4)}, and then for 7/8, 15/16, etc?
09:16:13 Quit bmbl ("Bye!")
09:16:29amiconnLinear interpolation is less precise than that though. Wonder whether it matters...
09:17:01amiconnHmm, not if it uses a LUT.
09:17:10*amiconn should look closer first
09:20:50 Join angelwolf71885 [0] (
09:21:07Unhelpfulalso i wonder if it's wise to special-case val==0 in these... surely it's uncommon enough for them to be called with *precisely* 0 that this is a losing proposition?
09:22:14 Quit angelwolf71885 (Client Quit)
09:24:33pixelmatomers: I think it makes the button/action assignments even harder to read which are already inconveniently spread across two files when pluginlib actions are used... action assignment in the switch statement but dependent on the used action contexts at the beginning of the file, button definitions in the pluginlib file.... much fun to find out when buttons/actions don't work
09:25:43n1spixelma: was a desicion reached on what to do with the whole PLA thing?
09:29:30 Join kcynice [0] (n=kcynice@
09:29:39pixelmaI think we could agree on reducing PLA to only one context which could be used for some plugins and if you need more actions then you should have actions/button definitions in the plugin itself. There is a patch in the tracker which implements that and changes all plugins that currently use PLA to conform with that. It works quite well
09:30:58pixelmadidn't try intensively yet but e.g. metronome which had problems due to PLA on two of my targets is controllable with this
09:31:21pixelma(on c200 and Ondio)
09:31:52n1sis that get-in-before release stuff or should it wait?
09:31:55Unhelpfuli felt that i had excellent success converting pictureflow from PLA to direct use of get_custom_action
09:34:44pixelman1s: I think it would be nice if it could get in before release because it actually fixes things (e.g. as mentioned metronome), it probably needs a bit more testing on different targets
09:35:30pixelmathe version of the patch I had was broken by LambdaCalculus' keymap additions though
09:35:42 Quit antil33t ()
09:36:14pixelmathat codebuster change breaks it too probably
09:40:56n1spixelma: i have a beast and h300 available if you need testing on any of those
09:42:40pixelmaFS #10387
09:43:16pixelmaI guess it needs n update
09:48:01n1si'll wait for one then :)
09:50:07 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
09:50:11 Join PaulJam_ [0] (
09:59:36 Quit BHSPitMonkey (Read error: 145 (Connection timed out))
10:00:29 Quit kcynice (Read error: 54 (Connection reset by peer))
10:03:56***Saving seen data "./dancer.seen"
10:07:59 Join Utchybann_ [0] (
10:15:33 Join TheSeven [0] (n=theseven@rockbox/developer/TheSeven)
10:23:00 Join Zaba_ [0] (
10:26:06tomerspixelma: I tried to make codebuster usable in Cowon D2. The 'Fire' action was occupied with 'Down' (see Touchscreen actions), which prevented me from pression 'validate' in codebuster.
10:26:51tomerspixelma: Do you have any better idea? I thought that adding some more PLA_* to the switch case wouldn't hurt.
10:28:03pixelmathat demonstrates the problems with PLA quite well :\
10:29:36tomerspixelma: I can't find anything wrong in my commit. Enlighten me please... I do agree that the PLA stuff is a bit complicated to understand and manage, though :-)
10:30:17pixelmaI'm not sure if adding more actions could break other targets (because the contexts were there before), but I don't like that you put the PLA inside the loop makes things even harder to read (one more place to look up)
10:30:41 Join DerPapst [0] (
10:31:10pixelmathe problem is that if you change things to improve the situation on one target, it's not that hard to break another
10:31:15 Join Kitr88 [0] (
10:31:22 Quit HBK (Read error: 54 (Connection reset by peer))
10:31:54pixelmaand especially if a combination of contexts is used
10:32:27 Join stoffel [0] (
10:32:34pixelmawhich shouldn't be done in first place though
10:33:00tomerspixelma: mmm... maybe indeed adding new PLA_'s to the switch case might break things. Do you have any suggestion on how to resolve for D2?
10:34:23 Join HBK [0] (
10:38:36 Quit Zaba (Read error: 110 (Connection timed out))
10:39:07pixelmaif it was me, I wouldn't use pluginlib actions at all (I could live with the much simplified one context version) - It obfuscates things and adds "invisible" connections. I'm sure that the original author does not know on which targets the controls work as intended and on which it doesn't, with the definitions inside the plugin (and only there) that wouldn't happen, even though it's a bit more work in the beginning
10:40:39tomerspixelma: What do you suggest to do? 1) Totally trash the pluginlib_action mechanism. 2) Consolidate context into one. 3) ?
10:41:47 Nick Guest14302 is now known as ThomasAH (
10:42:17 Nick ThomasAH is now known as Guest85111 (
10:43:13 Nick Guest85111 is now known as ThomasAH (
10:43:32pixelma2 would already be a relieve and there is a patch for it in the tracker - see FS #10387. It even includes changes to codebuster but the last version probably doesn't apply currently
10:45:45tomerspixelma: Maybe we should keep things how it is, waiting for FS #10387 to be committed? Maybe asking kugel where it stands?
10:47:26 Nick Zaba_ is now known as Zaba (n=zaba@about/goats/billygoat/zaba)
10:49:12 Quit Kitar|st (Read error: 110 (Connection timed out))
11:07:40CIA-6New commit by tomers (r23941): Invadrox: Add support for both touchscreen and buttons control of the game. ...
11:14:34 Quit Utchybann_ ("I like core dumps")
11:23:10CIA-6New commit by tomers (r23942): Fix wrong button used in touchscreen keymap
11:26:31 Join Jaykay [0] (
11:27:35 Quit alexbobp (Read error: 104 (Connection reset by peer))
11:28:48*tomers Hates robotfindkitten
11:35:39 Quit stoffel (Remote closed the connection)
11:40:29Jaykaytomas: imo rockblox 1d is worse :)
11:40:33Jaykay*tomers :D
11:44:50tomersJaykay: I definitely agree! I think we should start a discussion in the dev-list to discard of some of the obviously redundant, unusable, un-enjoyful and hateful 'games'
11:45:32Jaykayi never understood why rockblox1d is included in rockbox... :)
11:45:33pixelmaI wonder how you can "hate" a game und what it hurts having that
11:45:57tomersok, not hate, but definitely not fond of :-)
11:46:09Jaykayit increases binsize :D
11:46:33tomersJaykay: If we like to attract users, we should discard of some legacy. I guess Ubuntu had some discussion in what games to include in their distribution. I believe some old X games were not part of it either
11:46:43pixelmaby the way, tomers - your invadrox change caused red (and I don't understand it)
11:46:55tomerspixelma: Thanks. Checking
11:46:57pixelmaJaykay: no, they don't, they are plugins
11:47:49Jaykaypixelma: i mean download size :)
11:47:59pixelmanot by much
11:48:16Jaykaybut they do increase it :)
11:48:38tomerspixelma: The reds are because of touchscreen targets that haven't got some macros defind (LEFT, RIGHT, etc)
11:49:02 Join alexbobp [0] (n=alex@
11:49:04pixelmaI meant, that I don't understand your changes
11:49:37tomersI want to control the game either with the buttons *or* with the touchscreen
11:49:59tomersI find it hard to find an elegant solution for this requirement there...
11:52:13tomersI'll replace the if with switch... It won't hurt and make it more elegant
11:53:58pixelmathe ACTION_* in the beginning look very weird to me but I'm no coder (although I did some button or action commits). I think it doesn't do what you think it does, not sure
11:54:41tomerspixelma: It's ugly (but works). I'm working now on removing it.
11:55:05pixelmawhy the red then?
11:55:52tomerssome ifdef used (XXXX | LEFT) where LEFT was not defined (touchscreen target with no buttons assigned for left)
11:57:59pixelmabut | is no "or" there, is it? At least my impression is that you want to achieve that with this, I don't *believe* that it does (but call me confused about that)
11:58:24 Join petur [0] (
11:58:30 Join archivator [0] (n=archivat@
12:01:39 Join Sajber^ [0] (
12:03:57***Saving seen data "./dancer.seen"
12:04:40 Join GeekShadow [0] (n=Antoine@reactos/tester/GeekShadow)
12:04:42archivatorWhich target has a 160x128 grayscale screen?
12:05:23pixelmaH100, M5, greyscale Ipods (not Mini)
12:05:37archivatorpixelma: thanks.
12:06:14pixelmaI believe one of the Samsungs too but not sure which one
12:06:25archivatorI only needed one :)
12:09:12 Join Horscht [0] (n=Horscht2@xbmc/user/horscht)
12:16:16CIA-6New commit by tomers (r23943): Invadrox: Fix red caused by touchscreen targets with no buttons defined for movements
12:18:20tomerspixelma: I decided on some other solution from what I previously said. Not the prettiest, but works.
12:22:50pixelmaI'd still like to understand :\
12:23:41pixelmae.g. what does "#define TOUCHSCREEN_LEFT (BUTTON_MIDLEFT | BUTTON_BOTTOMLEFT)" mean?
12:25:20pixelmathe latter looks like a button combo to me but that doesn't make much sense for the game, or am I missing something?
12:25:26tomerspixelma: That's bitwise ORing of two mutually exclusive (bit-wise) values. The value of what you gave as an example translates to (Cowon D2): (0x80 | 0x400) −−> 0x480.
12:26:06pixelmadoesn't help me one bit
12:26:26pixelmano pun intended
12:26:35tomerstomers: Now, is you press BUTTON_BOTTOMLEFT, the value is matched against 0x480 using bitwise-AND: if (button | 0x480) then −−> because(0x400 | 0x480) == 0x400 != 0
12:26:35 Quit Horschti (Read error: 110 (Connection timed out))
12:27:10tomersRead the following:
12:27:48tomerspixelma: This methods is used extensively all over our code. You will benefit from understanding this
12:28:32pixelmawell, I want to understand from a users' point of view - doesn't it mean you need to press both parts?
12:29:45tomerspixelma: Good question. The answer it negative. You can use *either* of the keys :-)
12:33:35pixelmaok, I'm still not sure I understand correctly because that's not how I used button or action defines before and I only saw different ways to add alternative buttons for only some targets in other plugins (examples off the top of my head are brickmania and sudoku). I hoped someone else could chime in :(
12:37:28tomerspixelma: I'm currently working on the manual for Cowon D2. Do you happen to know who can help me with drawing an svg file of the player?
12:40:01 Quit yosafbridge ("Coyote finally caught me")
12:40:11 Join yosafbridge [0] (
12:40:49 Join pamaury [0] (
12:41:44 Join bertrik [0] (
12:41:50 Join MethoS- [0] (n=clemens@
12:46:05Unhelpfulhere's what i have so far on improving pow2_* in libfaad:
12:46:49Unhelpfulthe tables for pow2 and log2 are moved into iram, a multiply is removed from pow2, and the pow2 table precision is extended
12:52:03 Quit bmbl (Connection timed out)
12:57:00 Join kugel [0] (n=kugel@rockbox/developer/kugel)
13:00:03n1sUnhelpful: *looks* nice :)
13:00:40n1show does it affext speed?
13:03:58 Quit HBK (Read error: 104 (Connection reset by peer))
13:07:44 Join HBK [0] (
13:08:45 Quit HBK (Read error: 104 (Connection reset by peer))
13:15:00 Quit tomers (Read error: 113 (No route to host))
13:15:24 Join Tomis [0] (n=Tomis@
13:22:30 Quit bertrik (Remote closed the connection)
13:23:12Unhelpfuln1s: shaves off about .6MHz from svn on e200. pow2 is still not called that heavily.
13:27:42n1snice, what do you think about log2 in speex? it's basically a straight binsearch, no LUT, problem is that it's included in the core for voice so we can't use iram for the table and also it will likely be a bit of a binsize hit
13:28:25n1sthe iram thing can of course be solved with some #ifdef hackery
13:34:25 Join HBK [0] (
13:34:50 Quit HBK (Connection reset by peer)
13:38:13n1si guess the real question is: do we want to use the LUT in core? and i'd say that depends on the eventual speedup...
13:38:42n1sso i think i answered my own question and needs to run a benchmark :)
13:41:42 Quit kugel (Read error: 60 (Operation timed out))
13:45:33 Join HBK [0] (
13:48:55Unhelpfuln1s: you could use the debruijn-ish multiplier and the same table that core ffs already uses
13:52:13 Quit flydutch ("/* empty */")
13:53:11 Join KBH [0] (
13:53:50 Quit KBH (Read error: 104 (Connection reset by peer))
13:58:50 Join KBH [0] (
13:59:34 Quit KBH (Read error: 104 (Connection reset by peer))
14:02:04 Join kugel [0] (n=kugel@rockbox/developer/kugel)
14:03:02 Join teru [0] (
14:03:54 Quit HBK (Read error: 110 (Connection timed out))
14:04:00***Saving seen data "./dancer.seen"
14:04:14 Join HBK [0] (
14:04:35 Quit HBK (Read error: 54 (Connection reset by peer))
14:05:46 Join Thundercloud [0] (
14:06:50 Join HBK [0] (
14:08:26 Join KBH [0] (
14:08:44 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
14:08:59 Join KCYNICE [0] (n=KCYNICE@
14:09:03 Quit KBH (Read error: 104 (Connection reset by peer))
14:10:02n1sUnhelpful: i wonder if it's worth it, the whole idea being code unification :)
14:11:11 Join KBH [0] (
14:11:40 Quit KBH (Read error: 104 (Connection reset by peer))
14:19:55 Join KBH [0] (
14:20:37kugeltomers: (for the logs) that | for button pressing in invadrox only works because it's using &, all other plugins use a switch construct, which means both buttons need to be pressed
14:20:41 Quit KBH (Read error: 131 (Connection reset by peer))
14:21:00kugelso I can understand the commit is confusing at the first look
14:24:08 Quit pamaury ("exit(*(int *)0 / 0);")
14:25:13 Quit HBK (Connection timed out)
14:25:45pixelmathanks! Couldn't it be simplified a bit? It looks to me like it could
14:26:07 Join HBK [0] (
14:26:43 Quit HBK (Read error: 104 (Connection reset by peer))
14:28:21 Join domonoky1 [0] (
14:30:09 Join HBK [0] (
14:30:56 Quit HBK (Read error: 104 (Connection reset by peer))
14:32:28 Join stoffel [0] (
14:36:44 Join HBK [0] (
14:37:17 Quit BlakeJohnson86 (Remote closed the connection)
14:37:22 Quit KCYNICE (Read error: 104 (Connection reset by peer))
14:37:28 Quit HBK (Read error: 54 (Connection reset by peer))
14:37:33 Join KCYNICE [0] (n=KCYNICE@
14:39:19 Join HBK [0] (
14:40:05 Quit kugel (Read error: 145 (Connection timed out))
14:40:57 Join BlakeJohnson86 [0] (
14:43:45 Join Grahack [0] (
14:44:15 Quit domonoky (Read error: 110 (Connection timed out))
14:46:00KCYNICEhi, all.hi. how should i get/put ipod's synchronization mode? I want to prevent iTunes erasing all my changes of the track itmes db. because my ipod is not supported by the current version.
14:47:31KCYNICEi want to change the ipod's synchronization mode to "Manually manage musics and videos".
14:48:36Tomisitunes preferences>devices>prevent ipods and iphones from syncing automatically
14:48:42 Quit DerPapst ("Leaving.")
14:49:30 Quit BlakeJohnson86 (Read error: 104 (Connection reset by peer))
14:50:55KCYNICEyes, it is. but i want to do it by myslef, without iTunes. like realplayer does.
14:51:10Tomisdo what by yourself
14:51:35Tomisoh, you want to use the ipod like a mass storage device, not use itunes at all?
14:51:44 Join KBH [0] (
14:52:42Tomisthen set it to not "open itunes when attaching this ipod" and make sure it's not set to auto-sync, check "manually manage this ipod" as well
14:52:44Tomisand you're done
14:52:52 Quit KBH (Read error: 104 (Connection reset by peer))
14:53:17n1s...or uninstall itunes...
14:53:28Tomisyea but that's too obvious
14:53:43n1slol, indeed
14:54:26 Join KBH [0] (
14:54:26KCYNICEi think, this option do store somewhere in ipod. because realplayer can do it
14:55:21 Quit KBH (Read error: 104 (Connection reset by peer))
14:55:51n1sKCYNICE: that isn't really rockbox related and most rockboxers usually don't give the workings of Original Firmwares and synctols and whatnot much thought or use
14:56:49KCYNICEoh, i see. and thank you all the same.
14:58:06 Join BlakeJohnson86 [0] (
14:58:10 Quit HBK (Read error: 110 (Connection timed out))
14:59:08 Join webguest96 [0] (
14:59:38 Part KCYNICE ("out for shopping......")
15:00:41 Join HBK [0] (
15:00:56 Quit webguest96 (Client Quit)
15:01:38 Quit HBK (Connection reset by peer)
15:02:12 Join HBK [0] (
15:02:31 Quit HBK (Read error: 104 (Connection reset by peer))
15:05:05CIA-6New commit by teru (r23944): spacerocks: Fix FS #9380: spacerocks shows funny rocks. ...
15:06:38 Join kood777 [0] (
15:10:48 Join HBK [0] (
15:11:34 Quit HBK (Read error: 104 (Connection reset by peer))
15:13:31 Join HBK [0] (
15:14:35 Quit HBK (Success)
15:14:49Mode"#rockbox +o GodEater " by ChanServ (ChanServ@services.)
15:14:52Mode"#rockbox +b *!* " by GodEater (n=bibble@rockbox/staff/GodEater)
15:14:52WarnBanlist contains 18 bans!
15:14:54Mode"#rockbox -o GodEater " by ChanServ (ChanServ@services.)
15:21:40 Quit stoffel (Remote closed the connection)
15:23:57 Join TopyMobile [0] (
15:25:46 Quit MethoS- (Remote closed the connection)
15:29:48 Quit kood777 ("CGI:IRC (EOF)")
15:30:20n1si wonder if FS #10845 is a bug, IIUC he's saying, "glyphcache works as intended, which is suboptimal in his usecase"
15:30:38n1sand the "fix" is suboptimal ans quite a waste for everyone else
15:31:09 Quit Jaykay (Read error: 110 (Connection timed out))
15:31:43 Join kood777 [0] (
15:42:42 Join FOAD_ [0] (
15:45:57domonoky1rasher: do we have any download statistics for the theme page ?
15:46:47 Join Highlander [0] (
15:49:34 Join kugel [0] (n=kugel@rockbox/developer/kugel)
15:50:02 Quit FOAD (Read error: 145 (Connection timed out))
15:50:02 Nick FOAD_ is now known as FOAD (
15:54:01 Join efyx_ [0] (
15:55:40 Join fdinel [0] (
15:56:16 Quit teru ("Quit")
15:57:42 Quit Grahack (Read error: 110 (Connection timed out))
15:57:49 Quit kood777 ("CGI:IRC (EOF)")
15:59:35 Join Grahack [0] (
16:00:03 Join kood777 [0] (
16:04:02***Saving seen data "./dancer.seen"
16:09:07 Join guest001 [0] (
16:09:14 Join paulk_ [0] (
16:11:13paulk_Hi! I'm using rockbox api to do a simple game, and I need to do a random. My problem is that rb->current_tick is ALWAYS = 26 !
16:11:36 Quit kood777 ("CGI:IRC (Ping timeout)")
16:11:50kugelrb->current_tick is a pointer, you want *rb->current_tick instead
16:12:13guest001nls, I have a sansa e200 v1 and a modified font (~1 MB). With the fix on FS #7062, it saves a lot of boot time (from ~13s to 6s)
16:14:52paulk_kugel, It's the same with *rb->current_tic
16:15:16kugelon sim or target?
16:15:23kugelit seems wrong to me
16:15:53paulk_what is sim ?
16:15:53kugeli.e. I think you're doing it wrong, sine if current_tick was wrong rockbox wouldn't function properly
16:15:59kugel(or at all)
16:16:21paulk_i'm using a target (e250)
16:16:27kugelguest001: the only change besides the buffer is casting to long instead of short?
16:17:28 Join kood777 [0] (
16:17:51 Nick YPSY is now known as Ypsy (
16:17:58 Quit Grahack (Read error: 110 (Connection timed out))
16:18:05guest001kugle: no, it reserves more memory for the font and the cast is for addressing a number bigger than 65xxx
16:19:21paulk_kugel, sorry, it works with *rb->current_tic . I didn't modified the good string.
16:20:11kugelcurrent_tick isn't random though, it's incrementing
16:20:25paulk_kugel, I know, i juste need it for srand
16:20:25 Join Grahack [0] (
16:20:40kugelalright, I believe other plugins do it the same way
16:21:29kugelare there fonts with more than 65k glyphs?
16:21:43kugelI don't think rockbox could handle those
16:23:52guest001kugel: I don't know how many glyphs my font has, but you can try it with the official unifont. I guess you will get about the same result.
16:25:05 Join tomers [0] (
16:25:42n1skugel: the limit seems to be the size of the buffer in bytes not glyphs, and a large font can easily get to 60k i guess
16:27:00n1si think that cast-short*-to-long* sems like a hack though but didn't look closesly and i think in principle we should fix it if people want to use these lare font buffers but i don't see how the fragmented loading could take such a long time...
16:30:02guest001The patch is maybe not the right solution, but it's a big improvement for me and I think to many who uses non-latin fonts
16:34:04kugeln1s: I can imagine that this speeds up things for the existing font buffer as well
16:34:33n1sdo these non-latin languages you guys are using use a lot more glyphs than latin languages?
16:35:09n1skugel: well, i can too, in case the buffer is too small and we need to hit the disk to load a glyph but why is that done u
16:35:20n1ss/u/such a lot during boot?/
16:35:39 Quit fyrestorm ("Ur skills' fireproof like a wooden panel -- U got feds talking leet on your IRC channel!")
16:36:01kugelno idea, it should only load the glyphcache files, right?
16:36:21 Quit Highlander ("Quitte")
16:37:18kugelwrong window, oops
16:37:41n1skugel: yeah, something must be going wrong...
16:38:22n1smaybe this buffer should be settable like the max files in dir?
16:38:36n1sin addition to the bugs fixed, of course
16:39:44kugelmaybe a glyph was only partially in the font buffer; on boot it tries to load it, then it sees it's incomplete and so loads that glyph from disk, and since the font buffer was full loading that glyph from disk means the other glyphs are now incomplete and then it loads all from disk; just a theory
16:42:40 Quit PaulJam_ (".")
16:44:16Unhelpfuln1s: it might be worth making an exception for speex, though, as it's used in core. the debruijn-ish version is a good speed/size balance, and since the LUT is sparse it can probably overlap a fair bit with the existing one for ffs
16:45:58n1sUnhelpful: yeah, i'll check if the log2 is at all important for speed first i guess
16:46:31kugelis there any protection against incomplete glyphs in the font buffer?
16:46:32 Quit guest001 (Read error: 104 (Connection reset by peer))
16:46:41n1skugel: no idea
16:47:23kugelI can imagine that it doesn't memset() old entries to 0 so that garbage could be identified as a valid entry
16:47:42kugelthis would only be a problem on boot, since at runtime it's managed in a linked list AFAIK
16:48:55kugelbut I haven't looked at the related code at all, so it's all theory
16:48:58Unhelpfulif it's special-just-for-speex then it only has to support one mapping for 0 input :)
16:50:46 Join tg1 [0] (n=thomas@
16:51:53n1sUnhelpful: yeah
16:52:04n1si guess i should dig up a test file
16:52:25n1sdid you test my improved bitrev12 patch?
16:52:45Unhelpfulno, i spent a very long time on a very stupid bug in DESCALE
16:54:16Unhelpfulyou can also save an instruction on coldfire for doing log2 if you use a slightly different multiplier. i have a very long list. ;)
16:55:41n1shaha, i also thought about doing bitrev12 as a simple LUT, should only need 8kB ram so would actually be doable on some targets that arent using iram or have plenty left
16:55:49 Join CaptainKewl [0] (
16:56:18 Join CaptainKwel [0] (
16:56:43n1sit's also ironic that the beast seems to respond best to several optimizations, since it's already faster thatn everything else :/
16:59:02 Join Zaba_ [0] (n=zaba@about/goats/billygoat/zaba)
17:00:34 Quit mt_ (Read error: 104 (Connection reset by peer))
17:03:03Unhelpfuldefinitely time for bed here... i'll hack up a script later to see what the smallest total table size can be for overlapped log2/ffs table
17:03:27 Quit CaptainKewl (Read error: 60 (Operation timed out))
17:03:51n1sok, goodnight :)
17:06:09 Quit Zaba (Read error: 110 (Connection timed out))
17:06:25n1sthe tremor lib is using a structure that seems to effectively prevent inlining in a lot of places, i wonder if changing it help... will be intrusive as hell though
17:06:43n1ss/help/will help/
17:07:40 Join Highlander [0] (
17:12:29 Quit Sajber^ (
17:12:29 Quit Guest51958 (
17:12:29 Quit xavieran (
17:12:42NJoinSajber^ [0] (
17:16:11NJoinGuest51958 [0] (
17:16:11NJoinxavieran [0] (
17:16:39 Quit Grahack (Read error: 60 (Operation timed out))
17:18:40 Join Grahack [0] (
17:23:31 Quit Rondom (Nick collision from services.)
17:23:34 Join Rondom [0] (
17:24:45 Quit Grahack (
17:24:45 Quit yosafbridge (
17:24:45 Quit alexbobp (
17:24:45 Quit Slasheri (
17:24:45 Quit jon-kha (
17:24:45 Quit ThomasAH (
17:24:45 Quit grndslm (
17:24:45n1sshould commented debug code in imported code be kept around even if it's unlikely to ever be used again?
17:24:56NJoinalexbobp [0] (n=alex@
17:27:44NJoinyosafbridge [0] (
17:29:35pixelmakood777: could you stop cluttering up the logs with useless punctuation, please?
17:29:46kood777Ok sorry
17:30:56 Quit tg1 ("Leaving.")
17:31:18 Join tg1 [0] (n=thomas@
17:33:40NJoinThomasAH [0] (
17:35:27 Quit tg1 (Client Quit)
17:35:27 Quit kood777 ("CGI:IRC (EOF)")
17:35:43 Join tg1 [0] (n=thomas@
17:37:47CIA-6New commit by nls (r23945): Comment var only used by already commented out debug code, delete #if 0'ed block move a comment and fix some broken indentation
17:39:48 Quit Guest51958 (Connection timed out)
17:39:54 Join saratoga_ [0] (i=463f90ed@gateway/web/freenode/x-hurtpjczhnjsbrkz)
17:39:56NJoinjon-kha [0] (
17:39:56NJoinGrahack [0] (
17:39:56NJoingrndslm [0] (
17:39:56NJoinSlasheri [0] (i=miipekk@rockbox/developer/Slasheri)
17:41:07saratoga_i got a ClipV2 for $15 from woot a while ago if anyone wants it
17:46:29 Nick fxb__ is now known as fxb (
17:46:43 Quit archivator (Remote closed the connection)
17:47:33 Join archivator [0] (n=archivat@
17:51:02Lloreansaratoga_: It does appear the cygwin page doesn't mention a MIPS cross compiler yet.
17:51:26LloreanSo I guess the packages are probably out of date at least.
17:51:26saratoga_i noticed that, but i'm not even sure we have one compiled
17:51:51LloreanThen again, the proper solution to that is "well, we need updated packages"
17:55:01saratoga_compiling it for cygwin is easy enough but i don't know how to package it
17:55:29Lloreannor I
17:57:38 Quit Llorean ("Leaving.")
17:59:18 Quit kugel (Read error: 113 (No route to host))
17:59:39tmztpackage it why?
17:59:50saratoga_to update the cygwin packages
18:00:03tmztwith setup.exe?
18:00:29tmztcygwin setup.exe
18:00:42tmztotherwise just rar it and extract under /usr/local
18:00:48tmzttar not rar
18:01:46saratoga_we're talking about updating the cygwin packages on for mips, i'm not sure if thats what you are saying
18:04:05***Saving seen data "./dancer.seen"
18:04:12tmztah ok
18:04:48tmztare crosscompilers generally packeged to install with cygwin's setup.exe downloader?
18:07:03*domonoky1 plans to add a download counter to the themes site. i would just count accesses to the themes zip (redirected via a download.php), is that ok, or are there better ways ?
18:15:52archivatordomonoky1: if you do it that way (download.php), please, please, don't forget to set an appropriate content-type and content-disposition. application/x-octet-stream is not an appropriate type.
18:16:34*domonoky1 goes to look up how todo that :-)
18:20:11 Join toffe82 [0] (
18:29:21 Join kugel [0] (n=kugel@rockbox/developer/kugel)
18:33:18 Quit Grahack (Read error: 145 (Connection timed out))
18:36:28 Join Grahack [0] (
18:36:41CIA-6New commit by Domonoky (r23946): Add download counter to themepage. ...
18:41:22 Quit archivator (Remote closed the connection)
18:51:38 Quit saratoga_ ("Page closed")
18:54:41 Join Zaba [0] (
18:56:01CIA-6New commit by Domonoky (r23947): revert previous commit for the moment.
18:57:22 Quit J-23 (Broken pipe)
19:06:39 Quit Zaba_ (Connection timed out)
19:08:44 Join flydutch [0] (
19:23:07 Quit togetic (Read error: 104 (Connection reset by peer))
19:23:38 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
19:28:31 Nick Ypsy is now known as YPSY (
19:28:44 Join pixelma_ [0] (i=quassel@rockbox/staff/pixelma)
19:28:44 Quit pixelma (Nick collision from services.)
19:28:59 Quit amiconn (Nick collision from services.)
19:29:00 Join amiconn_ [0] (
19:29:04 Nick pixelma_ is now known as pixelma (i=quassel@rockbox/staff/pixelma)
19:29:06 Nick amiconn_ is now known as amiconn (i=quassel@rockbox/developer/amiconn)
19:29:23 Join panni_ [0] (
19:31:26 Quit petur ("burp")
19:36:54CIA-6New commit by kkurbjun (r23948): M:Robe 500: Fix FIQ's and make the audio DMA a FIQ, simplify the ADC code and make it more reliable. Fix ADC problems on initial boot.
19:49:35 Join archivator [0] (n=archivat@
19:54:48 Join einhirn [0] (
19:56:46 Quit togetic (Read error: 110 (Connection timed out))
19:57:37 Join togetic [0] (n=togetic@unaffiliated/ibuffy)
19:59:41 Quit paulk_ ("Ex-Chat")
20:04:08***Saving seen data "./dancer.seen"
20:14:31 Join kugel_ [0] (
20:17:05 Join pamaury [0] (
20:22:41 Quit Grahack (Read error: 110 (Connection timed out))
20:22:51 Quit togetic (Read error: 110 (Connection timed out))
20:23:17 Join Grahack [0] (
20:24:57 Quit saratoga (Ping timeout: 180 seconds)
20:27:11 Join archivator_ [0] (n=archivat@
20:30:50 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
20:33:34 Quit kugel (Read error: 110 (Connection timed out))
20:35:33 Quit Thundercloud (Remote closed the connection)
20:37:34 Quit einhirn (Read error: 104 (Connection reset by peer))
20:43:55 Join saratoga [0] (i=9803c6dd@gateway/web/freenode/x-ukpmoknkvblkatxa)
20:44:45 Join mt [0] (n=mtee@rockbox/developer/mt)
20:46:25 Quit martian67 ()
20:47:30 Quit Grahack (Read error: 110 (Connection timed out))
20:47:36 Join martian67 [0] (n=arkfar@about/linux/regular/martian67)
20:49:42 Join Grahack [0] (
20:49:53 Join solexx_ [0] (
20:53:59 Quit martian67 ()
20:54:39 Join martian67 [0] (n=arkfar@about/linux/regular/martian67)
20:54:44saratogamt and I are thinking about creating a branch in the rockbox svn for our mdct library while we slowly get it into shape
20:55:44saratogabut i don't think i've seen anyone do that before
20:57:12 Quit kadoban (Remote closed the connection)
20:59:53kugel_saratoga: sounds like a good idea
21:01:57kugel_we don't really utilize branches (but rather force developement with patches on flyspray) which is bad imo
21:02:05mtwe wanted to know the general opinion on this before branching, because this way it would be much easier for people to follow each others' changes, without actually affecting the trunk.
21:02:14 Quit solexx (Read error: 110 (Connection timed out))
21:03:31 Quit martian67 ()
21:04:14 Quit gevaerts (Nick collision from services.)
21:04:21 Join gevaerts_ [0] (n=fg@rockbox/developer/gevaerts)
21:04:59mtkugel_: We'll get to FS, just not now, the code is still in an early stage - and a bit too ugly :) - to have a task opened for it.
21:06:01kugel_developement via fs is rather painful
21:06:32kugel_go for the branch, I don't see what should speak against it
21:09:05 Quit TopyMobile (Read error: 110 (Connection timed out))
21:09:42pixelmawhy is not utilising branches "bad" though? I'm just curious and would like to hear the reasoning... e.g. I imagine branches can get confusing too if there are too many
21:09:48saratogamt: almost have a patch against current svn ready
21:10:17mtnice !
21:14:27kugel_pixelma: because branches are much more convinient than patches on fs
21:14:53kugel_and they have literally ~0 cost
21:15:07pixelmasounds like personal preference to me
21:16:00kugel_well, try it
21:16:05saratogaif a lot of people are working on something its much easier then patches
21:16:09saratogasince you don't have to sync them
21:16:23 Join martian67 [0] (n=arkfar@about/linux/regular/martian67)
21:16:58kugel_I work with git locally because working on branches (with commit history) is superb compared to plain source editing and handing out patches
21:17:14pixelmaI think the Gigabeat port started in a branch (maybe some other) but then it was hard to move their changes into the main tree
21:17:29saratogacompiles your code in the codec lib, but does not use it for any codecs (old mdct is compiled and linked as well)
21:19:44mtsaratoga: Great ! Could you add that to the wiki too ?
21:19:51kugel_pixelma: if a branch is not sync'd with trunk regularly merging can be difficult yes. but if you do it, it's a no brainer
21:19:55saratogamt: sure
21:20:49pixelmafor my work I do in Rockbox, I really don't see a need for a personal branch, really. And setting that up (maybe even a different environment) just means additional work hence won't be more convenient myself. I am not saying it doesn't make sense for some cases but I'll definitely will not "just try"
21:20:53mtkugel_: How does it have 0 cost ? (Never done a branch before :) )
21:21:48archivator_Doesn't branching do an actual copy of the data in svn? I think kugel's thinking about git.
21:22:03 Quit Grahack (Read error: 60 (Operation timed out))
21:22:20kugel_yes you copy, but it doesn't take the space of a full checkout
21:22:32 Quit flydutch ("/* empty */")
21:23:24 Quit toffe82 (Read error: 110 (Connection timed out))
21:23:48 Join Grahack [0] (
21:24:13mtpixelma: The reason for a branch here is that, currently, there are 3 developers working on the same code set, (and maybe later some more devs would like to join), so it's kind of annoying to keep moving files around through e-mails, or even FS.
21:24:55 Nick fxb is now known as fxb__ (
21:25:30 Join petur [50] (n=petur@rockbox/developer/petur)
21:25:38mtCurrently, everyone has to make sure the others didn't make any new changes before working on something new. So, having a branch would greatly simplify this whole process.
21:29:08saratogawell if we're agreed would someone mind branching (i'm not good at svn)?
21:29:21saratogai'd like to commit the next set of changes to the patch I just posted
21:29:35 Join fml [0] (
21:30:37kugel_why should someone complain? there's no real reason to complain
21:33:49kugel_internally a branch takes ~0 space
21:34:41mtkugel_: *if* there were problems, it's easily reversible, right ?
21:35:57 Quit Tomis (Read error: 104 (Connection reset by peer))
21:36:18bluebroth3ra copy in svn is a shallow copy. To be exact, a rename is a copy too.
21:37:15kugel_mt: that's the point of version control systems :)
21:37:17bluebroth3rbut the only version control systems that are actually *good* at merging are the distributed ones
21:37:39bluebroth3rsvn wasn't even able to track merges prior to 1.5
21:38:08fmlpixelma: hello. Have you had a chance to look at FS #10843 ? You're a master of arts here so I'm interested in your opinion.
21:39:09kugel_bluebroth3r: it was able, but it was rather stupid
21:39:30bluebroth3rcalling a merge a non-brainer is kinda optimistic for svn IMO. At least if you have concurrent changes
21:42:01 Join Tomis [0] (n=Tomis@
21:42:39pixelmafml: sorry, not yet and I'm off for a while again
21:43:27fmlpixelma: ok
21:44:48 Join Thundercloud [0] (
21:45:46bluebroth3rfml: wouldn't it be better to use \sfdefault insteadl of explicitly selecting cmbr?
21:46:08fmlbluebroth3r: I think it's also a problem with git. The complexity stems from the problem itself, not fron the "stupidity" of the tool.
21:46:38bluebroth3ror \textsf to be precise
21:47:06fmlbluebroth3r: I don't know. \textsf uses another font IIRC, not the cmbright.
21:47:24bluebroth3rfml: well, dvcs can generally cope much better with the task. At least from my experience.
21:47:39bluebroth3rhmm, \textsf should be the default serif font
21:48:11CIA-6New commit by mt (r23949): Created a branch for the faster MDCT work, to make it easier for developers to follow each others' changes, and to safely modify the codec library ...
21:48:14bluebroth3rat least from my understanding of the font system ... I've rarely changed fonts
21:50:14fmlbluebroth3r: I think the default sans serif font is not cmbright, hence the explicit selection.
21:51:17fmlI tried \textsf but it didn't give the desired result
21:52:24 Join toffe82 [0] (
21:52:42bluebroth3rhmm, but shouldn't the titles use the same sans serif font as the rest of the manual?
21:55:29fmlbluebroth3r: yes. Hence the patch. The title on the front page does not look like a "proper" title now.
21:57:42 Join toffe82_ [0] (
21:57:44 Quit Grahack (Read error: 110 (Connection timed out))
21:58:43CIA-6New commit by saratoga (r23950): Inital checkin of the MDCT library code. Compiles but is not actually used by any codecs.
21:58:55 Join Grahack [0] (
21:59:03CIA-6New commit by saratoga (r23951): Inital checkin of the MDCT library code. Compiles but is not actually used by any codecs.
21:59:53saratogaalright its in
22:00:27saratogamt: mind if we rename fft-fixed to fft-ffmpeg?
22:01:18mtsaratoga: Not at all.
22:04:09***Saving seen data "./dancer.seen"
22:04:14CIA-6New commit by saratoga (r23952): Add TRANSFORM_EQUAL optimization to fft. Reduces multiplications by about 10%.
22:05:59CIA-6New commit by saratoga (r23953): Rename fft-fixed to fft-ffmpeg to better reflect its history.
22:10:45CIA-6New commit by saratoga (r23954): Fix mistake.
22:10:52saratogaok i have to leave for a bit
22:12:52 Quit ps-auxw (Remote closed the connection)
22:13:20 Quit fml ("CGI:IRC (Ping timeout)")
22:14:44 Join ps-auxw [0] (
22:15:17 Quit toffe82 (Read error: 110 (Connection timed out))
22:19:25kugel_saratoga: 2x the same commit message?
22:19:45saratogakugel_: yeah the first commit really didn't do anything since I forgot to svn add
22:20:35 Join phanboy4 [0] (
22:21:20 Quit Grahack (Read error: 110 (Connection timed out))
22:23:44 Join Grahack [0] (
22:28:27 Join mister_pink [0] (
22:33:44 Join Omlet [0] (
22:34:23CIA-6New commit by tomers (r23955): Sokoban: Add keymaps for Cowon D2
22:35:13 Join fml [0] (
22:35:41fmlWhich packages do I need to install to build the manual under VMWare?
22:41:12 Join stoffel [0] (
22:47:33 Join TopyMobile [0] (
22:50:59 Join robin0800 [0] (
22:54:14 Join Tomis2 [0] (n=Tomis@
22:57:31 Quit tomers (Read error: 113 (No route to host))
22:59:48 Join robin0800_ [0] (
23:00:44 Quit stoffel (Remote closed the connection)
23:02:59CIA-6New commit by Domonoky (r23956): add a mechanism to add columns to tables into the adminview.
23:08:41 Quit Tomis (Read error: 110 (Connection timed out))
23:08:41 Nick Tomis2 is now known as Tomis (n=Tomis@
23:14:24 Nick kugel_ is now known as kugel (
23:14:55 Quit robin0800 (Read error: 110 (Connection timed out))
23:15:18CIA-6New commit by Domonoky (r23957): add a missing quote.
23:19:48 Quit Grahack (Remote closed the connection)
23:21:42 Nick gevaerts_ is now known as gevaerts (n=fg@rockbox/developer/gevaerts)
23:22:24 Quit fml ("CGI:IRC")
23:27:25 Quit petur ("burps")
23:38:00 Quit n1s (Read error: 110 (Connection timed out))
23:38:37 Quit martian67 (Nick collision from services.)
23:38:42 Join martian67 [0] (n=arkfar@about/linux/regular/martian67)
23:39:29 Quit martian67 (Nick collision from services.)
23:39:30 Join lol2 [0] (
23:39:42 Join martian67 [0] (
23:48:46 Quit lol2 (Read error: 54 (Connection reset by peer))
23:49:12 Quit martian67 (Read error: 104 (Connection reset by peer))
23:52:45 Join martian67 [0] (
23:52:48 Quit martian67 (Remote closed the connection)

Previous day | Next day