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 2007-08-26

00:02:33 Part oxygen77
00:02:42*bluebrother wonders when ohloh will spider svn again ...
00:03:31peturbluebrother: they are updating existing project again after somebody enterd three linux distros last week
00:03:48peturwhich took them all week to recover :)
00:04:03bluebrotherthree distros? Urgh, that's a lot of code ...
00:04:14rasherBagder: Ideally, you shouldn't have to change your scripts, but I guess you never know..
00:05:23Bagderoh, Henrik Backe strikes back
00:05:44Bagder(FS #7657)
00:05:46***Saving seen data "./dancer.seen"
00:06:27rasherSeems reasonable
00:06:37rasherAlthough it should of course be /usr/local/rockbox instead
00:06:40bluebrotherI dislike the folder name
00:06:42 Quit ToHellWithGA (Read error: 110 (Connection timed out))
00:07:08rasherYeah, not too fond of it either, but that's details, the general idea looks good
00:07:10bluebrotherdid he miss the patch?
00:07:14Bagderthe idea and approach seems fine though
00:07:41rasherbluebrother: looks like it
00:07:54Bagderwow, it was two years!?
00:07:59bluebrotherand the idea was already discussed, wasn't it?
00:08:03Bagder"I've not had the opportunity to do som rockboxing in almost two years."
00:08:05 Join saratoga [0] (i=98039a9c@gateway/web/cgi-irc/
00:08:26rasherI'd rather keep the directory hidden, perhaps .user-data?
00:08:28Bagderyeah he missed the patch...
00:08:31rasheror something like that
00:08:45*petur also likes splitting bin and data
00:09:04rasherAlthough I guess by using a non-hidden folder it makes it easier for users to do the right thing
00:09:27rasherparafin: putting "user" in it makes it clearer that this is something that you *should* modify, whereas .rockbox isn't
00:09:27bluebrotheraverage users should use rbutil anyway :)
00:09:37rasherbluebrother: Bias
00:10:12Bagderwell, average users should use rbutil but it won't prevent that they will do occasional manual poking too
00:10:38rasherI have all sorts of crap in my .rockbox for this very reason. I found myself wondering about polski-ascii the other day. Turns out that was removed half a year ago
00:11:20parafinthe bad thinks about .rockbox is plugin saves, which are found all over the place
00:11:23bluebrotherrasher: ;-)
00:11:55 Quit MySic (SendQ exceeded)
00:12:07Bagderthat's _one_ bad thing about it
00:12:37rasherHaving user-installed cruft along with the "official" themes/fonts etc is another
00:12:58rasherMakes it difficult to do an upgrade without the risk of orphan files
00:13:32Bagderyeah, it'll be good to more or less be able to make .rockbox read-only
00:13:41parafinthe point is that i want to be able delete .rockbox and install new version of rockbox without losing my preferences
00:14:00Bagderparafin: I think we all basically agree on this
00:14:32Bagdernow there's a patch available...
00:14:37 Quit Rondom ("Ex-Chat")
00:14:43bluebrotherwell, not yet ;-)
00:14:47 Quit petur ("beer + irish coffee = Zzzzzz")
00:14:50Bagdercheck again
00:15:01Bagderhe added it
00:15:16 Join henrikb [0] (i=53e3cf78@gateway/web/cgi-irc/
00:15:17bluebrotherah :)
00:15:48bluebrotherbut why look in /.rockbox anymore at all?
00:15:52henrikbHi, the patch is uploaded now (my first task since the move to flyspray)
00:16:06Bagderbluebrother: for things we ship by default I guess, like themes
00:16:18bluebrotherok, makes sense.
00:16:32Bagderhenrikb: yeah, we noticed. welcome back btw!
00:16:41bluebrotherbut it's not needed for recent bookmark file, plugins config etc.
00:18:04Bagderunless you want to allow a transition time
00:18:20rasherThat's what the plugin's for!
00:18:24rasherI assume
00:18:37bluebrotherdo we want such a thing? Moving the plugins to the categories system also hadn't one ...
00:18:45rasherI certainly think it should get removed again eventually.
00:18:57rasherIf not completely left out
00:19:24rasherI don't see the need to run it at *each* boot?
00:19:30bluebrotheras we don't have an official release I think we can safely leave out any transition period
00:20:18rasherMaybe just include the plugin and state somewhere (frontpage news? MajorChanges? Forum post? Mail to -users?) that you need to run this plugin once to migrate settings
00:20:39BagderI think I'll prefer that approach
00:21:26rasherAnd then stop building the plugin after a month or two
00:21:43henrikbWell, Its either on every boot, or it'll have to be run manually once, and not runnig it will cause probelems ehen config.cfg is saved.
00:22:11bluebrotherwhy will it cause problems? Because the folder structure is missing?
00:22:14rasherI think that's the user's problem, to be honest. Requiring a single manual run is fine as far as I'm concerned
00:22:17Bagderbut now I'll sleep...
00:22:29parafinwhy not place some flag if conversion has been already made?
00:22:42bluebrotherwell, you could fallback to the old location if the folder is missing ...
00:23:25 Join H10_007quick [0] (
00:23:32 Quit H10_007quick (Client Quit)
00:23:34rasherparafin: might be sensible, but then again, that means a disc access on every boot
00:23:38henrikbthe flag could be that /rockbox (or whatever is the proper name)n is created.
00:24:41bluebrotheryou would still either need to have additional code in the core that is superfluous after conversion or run a plugin upon boot, which would require a disc access.
00:25:21henrikbThats right but there is alredy plenty of disc activity on boot
00:25:51 Quit davina (Remote closed the connection)
00:25:59henrikbbut code bloat is always an issue
00:26:11bluebrotherstill it would increase boot time
00:26:34bluebrothersomething that is not wanted. I think it's absolutely safe to require the user to do a step manually
00:26:50bluebrotherand btw, we could include the folder structure in the distributed zip file
00:27:18rasherbluebrother: then you run the risk of retarded unzippers potentially messing it up :|
00:27:34rasherBut I guess we'll have to. What else would new users do
00:27:42bluebrotherand tell the user to manually move the config files or create new ones
00:27:48rasherIgnore me
00:28:11 Nick _pill is now known as pill (
00:28:41rasherRegarding the recent post to -dev, perhaps a new plugin category is needed? System or something
00:29:42 Join EnterUserName [0] (n=dave@pdpc/supporter/student/GeekZoid)
00:29:42 Join linuxstb_ [0] (i=5343d4aa@rockbox/developer/linuxstb)
00:29:43EnterUserNamehi all
00:29:50EnterUserNamehow long should the battery last on a Sansa?
00:29:58EnterUserNameif watching videos..
00:31:37henrikbI put the plugin in apps, i think at least one more system plugin is needed before creating a new category.
00:31:50 Quit HellDragon (Read error: 104 (Connection reset by peer))
00:32:21bluebrotherI believe we could safely go without the plugin.
00:33:11rasherbluebrother: I wouldn't say entirely without it, just tell users to run it manually
00:33:28rasherand remove the plugin again in a while
00:33:49rasherhenrikb: I meant for plugins that are used by the core (such as properties and credits)
00:34:07bluebrotherthere will be enough users that will not have updated in that time frame and complain ...
00:34:35 Quit Arathis ("Verlassend")
00:34:47henrikbrasher: properties is a viewer bu credits would qualify
00:35:02linuxstb_bluebrother: Yes, as demonstrated by the ipod users who still haven't updated to the latest bootloader....
00:35:07bluebrotheras we'll get support questions out of such a move anyway why bother about the plugin at all?
00:35:32rasherbluebrother: hit them with a cluebat.. at that point it really isn't our problem I say. HAving the plugin in a short transition period seems the nicer thing to do
00:36:23bluebrotherhow much configuration files is it really? config.cfg
00:36:36bluebrotherand for the plugins we had the same issue short ago ;-)
00:36:39rasherbluebrother: user themes, plugin savefiles, user fonts...
00:36:53bluebrotherwell, they can still live in the old location
00:37:07rasherBut the point is to help people do the right thing
00:37:14rasherUnless they're too damn slow, of course
00:37:27bluebrotherbtw, all theme zips would need to be adjusted too.
00:37:50henrikbI think that the part that creates the directory structure could stay and only the file copying part be removed after the transition period.
00:38:02henrikbThemes will work in both places
00:38:14rasherBut they shouldn't be put in the old location anymore
00:38:23bluebrotherhenrikb: why use a plugin for creating the folder structure? Just add the structure to the zip file
00:38:42bluebrotherusers will stop using the old structure once theme files are updated.
00:39:09bluebrotherand if a user grabs a theme zip with old structure he will extract it to the old structure ...
00:39:31henrikbI'ts easier to to figure out which directories ar needed for each platform in .c than in perl
00:40:01rasherbluebrother: and then lose it in the next update and bitch at the author (in a perfect world, in reality we'll bitch at us, but we'll tell him it's the author's fault)
00:40:45henrikbrasher: why would he loose it at the next update
00:40:45 Join ender [0] (
00:40:50rasherhenrikb: the structure could be created by the build system prior to running
00:40:54bluebrotherhenrikb: not sure ... we already do that, so the buildzip script would just need to mirror the folders
00:41:28rasherhenrikb: well, he might. If he deletes it (because he's been told that you can safely delete .rockbox when upgrading now)
00:42:41henrikbrasher: don't tell the user that until the themes sites have fixed all there themes.
00:43:06rasherThat'll be approximately never
00:43:11 Join HellDragon [0] (n=Nocebo@unaffiliated/helldragon)
00:44:35henrikbTime to go to bed, bye.
00:45:39saratogalinuxstb: around?
00:45:50saratogalibuxstb_: you too
00:46:15linuxstb_I am
00:46:33saratogasomeone sent me a really intersting file
00:46:52saratogait decodes correctly in that old test program you sent me a month or so back
00:46:59saratogabut not in the current version of rockbox
00:47:29saratogahow different are those two parsers?
00:47:29 Quit henrikb ("CGI:IRC (EOF)")
00:47:38saratogai want to dig into this issue
00:47:50linuxstb_Which old test program are you talking about? Which parser was it using?
00:48:00saratogathe one you sent me on July 1 or so
00:48:11saratogathe original version you wrote
00:48:24saratogai tried it on this file and amazingly it works, while the current svn does not
00:48:29linuxstb_I can't remember - was that using libasf?
00:48:38saratogai think so
00:48:51saratogaits got a libasf folder, but i thought it was the parser you wrote?
00:49:19 Join bush_pma [0] (i=PolarisO@
00:49:49linuxstb_If it's got a libasf folder, then it's using libasf
00:50:00 Quit bush_pma (Client Quit)
00:50:06 Join bush_pma [0] (i=PolarisO@
00:50:45linuxstb_You just need to add (or enable) some debugging DEBUGFs in wma.c and see what's going wrong.
00:50:45saratogaok, nerochiaro said it wasn't quite libasf in there, so i assumed it was your parser
00:50:49saratogaperhaps he was mistaken
00:51:21saratogaok i'll do that
00:53:03linuxstb_saratoga: Did you test if the asf parser is passing the final packet to the decoder? The other day you said you didn't think it did.
00:55:06saratogalinuxstb: I'm pretty sure it doesn't
00:55:15saratogasince the last packet is often smaller then the others
00:55:31saratogait tends to just return -1 and end the song 200 ms early
00:55:49 Part bush_pma
00:56:11linuxstb_I don't think the file duration being reported by Rockbox is correct though.
00:56:11 Quit ender` (Read error: 110 (Connection timed out))
00:56:12saratogai didn't really look into it though
00:56:17saratogaits not
00:56:32saratogamy test file is always reported being 2 seconds longer
00:56:41saratogabut i'm not sure if thats related
00:56:57saratogaat least the difference in length and the difference in playback time doesn't seem to be the same, so i'm assuming its two different problems
00:58:20saratogai've really been unenthusiastic about working on the parser i guess
00:58:24saratogaits just less interesting to me
00:58:33saratogaand also much more annoying to try and debug from a windows machine
00:59:20linuxstb_Yes, but it's just as important as the decoder...
00:59:55rasherHrm, that's not too bad. Make voice took 3:40 on my 1.3ghz laptop in cygwin.
01:00:42linuxstb_rasher: How quick is a Rockbox build on your laptop in cygwin?
01:01:14rasherNo idea.. let me try
01:01:41 Quit Robin0800 (Read error: 104 (Connection reset by peer))
01:04:26rashermake[1]: *** No rule to make target `sysfont.h', needed by `/home/rasher/rockbox/build/apps/bookmark.o'. Stop
01:04:31rasherRemind me again what this means
01:04:45rasherIt's something entirely unrelated to sysfont, I remember that, but not much else
01:05:10linuxstb_Do you have the compiler installed?
01:05:22 Quit ender (" # Basic IBM dingbats, some of which will never have a purpose clear # to mankind -- /usr/src/linux/drivers/char/cp437.uni")
01:06:45rasherHm.. somehting might have gone wrong
01:06:55rasherI entirely blame
01:07:26 Quit bluebrother ("leaving")
01:07:56 Quit Bjerrk ("Dead Men Tell No Tales")
01:08:19rasherI do have m68k-elf-*.. strange
01:08:37rasherarm failed to install at all, it seems
01:09:20rasherI do *not* have the patience to debug on cygwin..
01:09:28rasherLet me install from the rockbox cygwin mirror
01:09:59 Join ddalton [0] (
01:10:19saratogalinuxstb: i'm not seeing anything obvious wrong with this file
01:10:33saratogaits a vbr file, all packets ar ethe same length, and i'm not getting any errors
01:10:40saratogabut about 9000 samples decode as nonsense
01:10:47saratogaany ideas where i should be looking?
01:12:21linuxstb_Which 9000 samples?
01:12:32saratogatowards the end of the first packet
01:12:48linuxstb_And then subsequent packets are OK?
01:13:15linuxstb_Does it have multiple payloads/packet?
01:13:15saratoga26592 samples in the first packet, samples 23000 to 28900 are off
01:13:21saratogano multiple payloads
01:13:36saratogaand subsequent packets are mostly ok, though a few others do the same thing (i think)
01:13:45linuxstb_Have you checked how much data it's passing to the decoder, and compared that to the working parser?
01:14:36saratoganot yet, still finding my way around the working one
01:14:39 Quit lee-qid ("aufwiederbyebientotsayonara")
01:15:01saratogahowever, since most subsequent packets decode normally, I assume the data passed is right
01:15:11saratogai think my decode would not resync if the wrong amount of data were passed
01:15:36linuxstb_It sounds to me as if the problem is more likely to be the decoder then. There's not much the parser can get wrong.
01:16:02saratogai loaded the svn decoder into the libasf code and it decodes normally
01:16:18saratogathe only thing i changed to use it with libasf is the code for decoding one frame at a time
01:16:25saratogasince libasf still returns 1 superframe
01:16:38saratogado you think thats likely to be the problem?
01:18:25saratogathe width of the problem areas made me think parser, since its much larger then 1 frame, but i suppose if an entire superframe weren't decoded right, that might be big enough
01:18:44saratogai will look into that and see if the errors line up with superframe boundaries
01:19:07 Join sarixe [0] (
01:20:42 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
01:21:19 Quit chrisjs169 (Read error: 110 (Connection timed out))
01:25:46 Quit alienbiker99 ("( :: NoNameScript 4.02 :: )")
01:28:36rasherlinuxstb_: I can't manage to build with cygwin.. getting all sorts of errors for the various archs
01:29:22saratogalinuxstb: are superframes always equal to one packet?
01:30:00 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
01:30:28 Join killerz [0] (
01:32:09 Quit Rincewind_ ("Cya")
01:38:50 Join JdGordon [0] (
01:39:26rasherkillerz: If you're waiting for a response before asking a question, you might end up waiting a while. Just ask your question if you have one.
01:40:29 Join homielowe [0] (
01:43:21homieloweWho was the person who bought a gigabeat second hand and had problems with audio?
01:44:45 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
01:45:31 Part homielowe
01:45:55 Join H10_007quick [0] (
01:46:16 Quit H10_007quick (Client Quit)
01:47:43saratogawhat does this code actually do: ((bit_offset + 7) >> 3)
01:47:51saratogasome sort of rounding operation?
01:54:58 Join BHSPitLappy [0] (
01:57:20saratogalinuxstb: i'm looking at libasf's asf_data_read_payloads and asf_data_get_packet functions
01:57:29saratogathey're very similar to what you put in wma.c
01:57:42saratogawhy did you need to change them for rockbox?
01:58:18 Join hannesd__ [0] (
02:03:05 Quit idnar (Nick collision from services.)
02:03:07 Join idnar_ [0] (n=mithrand@unaffiliated/idnar)
02:03:40 Nick idnar_ is now known as idnar (n=mithrand@unaffiliated/idnar)
02:05:48***Saving seen data "./dancer.seen"
02:13:55 Quit hannesd (Read error: 110 (Connection timed out))
02:13:55 Nick hannesd__ is now known as hannesd (
02:16:04 Quit bdgraue (Remote closed the connection)
02:22:50 Join virGr [0] (i=d4cd8bf8@gateway/web/cgi-irc/
02:23:17 Join perrikwp [0] (n=chatzill@
02:25:22 Quit virGr (Client Quit)
02:30:54 Quit Nico_P (Remote closed the connection)
02:32:54 Quit DerPapst (Read error: 104 (Connection reset by peer))
02:33:34 Quit BHSPitLappy ("Leaving")
02:43:12 Quit thegeek_ (Read error: 104 (Connection reset by peer))
02:43:49 Join thegeek [0] (
02:47:51 Quit sarixe ("Peace")
02:54:20 Quit ctaylorr (Read error: 110 (Connection timed out))
02:58:14 Join tedrock [0] (
03:06:01 Quit Thundercloud (Read error: 104 (Connection reset by peer))
03:15:22 Join ddalton [0] (
03:17:19 Join perrikwp_ [0] (n=chatzill@
03:26:03 Join chrisjs169 [0] (
03:28:30 Join Xavier_Airport [0] (i=d4cd8bf8@gateway/web/cgi-irc/
03:30:54 Quit Xavier_Airport (Client Quit)
03:36:14 Quit perrikwp (Read error: 110 (Connection timed out))
03:40:57 Quit spiorf ("Read error: 110 (Connection timed out)")
03:41:01 Nick perrikwp_ is now known as perrikwp (n=chatzill@
03:42:30 Join midkay [0] (n=midkay@rockbox/developer/midkay)
03:43:21 Join ddalton_ [0] (
03:45:07 Nick ddalton_ is now known as dalton (
03:51:36 Join stevenm [0] (n=stevenm@
03:51:48stevenmI must say, whoever optimized MIDI to realtime on Iriver.. I am very impressed
03:52:36 Join ctaylorr [0] (
03:56:35daltonI have the line + ++ptr; from a patch. how many + signs should it have in the actual code?
03:57:18scorcheat least from that paste
03:57:52scorchedoes your screen reader not see the spaces between teh first + and the other 2?
03:58:41daltonthere were three + signs. Maybe I left one out. so should one be taken out in the code?
03:58:58daltonok thanks
03:59:04saratogathough +++ptr and ++ptr do the same thing
04:02:27 Join spiorf [0] (
04:05:51***Saving seen data "./dancer.seen"
04:08:39 Nick dalton is now known as daniel_dalton (
04:10:18 Join Llorean [0] (
04:10:48 Quit Llorean (Remote closed the connection)
04:11:03 Join Llorean [0] (
04:11:07 Quit Llorean (Remote closed the connection)
04:11:09 Join H10_007quick [0] (
04:11:16 Quit H10_007quick (Client Quit)
04:11:22 Join Llorean [0] (
04:18:05 Quit spiorf (Remote closed the connection)
04:22:46 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
04:23:27 Nick daniel_dalton is now known as dalton (
04:23:33 Nick dalton is now known as ddalton (
04:24:29 Quit Galois (Remote closed the connection)
04:24:37 Join Galois [0] (
04:38:16 Join aliask [0] (
04:46:27 Join jhulst [0] (
04:53:21 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:53:32 Join miepchen^schlaf [0] (
04:54:51 Quit pabs ("Lost terminal")
04:55:42 Join pabs [0] (
04:56:26 Quit pabs (Client Quit)
04:57:40 Join pabs [0] (
05:07:22 Quit SirFunk (Remote closed the connection)
05:14:35 Quit stevenm ("Connection reset by beer")
05:14:35 Quit jhMikeS (Read error: 104 (Connection reset by peer))
05:16:22 Quit perrikwp ("ChatZilla [Firefox]")
05:16:52 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
05:52:31 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
05:55:20 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
05:57:00 Join daurnimator [0] (i=daurn@unaffiliated/daurnimator)
06:05:55***Saving seen data "./dancer.seen"
06:09:53 Quit linuxstb (Read error: 110 (Connection timed out))
06:26:45 Quit chrisjs169 (Read error: 110 (Connection timed out))
06:39:30 Quit qweru ("KVIrc 3.2.0 'Realia'")
06:40:25 Quit kubiixaka ("Miranda IM! Smaller, Faster, Easier.")
06:41:54 Join Genre9mp3 [0] (n=yngwiejo@rockbox/contributor/Genre9mp3)
06:45:49 Quit saratoga ("CGI:IRC (Ping timeout)")
07:01:24 Join RaRe` [0] (
07:11:32 Join sarixe [0] (
07:20:15 Quit RaRe (Read error: 110 (Connection timed out))
07:55:59 Part toffe82
08:06:00***Saving seen data "./dancer.seen"
08:16:12 Join hiroaki666 [0] (i=3b70345f@gateway/web/cgi-irc/
08:17:59 Quit hiroaki666 (Client Quit)
08:24:55 Quit jhulst (Remote closed the connection)
08:28:41 Join kubiixaka [0] (
08:35:03 Join onakiefmj [0] (i=0@
08:42:00 Join desowin [0] (n=desowin@
08:52:34 Quit RaRe` ("Quit msgs should be longer.")
08:58:17 Join n1s [0] (
09:02:20 Join TiMiD[FD1 [0] (n=TiMiD[
09:05:35 Join RaRe [0] (
09:08:01 Quit sarixe ("Peace")
09:15:21 Quit TiMiD[FD] (Read error: 110 (Connection timed out))
09:24:44 Join Rondom [0] (
09:28:15 Join davina [0] (
09:30:42 Quit Rick (Read error: 104 (Connection reset by peer))
09:31:38 Join Rick [0] (
09:32:28 Quit lazka (Remote closed the connection)
09:35:27 Quit tedrock (Client Quit)
09:38:08 Quit rvvs89 (Nick collision from services.)
09:38:29 Join rvvs89 [0] (
09:40:33 Join Rob222241 [0] (
09:43:55 Join bluebrother [0] (i=OruT3B7F@rockbox/staff/bluebrother)
09:44:14 Join freqmod [0] (
09:50:16 Join oxygen77 [0] (
09:55:12 Quit phalax (Read error: 104 (Connection reset by peer))
09:59:05 Quit Rob2222 (Read error: 110 (Connection timed out))
10:00:36 Join darkraven [0] (
10:06:02***Saving seen data "./dancer.seen"
10:06:05scorchebluebrother: oye...user-ML
10:06:18scorche"I can't be bothered with the 'tunnel vision' style of 'obey or leave'."
10:08:58 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
10:09:27 Join miepchen^schlaf [0] (
10:14:53n1shaha, so they want people who wants the list to follow guidelines to ignore their posts and also want help from the same people...
10:18:28bluebrothersure −− why not? ;-)
10:20:38LloreanMaybe we should arrange a scripted weekly or bi-weekly mailing of the guidelines?
10:20:43 Join stripwax [0] (
10:20:48LloreanIf it can be trimmed from the archives.
10:21:31bluebrotherwell, I don't think it hurts too much in the archives
10:21:46bluebrotherjust give the mail a good subject
10:22:08Llorean"Weekly guideline reminder"? Heh
10:22:17n1sLlorean: sadly it wont help, people don't read guidelines anyway
10:22:20bluebrother"Weekly ignorance reminder" :)
10:23:30scorcheit isnt about whether top or bottom posting is is about willful ignoring of guidelines...and no one has said anything remotely close to "obey or leave"
10:24:12LloreanMy problem is the assumption that any reminder of the rules is automatically hostile.
10:24:23LloreanI don't think any of us feel ill will the first time someone top posts. It's a case of "Oh, they didn't see the guidelines"
10:27:13bluebrotherwell, usually someone writes a reminder to the etiquette. But this also only happens sometimes
10:27:32bluebrothermaybe this should be done more often? Otoh that will clutter the archives even more :(
10:28:56 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
10:29:26LloreanI just find myself repeatedly caught off guard by how indignant they get when they find out they're actually expected to follow the rules.
10:29:30 Join kk [0] (n=kkit@
10:29:38LloreanI'm honestly curious why they think anyone even bothered to write them down if they're not supposed to matter.
10:30:04bluebrotherbecause the downloaded a "product"?
10:35:55 Join bdgraue [0] (
10:40:01 Quit krazykit (Read error: 110 (Connection timed out))
10:40:49n1sjhMikeS: ping
10:45:45n1sanyway, for the logs probably, could you have a quick look at the "play_flush-voice-fix.diff" patch in FS #6239 and comment on if you think it will break anything?
10:48:23n1sit touches scary playback.c land and queues none of which I'm comfortable with :-)
10:51:15hcsI'm glad to report that I haven't experienced a freeze with my ipod color in a long time; every now and again playback will freeze until I pause and unpause, but this is much less irritating than having to reboot every half hour.
10:51:22 Quit Rondom ("Ex-Chat")
10:54:57 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
10:55:09 Join miepchen^schlaf [0] (
10:59:57 Join ender` [0] (
11:02:09 Quit Galois (Read error: 110 (Connection timed out))
11:02:52 Join Buschel [0] (
11:04:14 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
11:04:16 Quit Buschel (Client Quit)
11:15:09 Join Buschel [0] (
11:22:51 Quit freqmod (Remote closed the connection)
11:23:39 Join freqmod [0] (
11:25:12 Join Buschel_ [0] (
11:28:19 Join seablue [0] (
11:29:05 Join spiorf [0] (
11:31:05 Quit spiorf (Client Quit)
11:31:17 Join spiorf [0] (
11:33:15 Quit qwm (Remote closed the connection)
11:33:19 Join qwm [0] (
11:35:44 Quit ompaul ("Leaving")
11:38:21 Quit Buschel (Read error: 110 (Connection timed out))
11:44:45 Join petur [0] (n=petur@rockbox/developer/petur)
11:49:13 Join datachild [0] (
11:54:08 Quit freqmod (Remote closed the connection)
12:04:13 Join falen [0] (i=51edfa38@gateway/web/cgi-irc/
12:04:26*rvvs89 activates conversation
12:05:13rvvs89hmm, I seem to have swapped my #rockbox window with my #uncyclopedia window...
12:05:16rvvs89sorry about that
12:06:05***Saving seen data "./dancer.seen"
12:06:37falenAnybody knows the name of that c++ channel on freenode?
12:07:48bdgrauefalen: /join ##c++
12:08:14falenThank you!
12:08:39 Quit falen (Client Quit)
12:08:44 Join falen [0] (i=51edfa38@gateway/web/cgi-irc/
12:08:49 Quit falen (Client Quit)
12:08:59 Join Dalton [0] (
12:09:37 Quit Dalton (Client Quit)
12:09:44 Join Falen [0] (
12:10:48 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
12:11:09 Join miepchen^schlaf [0] (
12:12:10 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
12:12:47 Quit petur ("switching")
12:12:56 Join petur [0] (n=petur@rockbox/developer/petur)
12:15:28 Join miepchen^schlaf [0] (
12:25:18 Join ddalton [0] (
12:33:02 Quit desowin (Read error: 104 (Connection reset by peer))
12:33:40 Join Arathis [0] (
12:37:11 Join webguest82 [0] (i=59661ad4@gateway/web/cgi-irc/
12:37:21 Quit webguest82 (Client Quit)
12:39:04 Join obo [0] (n=obo@rockbox/developer/obo)
12:41:36 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
12:44:11 Join Lear [0] (i=chatzill@rockbox/developer/lear)
12:51:55 Quit jhMikeS (Nick collision from services.)
12:52:01 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
12:52:52 Join Domonoky [0] (
12:55:49 Join webguest82 [0] (i=59661ad4@gateway/web/cgi-irc/
12:58:46webguest82hello can anyoen help me ? :p
12:59:03peturwe could if you would ask the question ;)
12:59:06webguest82i have problem with dell latitude d600 hdd
12:59:07ddaltonwhat with?
12:59:18webguest82hdd have password protect
12:59:20peturwebguest82: wrong channel?
12:59:36ddaltonyeah I think he is on the wrong one.
12:59:43webguest82really? :D
12:59:45webmindlooks like it yes
12:59:45peturI know why
12:59:59ddaltonthis is the rockbox channel.
13:00:11 Join Galois [0] (
13:00:11peturgoogle points to rockbox when searching for harddisk password
13:00:17scorchewebguest82: unless you think that has something to do with rockbox (you are even using our webclient), it is offtopic in this channel
13:00:53webmindwhy do people blindly follow google? without -reading- the website ?
13:01:02webguest82who cares :p
13:01:07scorchewe do
13:01:10webguest82so sry and gl
13:01:18 Part webguest82
13:01:24Domonokyin the very early times of rockbox, there were problems with harddisk password lock.. so google links to rockbox.. but we cant help.. :-)
13:01:38scorcheit happens...
13:01:40*petur wanted to point him to
13:02:36bluebrothertoo late
13:05:20 Join DerPapst [0] (
13:11:22 Join My_Sic [0] (
13:12:10 Join Thundercloud [0] (
13:13:50*petur wonders where the hell austriancoder went...
13:14:13n1spetur: how is the usb stuff coming along?
13:14:29peturzero progress
13:14:48bluebrotherdoesn't sound good :(
13:14:54scorchepetur: so did you end up failing or passing?...isnt tomorrow the deadline?
13:14:58peturaustriancoder hasn't been online in almost 6 days
13:15:14 Join tvelocity [0] (
13:15:53peturI have not done his evaluation, I have mailed him that I have set him a deadline on tonight. If I do not see code I will fail him
13:17:52scorchedid he give any warning, or real-life occurances that would explain these absences?
13:18:43peturthe last mail I got mentioned his mother having a car crash but she was ok only he had to do more real-life stuff
13:18:53 Nick fxb__ is now known as fxb (
13:19:04 Quit Captain_A (Read error: 110 (Connection timed out))
13:19:28peturthat was on 20/8
13:19:44Domonokyhm, nobody has a phone numer of austriancoder, and ask him whats up ? would be nicer than just fail him on gSoc
13:20:17 Join Natanzuelo [0] (
13:20:33peturthat is why I set him a generous deadline, following google rules would have failed him already
13:20:57peturmaybe I should try to find a tele number
13:21:02 Part Natanzuelo ("Saliendo")
13:21:20peturbut he could at least have answered one of my mails by now
13:21:27ddaltonwhat was he doing?
13:22:00Domonokybluebrother: any reason why release installing in rbutilQt doesnt work ?
13:22:47bluebrotherDomonoky: you mean installing a release build?
13:22:52*petur goes to lunch
13:23:04bluebrotherit isn't completely finished yet :)
13:23:36Domonokybluebrother: the code tells me, it just needs the correct filename, right ?
13:23:53bluebrotherit shouldn't me more
13:24:05Domonokythen i will fix it :-)
13:24:10bluebrotherbut I haven't looked into that
13:24:31bluebrotherI started looking into the logfile stuff. I now have a nice log viewer tab :)
13:24:44Domonokynice.. :-)
13:26:06bluebrothernow I need to move the saved numbers to revisions ... which is a bit more work
13:26:33 Quit DerPapst (Read error: 110 (Connection timed out))
13:34:48Domonokynow the installing of release builds works :-)
13:36:22 Quit ddalton ("I was using BOFHNet IRC version 1.2 by fmillion - get your copy today from !")
13:37:17 Quit spiorf (Read error: 110 (Connection timed out))
13:37:32 Join spiorf [0] (
13:37:40*petur goes to enjoy real life... back tonight
13:37:46 Quit petur ("later!")
13:37:55rasherLear: Which voice were you talking about for cygwin?
13:38:15bluebrotherDomonoky: nice :)
13:38:47rasherLear: Ah, I read your question wrong. I thought you were pointing one out to me. No, I don't know of any.
13:39:34 Join MySic [0] (
13:40:15 Join Falen_ [0] (
13:42:12 Join Falen__ [0] (
13:43:28 Quit datachild ("Communism is like one big phone company.")
13:49:06 Quit jhMikeS (Nick collision from services.)
13:49:12 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
13:52:36Domonokybluebrother: for the Small and the Complete Installation, we need something to detect an installed bootloader (with version or timestamp), so we dont install the bootloader if it isnt neccessary.. am i right ?
13:53:33 Quit Falen (Connection timed out)
13:53:53bluebrotherwell, on Ipods it's safe to just overwrite the bootloader
13:54:04bluebrotherno idea about the sansa though.
13:54:37Domonokywhere we have the patcher, its not a problem.. but the other targets.. because they need user interaction..
13:54:40bluebrotherI was wondering if it makes sense to store bootloader information in the log file
13:55:25bluebrotheranother way would be to read the Ipod bootloader and md5 it. That way we could detect the version
13:56:13Domonokybluebrother: is this should work on all targets, we need a file with the info, we cant detect the bootloader on all targets..
13:56:32bluebrotherwell, we can't do that on h100 and h300
13:56:34rasherNot to mention, it's impossible to detect bootloader on h3xx or h1xx
13:57:03 Quit My_Sic (No route to host)
13:57:05Domonokythats what i meant, so we need a file, that rbutilqt creates, when it installs the bootloader
13:57:20bluebrotherwell, just add the information to the log file
13:57:47bluebrotherthat way we can display bootloader version in the info tab I'm working on
13:58:19Domonokyhm.. the log is inside .rockbox, am i right ? i think it would be better to have this info in the root (or extra folder) so it persits when you delete .rockbox
13:58:50bluebrotherwell, if a user uses rbutil he shouldn't need to delete that folder manually
13:58:53Domonokybut we could start with adding the info to the log..
13:59:09bluebrotherbut once we have a split between system and settings we can just move the log file
13:59:21Domonokyyes.. thats good..
13:59:46Domonokynow i only need to get that info (version/timestamp) for the bootloader :-)
14:00:02 Quit Falen_ (Read error: 110 (Connection timed out))
14:00:16bluebrotherwe could add the md5sums to rbutil.ini
14:01:36Domonokybluebrother: could work, but then we need to update rbutil, if we release a new bootloader..
14:01:57Domonoky(we could of course build a function to load a new rbutil.ini from a server) :-)
14:02:42 Quit Arathis ("Verlassend")
14:03:18bluebrotherit could also be a file similar to build-info
14:03:46bluebrotherwhich would be required to hold md5sums for old bootloaders too
14:04:25Domonokyyes a bootloaders-info file on the server would be good..
14:04:31bluebrotherotherwise we could only assume it being an outdated bootloader if the md5sum doesn't match
14:05:23Domonokyif we have a bootloaders-info file, we could also use version numbers..
14:05:46bluebrothergiven that the file provides them, yes
14:06:07***Saving seen data "./dancer.seen"
14:06:10bluebrotherotoh a user always wants to have the most recent bootloader
14:06:27bluebrotherso it would be sufficient to detect if the bootloader matches the most recent version
14:06:28Domonokyno, if you put a new bootloader on the server, you have to modify the bootloaders-info file, and there you just bump the versio.. schould work..
14:07:59Domonokyof course with md5 you could automate the updating of the bootloaders-info file.. but there arnt to often new bootloaders..
14:08:08 Join datachild [0] (
14:08:42 Join DerPapst [0] (
14:08:50bluebrotherwell, if I want to detect the bootloader version regardless of rbutil.log I should use md5
14:08:53bluebrotheror something similar. are right..
14:11:10Domonokyso we need someone with write rights on, to create the file :-)
14:15:27 Quit darkraven (Read error: 104 (Connection reset by peer))
14:19:33 Quit MySic ("Miranda IM! Smaller, Faster, Easier.")
14:24:04 Join My_Sic [0] (
14:31:57 Quit Falen__ ("ChatZilla [Firefox]")
14:34:13Domonokycan somebody (bagder?) put this file : domonoky/bootloaders-info"> onto please ??
14:36:09 Quit DerPapst ("So Long And Thanks For All The Fish!")
14:36:19 Join DerPapst [0] (
14:41:08 Join ToHellWithGA [0] (
14:49:33 Quit aliask ("ChatZilla [Firefox]")
14:54:20Nico_P"FS #7660 - Plugin espeak TTS engine for rockbox"... is that by the TTS SoC student ?
14:55:43 Join qweru [0] (
14:56:55JdGordonNico_P: i dont think so....
14:57:55JdGordonNico_P: hows the playback hacking going?
14:58:27Nico_PJdGordon: nothing new... been on vacation and now I have to work for resits :(
14:58:42JdGordonhow many resits?
14:58:44Nico_PI hope to be able to do a bit of hacking though
14:58:52Nico_Pall math
15:00:51 Quit Lear ("ChatZilla [Firefox]")
15:01:50Nico_PJdGordon: the resits are on the 3rd and 4th of september, so after that I should be mostly free to hack
15:02:48JdGordonplenty of time to cram :)
15:03:07Nico_P"cram" ?
15:03:12JdGordonstudy your ass off
15:03:35 Quit killerz ("I-n-v-i-s-i-o-n 2.0 Build 3515")
15:03:40Nico_Pah right, yeah... hopefully I'll be ready :)
15:03:56JdGordonresits because you couldnt do it at the time? or fails?
15:04:09JdGordoncoz if we fail a exam we have to retake the whole subject, not jst the exam
15:04:42Nico_PI failed the math part of my year so they gave me some exams to retake
15:05:20Nico_Pthe rest (CS and everything else) was perfectly fine
15:05:29JdGordonof course :) noone likes math
15:05:39*JdGordon somewhat jelous...
15:05:46Nico_Pwhy ?
15:06:07JdGordonI could be finishing this semester if i was alowd to resit exams instead of redoing them the next semester
15:06:34 Part oxygen77
15:07:24 Join Febs [0] (
15:08:07*JdGordon does ESP thing to get Linus online....
15:08:47*bluebrother lost quite some time during studying because resits were never the same semester
15:10:14daurnimatorJdGordon: knock him on the head
15:10:34*JdGordon doesnt think thats the best idea :p
15:10:50daurnimatorrear a calf instead
15:11:22JdGordonfark, its only 11.10.. i tohught it was 12.10... :'( gonna run out of download limit again this month
15:11:43 Join Robin0800 [0] (
15:11:54daurnimatorwhat you getting?
15:12:04daurnimatorJdGordon: what does 1 hour do anyway?
15:12:14JdGordonfree time starts at midnight
15:12:16 Quit BobShield (Read error: 104 (Connection reset by peer))
15:12:25daurnimatorso... delay what you're doing for 50 mins
15:22:20 Join Vortex [0] (
15:28:18 Join Kingstone [0] (i=A@
15:40:39 Join Arathis [0] (
15:48:05*Domonoky has now the code ready to include the md5sum for the bootloaders into the installlog of rbutil, now someone has to move the bootloaders-info file to ...
15:50:03 Join BobShield [0] (
15:53:31 Join desowin [0] (n=desowin@
15:57:03 Join iamben_ [0] (
16:06:11***Saving seen data "./dancer.seen"
16:07:46 Quit ctaylorr (Read error: 110 (Connection timed out))
16:09:02 Join freqmod [0] (
16:09:30bluebrotherDomonoky: you noticed that Qt has md5 functions shipped?
16:10:41*Domonoky knows it but at the moment i only use the info from the bootloader-info file.. checking for new bootloaders isnt done yet.. :-)
16:12:12 Quit JdGordon ("Konversation terminated!")
16:12:29 Quit iamben (Read error: 110 (Connection timed out))
16:14:56 Quit tvelocity ("Αποχώρησε")
16:16:54*bluebrother wonders if it was feasible to not download that much foo-info files but simply get a php script that combines those ...
16:20:40 Join Gursikh [0] (i=khalsa@tremulous/officialdevannoyer/khalsa)
16:20:51 Part Gursikh
16:24:54 Quit indro (Read error: 104 (Connection reset by peer))
16:26:50Domonokybluebrother: i think this wouldnt be bad.. we now have at least 3 different info files..
16:30:40 Join indro [0] (i=indro@
16:37:16 Join advcomp2019- [0] (n=advcomp2@
16:37:37 Join Kevi [0] (
16:38:21Kevican someone help me? i plugged in my ipod and the e drive only comes on when i sync and goes of when its done
16:38:25Kevihow do i stop this?
16:40:09 Quit pill (Nick collision from services.)
16:40:32 Join pill [0] (
16:41:05 Quit pill (Nick collision from services.)
16:41:16 Join pill [0] (
16:41:22 Join linuxstb__ [0] (
16:41:26 Join toffe82 [0] (
16:41:48 Quit pill (Nick collision from services.)
16:42:01 Join _pill [0] (
16:43:03 Quit Kevi ()
16:49:11 Join BigMac [0] (
16:52:02 Quit Buschel_ (Remote closed the connection)
16:54:38 Quit linuxstb_ (Read error: 110 (Connection timed out))
16:54:51 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
16:55:32 Quit advcomp2019 (Read error: 110 (Connection timed out))
16:56:00 Join ctaylorr [0] (
16:56:22 Join Buschel [0] (
17:00:55bluebrotherDomonoky: shouldn't we use something more descriptive than "rockboxbase" in the log file?
17:01:22bluebrotherQSettings can cope with spaces and stuff, so "Rockbox Base" or similar shouldn't give any trouble
17:02:28Kingstonewill there be a rockbox on iphone?
17:02:34Kingstoneor it's not even a question?
17:02:56 Quit BobShield (Read error: 104 (Connection reset by peer))
17:03:44DerPapstKingstone: trying it here as well huh? :P
17:04:52bluebrotherwill there be someone paying me for working on Rockbox?
17:05:49Kingstonenope, it's an open source project
17:05:58Kingstonei didn't obliged you to do it
17:06:02thegeekI suggest you do it Kingstone
17:06:11Kingstonei don't have an iphone..
17:06:16thegeekthen get one?
17:06:18Kingstonei don't leave in the US lol
17:06:22thegeekebay man
17:06:33 Quit linuxstb__ (Read error: 110 (Connection timed out))
17:06:36thegeekit's even unlocked now
17:06:37Kingstonewell i would want to see it work in order to try and port it
17:06:39thegeekso you can use it wherever
17:06:42Kingstonenot working here
17:06:46Kingstonenot entirely true
17:06:56Kingstoneonly suitable networks
17:07:01Kingstonenot in my country
17:07:01thegeekthere is a 100% hardware unlock
17:07:08thegeekand two different sim unlock solutions
17:07:12thegeekall of them work
17:07:23thegeekand there is a supposed software unlock coming soon too
17:07:27thegeekso yes
17:07:38Kingstonewell still even if i get it to work, i'm not going to pay 4000$ for this one
17:07:39thegeekit will work in your country
17:07:46thegeekoh I see
17:07:51thegeekso you expect the devs to do so?
17:08:03Kingstoneif someone of you
17:08:05Kingstonealready has one
17:08:12Kingstonebecause he wanted to buy it for himself
17:08:19Kingstoneand is willing out of his own good will
17:08:26Kingstoneto also work/try this project
17:08:29Kingstonebe my guest
17:08:38Kingstonei never said anyone must've do it
17:08:49Kingstoneme myself helped develope stuff for ipod
17:08:54Kingstonebut i can't do it for iphone
17:09:10thegeekI you've developed for rockbox you should know how it works
17:09:30thegeekand you should also know why noone can say for sure if there will be an iphone port
17:10:50bluebrotherand it shouldn't be hard to guess that there is not much interest in the iphone.
17:10:55bluebrotherat least not in this channel
17:11:39 Join Nick_Brackley [0] (
17:15:16bluebrotherDomonoky: what happened to moving the configuration of the talk creation window to the configure dialog?
17:17:05 Quit spiorf (Read error: 110 (Connection timed out))
17:17:41 Join spiorf [0] (
17:26:16 Nick _pill is now known as pill (
17:27:23 Join petur [0] (n=petur@rockbox/developer/petur)
17:29:23Domonokybluebrother: i never finished it.. it was getting complicated, so i will rewrite its someday, when i find time :-), if you dont beat me to it :-)
17:31:29Domonokyand about the Section "rockboxbase" in the log file, rename it if you like.. when i named this section, we didnt have something to show them :-)
17:32:43n1sAnybody using the recording trigger around?
17:34:32n1sI was thinking about changing the recording trigger screen to a regular menu, that will of course remove the peakmeters at the bottom, are those very useful?
17:34:41bluebrotherI'm about committing the log rework ... just trying on windows first.
17:36:08 Join lazka [0] (
17:43:32 Join lee-qid [0] (
17:48:17 Join SirFunk [0] (
17:49:11 Quit Nick_Brackley ("ChatZilla [Firefox]")
17:49:13 Nick advcomp2019- is now known as advcomp2019 (n=advcomp2@
17:53:22_jzi found a bug in the 0924 version
17:53:49_jzwhen hitting "next" with crossfading enabled, but right before the end of the song, some reaaaally weird shit happens
17:54:03_jzno not hitting next but selecting some other track
17:54:14bluebrotherwhich version are you referring to?
17:54:24_jzwhen you select some other track and expect it to crossfade to that track, it instead breaks and start playing the next song in the playlist
17:54:33_jzthen after some seconds crossfades to the track you selected
17:54:36_jzthat's awful :/
17:54:46_jzbluebrother: on second
17:55:03n1s_jz: does it happend every time?
17:55:13_jzi think so
17:55:16_jzi'll test again
17:55:30*bluebrother wonders what _jz meant with "0924 version" then
17:55:42_jzi meant 0824
17:55:45n1sI guess it has to spin up the hard drive to load the track you selected and keeps playing what's in the buffer until it can get the file
17:55:45_jzbut i was wrong :)
17:55:47_jzit's 0823
17:55:56_jzit worked before !
17:56:08_jzi did any kind of acrobatic crossfading from beginning / middle / end of tracks before
17:56:16bluebrotherplease always use complete version numbers.
17:56:22_jzyep sorry
17:56:42 Join Chewie[] [0] (
17:57:05 Join Phalangees [0] (
17:57:09bluebrothercan you fast forward to the end or does it only occur if you let it play all the time?
17:57:15n1s_jz: but if you are at the end of a track, spinning up the drive takes about 1-2 seconds, it will start the next song in the buffer until it can read the new file
17:57:33_jzi got flash in my sansa
17:57:43bluebrotherhmm, I can't reproduce it here. Crossfading works fine.
17:57:47_jzhmm i can't reproduce it now i reduced the fade-in/fade-out times to 10 secs instead of 15
17:57:51n1saha, it could have helped if you said which device you used
17:57:53_jzmaybe it only happens at 15sec
17:58:31bluebrotherfade-in delay? fade-out?
17:58:44_jzno delay, duration
17:58:46_jzI try again
17:59:26bluebrotherstill works here, even with duration set to 15s
18:00:16_jzI just redone it right now
18:00:30_jzmaybe you don't have just to skip to the end of the song for this to happen
18:00:38_jzbut maybe listen to the end of the song
18:00:39bluebrotherI suggest to update to the most recent build before doing further tests.
18:00:45_jzyep good idea
18:00:49bluebrotherwell, can you still reproduce it?
18:00:53_jzi'll do it right away
18:01:01_jzi'll tell you with most recent version in a few minutes
18:01:04_jzi just reproduced it
18:01:12 Quit Buschel ()
18:01:38_jzbtw i had a feature idea... soooo simple and useful ... :)
18:02:01_jzin the contextual menu for one track having the "play next" or "enqueue" to insert that track just after the currently playing track !
18:02:19peturn1s: re the rec trigger screen, I wouldn't remove the peakmeters there, I would even think it were better if the trigger points were marked better on them
18:02:19_jzno need to have a long queue, with a queue manager (though it would be nice ;) just one queue would be awesome already
18:02:22_jzwhat do you think ?
18:02:44kk_jz, er, you mean like "insert next"?
18:02:58n1spetur: ok I won't touch it, it's just that the custom menu is anoyying :-)
18:03:06kkoh, dammit freenode's nickserv.
18:03:08_jzkk: "insert as the next track in the playlist"
18:03:17kkright. that's already there
18:03:22_jzhoooo ?
18:03:24 Quit Xerion (Read error: 104 (Connection reset by peer))
18:03:27_jz<3 i love you !! <3
18:03:33_jz<3 i love rockbox ! <3
18:03:39_jzhow is it called ?
18:04:01Domonoky"insert next" :-)
18:04:03kkcontext menu -> playlist -> inster next
18:04:07kkinsert, rather
18:04:10 Nick kk is now known as krazykit (n=kkit@
18:04:26*petur reads mail from austriancoder
18:05:02_jzhaaa ! in the playlist submenu
18:05:11_jzthis is a place where i didn't even thought looking for it :)
18:05:17bluebrother_jz: just wanted to tell you ;-)
18:05:19_jzyou guys seriously kick ass :)
18:05:29n1s_jz: we do have a fairly decent manual...
18:05:35_jzi know
18:05:38_jzi read most of it
18:05:51_jzanother thing that might not be as simple :
18:06:13_jzwhen crossfading, if skipping into the incoming track, that the volumes remains the one it would have been when you press the skip button
18:06:16***Saving seen data "./dancer.seen"
18:06:17_jzhmm maybe it's not clear
18:06:42_jzif the track is "entering" with a level of 5/100 and you skip to 30sec, that the playback takes back at 30sec at 5/100 volume ?
18:06:42bluebrotherI agree that the trigger setup screen is really awful
18:07:25bluebrotherI'd like to have a "display settings in lists" like FS #5833 did
18:07:36bluebrotherbut having one screen behaving different is ugly
18:07:44krazykit_jz, not sure about your fractions, but the volume control's dB level won't change. if the song is louder half a minute in, it'll be louder.
18:08:29krazykitor am i misunderstanding what you're asking?
18:09:26_jzkrazykit: i'm talking about crossfading
18:09:41_jzkrazykit: when track A is fading out, track B is fading in
18:09:56_jzbut as soon as the fading happens, you see track B control on the screen
18:10:08_jzbut if track B is very low volume, you can't hear it at all
18:10:23_jzand you just skip 15 seconds (which takes about 2 seconds), then it comes with full volume
18:10:32_jzinstead of continuing the fade in
18:10:48krazykiti don't know. i hate crossfading, so i don't use it
18:10:58_jzha :)
18:11:05_jzit's good for pseudo-DJing at parties :)
18:11:26_jz(which is better than DJing at pseudo-parties ;)
18:11:32bluebrotherbut in that case you won't skip at the end of a song ;-)
18:11:39_jzno i wont ;))
18:11:42_jzjust after the intro maybe
18:11:51n1sbtw regarding the UI I had an idea before: How about grouping fonts into families, and requiring each family to have one font no larger that say 10p So that we have a smaller fallback font wherever we need it (where we now fall back to sysfont) so if I had foo-font-32 in the lists and went into the quick screen it would select the largest one that fits in the family, and fall back to the 10p version if there is no font between 32 and 10
18:11:53_jzor right in the middle of the action if things are getting nasty :)
18:12:11_jzn1s: reordering the fonts would be a great idea indeed
18:12:26_jzwhy not grouping them by family and by sizes ?
18:12:49_jzbecause changing a fond with a tightly made wps really depends on the size and nothing else...
18:13:55n1s_jz: this is more a suggestion for the rest of the UI, not so much wps
18:14:19_jzbut it would help when changing just the font, not the whole wps
18:14:23bluebrotherbesides, falling back isn't that easy −− you currently can have only one font
18:14:24n1sCurrently we try to use only one font and fall back to sysfont which is about 8p
18:14:33bluebrotherthe only exception is sysfont which is always present.
18:14:51_jzBug reproduced !!!
18:14:56_jzthe track was 8:07
18:15:03n1sbluebrother: I mean that if we need to fall bak, just load glyphs from a smaller font in the same family
18:15:07_jzat 8:00 i switched to an Adam Beyer track
18:15:12_jzthen it went to the The Hacker track
18:15:16_jzthat was the next in the playlist
18:15:21_jzit played for 15 to 20 seconds
18:15:30_jzthan cut / here comes the Adam Beyer track
18:15:59_jzmy first track just fade out, the The Hacker track didn't fade in but just happened at the beginning, and same for the adam beyer track
18:16:06bluebrotherI assume you already updated?
18:16:20_jzbut I listened to 1min of my first track before arriving to 8:00
18:16:21n1sbluebrother: because IMO it is not realistic to make the UI able to use a 32 pixel high font everywhere...
18:16:22_jzyes I did
18:16:47bluebrotherthen it seems to be something for the tracker ...
18:17:14n1sThis would cause extra disk spinups when using very large fonts or another or bigger glyphcache
18:17:19_jzi hate trackers...
18:17:25_jzbut i like rockbox sooo much ....
18:17:39bluebrothertry to search it first. Perhaps someone already reported it ;-)
18:18:30_jzdoesn't look like it
18:19:29n1sBecause in most places where sysfont is used a larger font will fit, just that not all fonts will fit
18:20:05n1salso we have screens that do not fall back to sysfont but instead just draws stuff offscreen instead, not nice
18:20:36n1sfor example the set Time/date screen
18:22:34n1sthe Trigger screen draws things on top of each other with a large font (32)
18:25:04n1sAnd the sysfont strings are a nasty hack, making parts of rockbox impossible to translate to certain languages, and making voice files bigger
18:25:59 Quit petur ("switching...")
18:26:08 Join petur [0] (n=petur@rockbox/developer/petur)
18:26:26 Join Arathis_ [0] (
18:26:28bluebrotherwell, I agree that we need a better solution than the current one. I'm just wondering if grouping fonts is a good one ...
18:26:48 Quit Arathis (Nick collision from services.)
18:26:54 Nick Arathis_ is now known as Arathis (
18:26:57 Join Spiorf_ [0] (
18:27:24 Quit Spiorf_ (Client Quit)
18:27:25n1sbluebrother: I am interested in other approaches :-)
18:28:12 Join The-Compiler [0] (
18:28:27peturimho, we need 1) multiple fonts loading 2) GUI capale of coping with different fontsizes
18:29:14bluebrotheran idea would be to allow screens to scroll
18:29:14n1spetur: but how do you make the quick screen cope with only being able to display one line? for example
18:29:21peturthe reason stuff like the recording screen is still the way it is, is because we are waiting for viewports to rewrite the GUI
18:29:31bluebrotherbut I'm not sure if that would work nicely ;-)
18:30:39peturn1s: in the same way that a browser isn't usable anymore when the font is sooo big you can't read anything
18:31:21peturif the user selects a font that is too big, it is his problem, he'll change back ;)
18:32:26 Join Siku [0] (
18:33:19n1sI think it would be nice if a gui screen could just say "I want a font that is no taller than 10pixels" and then an available font is the same family as the userfont would be loaded or if it din't exist sysfont could be used
18:34:57peturor we have two fonts to load: one for WPS/..., the other for menus/...
18:35:03n1sof course this would need some changes to font loading an the glyph cache as it would be bad to have to unload the user font on many screen swithces
18:35:23n1spetur: that's an entirely different issue
18:36:23*bluebrother leaves for a while
18:36:27n1sbut of course the wps could probably make use of the fall back system and specify font height, wouldn
18:36:56n1s't work well for some wps's but maybe better than how we do it now
18:37:47Domonokybluebrother: this new info tab looks nice :-)
18:39:05bluebrotherthanks :)
18:39:41bluebrotherit shouldn't be too hard to extend it to a "cleanup" tab (which removes all files in a section but a specified version string)
18:40:41Domonokyi think i will now commit my changes to the bootloader installation.. but at the moment the ini file points to my webspace for the bootloaders-info..
18:41:33bluebrotherno real issue imo −− we haven't released yet ;-)
18:41:46Domonokytrue.. i will commit :-)
18:41:47bluebrothermaybe I'll give caching a shot later today.
18:42:04bluebrotherI think that's the only thing that is missing to be en par with the wx version
18:42:16 Quit pill (Nick collision from services.)
18:42:25bluebrother(but imo we already have some improvements over the wx version ;-)
18:42:34 Join pill [0] (
18:45:02Domonokyyes, we are now better then the wx version.. and caching.. i dont know if we really need this..
18:46:33Domonokyhm, for the complete and small installation.. should it ask which build/daily/release version it should install, or should it just install the current ?
18:46:41 Quit pill (Nick collision from services.)
18:47:04 Join _pill [0] (
18:47:27n1sDomonoky: that question will probably confuse our users :-)
18:47:53Domonoky:-) true
18:48:16Domonokyso just install the current.. and if it isnt aviable at the moment.. the daily :-)
18:49:07_jzOK I submitted my bug
18:49:12_jzI hope it's clear enough
18:49:18_jzand I hope it will be helpful
18:49:45_jzplease vote for it! ;))
18:50:17n1sme never understood the voting in the tracker...
18:57:43PhalangeesCould someone possibly help me out with displaying a bitmap in a plugin?
18:58:08PhalangeesI'm fairly new to coding and everything so it's pretty confusing to try and understand by looking at others code
18:58:12n1sPhalangees: it's done in many plugins, you could look at them
18:58:29PhalangeesI'll try looking at the simpler ones.
18:58:39PhalangeesThanks nld
18:58:51Phalangeesor n1s? :p Thanks
19:00:55 Join datasleep [0] (
19:01:14 Quit datachild (Nick collision from services.)
19:01:50 Nick datasleep is now known as datachild (
19:02:24_jzOK now you'll have to pay the price for having me create an accout on the flyspray :)
19:03:18_jzhey ! this seems directly related to my crossfading issue !
19:03:41_jz.... or not... ?
19:12:11 Quit Siku (Nick collision from services.)
19:12:17 Join Siku [0] (
19:15:09 Join Adler [0] (i=47b7b9d7@gateway/web/cgi-irc/
19:15:23 Quit SaLoMoN (Read error: 104 (Connection reset by peer))
19:17:25 Join MyAdler [0] (i=47b7b9d7@gateway/web/cgi-irc/
19:17:25 Quit Adler (Client Quit)
19:19:00 Join Rondom [0] (
19:20:52qweruIn the rockbox file browser, is there a possibility to select more than one object at a time?
19:21:04Nico_Pqweru: no
19:21:05qweruWould be good to be able to move/delete more than one file.
19:21:05 Join chrisjs169 [0] (
19:21:16 Join merbanan [0] (n=banan@
19:21:22qweruI thought so because I couldn't find anything in the manual.
19:21:46MyAdlersomeone should make that possible...
19:24:41 Quit MyAdler ("CGI:IRC (EOF)")
19:27:45_jzthe behaviour of the "rec" key on a Sansa should be assignable to different actions
19:28:02_jzlike during playback this could lead directly to the pitching of a track, this would be awesome :)
19:28:22*bluebrother would _really_ like people using real names in the tracker :(
19:28:50_jzbluebrother: was this directed to me ?
19:28:54_jzi put my real first name now
19:29:12bluebrotherwell, not directly. There are tons of other users using nicknames only.
19:29:26peturFebs: reading your comment on 7664, close it?
19:29:41DerPapstadd the real name policy to the tracker
19:29:56bluebrotherI'm fine with nicks on irc, but for other communication I prefer to know with whom I'm talking
19:30:14bluebrother(and for the regulars in here I know their real names ;-)
19:30:49n1sheh, a quick reordering of the playlist_info struct saves over 700bytes, sacrafices some readability though, I wonder why we can't tell gcc to do this itself...
19:30:53_jzi don't display my first and last name plublicly "anywhere
19:30:59webmindbluebrother, I've changed my realname more often then my nick
19:31:02webminddoes that help?
19:31:05_jzi mean i don't think rockbox trackback system is evil or whatever
19:31:17_jzbut that is private personnal data, i'm paranod about it, and i hate forms :)
19:31:18bluebrotherwebmind: hehe ... I guess that is really rare
19:31:38_jzwebmind: have you married different times ? :)
19:31:43bluebrotherwell, we don't enforce real names in the tracker (yet) ...
19:31:53webmindonly my firstname btw
19:32:15bluebrotherwhy changed it? You disliked it?
19:32:34webminddidn't like my firstname, started using different new names depending on where I was
19:33:00webmindmy nickname is slightly more consistent than my realname
19:33:20Febspetur: I wouldn't object to closing it.
19:33:48webmindanyway, I prefer nicknames as they're more unique
19:36:41peturwtf... sim crashes when trying to play a file :(
19:38:09*bluebrother shakes head at this top-posting mail :(
19:42:06 Join kk [0] (n=kkit@
19:49:02 Join Lear [0] (i=chatzill@rockbox/developer/lear)
19:52:27 Quit krazykit (Read error: 110 (Connection timed out))
20:00:15 Quit thegeek ()
20:04:11 Join thegeek [0] (
20:06:19***Saving seen data "./dancer.seen"
20:16:04 Join shot1 [0] (
20:16:31 Join jhulst [0] (
20:16:44 Quit jhulst (Read error: 104 (Connection reset by peer))
20:17:46 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
20:19:30 Part shot1
20:21:34 Join webguest22 [0] (i=50d81efc@gateway/web/cgi-irc/
20:22:04webguest22petur: Any news from austriancoder?
20:22:26peturI got code from him
20:22:40webguest22petur: Wow! interesting!
20:22:53peturgot some link errors to fix before trying
20:23:08peturbut he says that both serial and UMS drivers do not yet work
20:24:19*petur wonders if he should put it online himself or wait for austriancoder to do that...
20:24:44webguest22petur: well, it's a good start now that tx works atleast? getting the serial should not be that difficult (considering MrH's work) ?
20:25:56peturjust need some time to look at the code...
20:31:09webguest22petur: After sending started to work, has he been able to work a lot on getting the stack and serial/UMS to work?
20:32:02 Quit desowin ("use linux")
20:32:08peturwebguest22: how could I know? He hasn't been online since then. He did work on it, that I can see
20:33:28webguest22petur: sorry, i'm just curious what he wrote in the mail :) i'll wait and look at the code when it's up
20:34:04 Nick _pill is now known as pill (
20:35:50peturThis is only the technical stuff, I removed the gsoc specific talk
20:36:08 Quit ToHellWithGA ("You know you'll miss me a lot.")
20:36:19 Join DataGhost [0] (
20:40:28 Nick Vortex is now known as Vortex375 (
20:40:36peturlooks like I miss at least two files :(
20:48:04 Quit ompaul ("Leaving")
20:51:39 Join ompaul [0] (n=ompaul@freenode/staff/gnewsense.ompaul)
20:51:44 Quit darkless (Read error: 110 (Connection timed out))
20:52:14PhalangeesHow do I fix the warning "function declaration isn't a prototype?"
20:52:59peturdid you write myfunction(); ?
20:53:18peturchange it to myfunction(void);
20:53:24Phalangeesah. Ok, thanks
20:53:41xaqueFer the record, I am really enjoying rockbox. Thanks to all the devs!
20:54:12*petur finds the missing usbstack functions and wonders why the linker doesn't
20:56:26n1spetur: are the files included in SOURCES ? so they are compiled?
20:56:29 Join desowin [0] (n=desowin@
20:57:45peturn1s: looks like it...
20:57:47Nico_Ppetur: you can tell ac he's welcome to use my git repo
20:57:59Nico_PI can give him push access
20:58:11peturI'll mail it to him....
20:58:29bluebrotherwow ... even more users completely ignoring the list etiquette
20:58:40peturor rather, you could...
20:58:52Nico_Ppetur: true :)
20:58:58Nico_PI'll do that
21:00:01*Domonoky has now a nearly working small and complete installation for rbutilqt :-)
21:03:23Febsbluebrother: I think Kane Brolin doesn't understand what top-posting is. Notwithstanding his/her LONG post purporting to explain it to others.
21:03:31peturNico_P: good luck on getting a quick answer ;)
21:04:30 Join robin_0800 [0] (
21:06:45 Quit Lear ("ChatZilla [Firefox]")
21:07:52Nico_Ppetur: I don't really need an answer ;)
21:08:03peturlucky you!
21:08:54webguest22petur: so what will the GSoC result be?
21:11:01peturstill looking to get it compiling...
21:12:48 Quit ctaylorr (Read error: 110 (Connection timed out))
21:19:07bluebrotherFebs: well, he might understand it or not. But the way re reacted is completely inappropriate imo
21:20:31bluebrothermaybe a rockbox-blind list which would be allowed to top-post would be helpful?
21:22:12*n1s is glad he doesn't subscribe to the users list, dev list has much less noise :-)
21:22:27Domonokybluebrother: what do you think, would it be good to merge a theme.cfg with .rockbox/config.cfg after installing a theme ( the it would already be selected in rockbox, when you reboot)
21:23:01*bluebrother is about to unsubscribe from the users list
21:23:24bluebrotherDomonoky: I think a more complete configuration-editor (possibly limited to some sane defaults) is better
21:24:18bluebrotherdos line endings again ... ;-(
21:24:31 Join darkless [0] (
21:24:36kkthat thread is STILL going in the user list? ugh.
21:25:01bluebrotherit is. And some seem to be really brain-dead.
21:25:17*bluebrother wants to hand out some "Merkbefreiungen"
21:25:28 Nick kk is now known as krazykit (n=kkit@
21:25:47Domonokybluebrother: dosline endings in my code ? ifso i am sorry
21:26:32bluebrotherin the ts file ... and I had some translations pending.
21:26:47bluebrotherwell, not much of a problem. It's only the translation :)
21:27:12Domonokyups, this was by accident.. didnt want to commit the .ts file..
21:27:14bluebrotherbut it's kinda strange as svn is supposed to handle the translation itself properly :/
21:30:59 Quit The-Compiler (Remote closed the connection)
21:30:59 Join matsl [0] (
21:36:26bluebrotherwouldn't it be better to let BootloaderInstaller ask for the OF file itself if needed?
21:38:54Domonokybluebrother: undoAbort is there because some installers call abort when they are finished.. so the button changed from abort to ok, but we want to install more, so i change the button back :-) ( i didnt want to find and change all other places, so they dont do the abort...
21:39:36Domonokyabout the asking for the OF file.. i thought about the possible commandline interface, and the we dont want gui windows in the installers..
21:40:19 Quit pill (Nick collision from services.)
21:40:37 Join pill [0] (
21:41:18 Quit pill (Nick collision from services.)
21:41:43 Join _pill [0] (
21:42:09bluebrotherwtf ...? It doesn't build on linux
21:43:04 Join nerochiaro [0] (
21:43:14bluebrotheralso, couldn't we call the installer stuff from MultiInstaller from rbutilqt.cpp?
21:43:27bluebrothercurrently it looks like much duplicated code to me
21:44:27Domonokyyes, there is much duplicated code there.. but the functions are unfortunatly not exactly the same.. but i will try to improve :-)
21:44:56bluebrotherthis is really strange: a clean build works fine on windows but not on linux
21:45:25Domonokystrange.. it doesnt build .. or doesnt run ?
21:45:34bluebrotherit doesn't build
21:45:53Domonokystrange.. and the error is ? :-)
21:47:37chrisjs169random question, but is there a reason why the battery menu doesn't appear on the Sansa any more?
21:47:46bluebrotheraaah ... found it :)
21:51:39bluebrotherDomonoky: fix committed
21:55:06 Join x1jmp [0] (
22:02:04Phalangeeshow do you do square root in C? sqrt();
22:02:26bluebrotheryou could also pow(x, .5) ;-)
22:02:56PhalangeesEither way will do what I need. :D Thanks again!
22:03:11 Join jhulst [0] (
22:04:15 Join tvelocity [0] (
22:05:39 Quit desowin ("use linux")
22:06:21***Saving seen data "./dancer.seen"
22:06:43 Join Lars_G [0] (n=lars@unaffiliated/lars-g/x-000001)
22:06:53PhalangeesThe sqrt and pow both give errors.
22:06:54Lars_GI'm about to pull. will I find a lot of nifties?
22:07:00Lars_GBtw I wanted to tell you something.
22:07:02PhalangeesDo I need math.pow or something like that?
22:07:36Phalangeesanother include?
22:07:43Lars_GI was the other day in one of the very few (2 or 3) apple distributors (not even official stores) in my country, and flaunted my rockboxed nano to everybody. they were all awed
22:09:30merbananPhalangees: you need to link with the math lib
22:10:12Phalangeesmerbanan: How do I do that? Just do an #include
22:11:06merbananPhalangees: search google
22:11:09 Join amiconn [0] (
22:12:40*amiconn back from pauze ;)
22:13:07amiconnpetur: Why are the peal clip counter lang strings catchall? They are only used on recording targets...
22:13:34 Join gtkspert_ [0] (
22:14:49*amiconn also wonders what that clip counter would be good for :\
22:16:40 Quit Arathis ("Verlassend")
22:18:34PhalangeesDoes anyone know of any plugins that use sqrt() or pow()?
22:20:02bluebrotherPhalangees: just grep for it ;-)
22:20:11bluebrothervu_meter.c uses fsqrt
22:20:31bluebrother(declared in fixedpoint.h)
22:20:45 Quit gtkspert (Read error: 110 (Connection timed out))
22:21:00PhalangeesSo fixedpoint.h has sqrt function? Sorry if I'm being really stupid. I'm fairly new to coding.
22:21:17bluebrotherregarding rockbox, yes.
22:21:37Phalangeesdoes it have pow() as well?
22:22:20 Quit webguest22 ("CGI:IRC")
22:22:29PhalangeesWell, I don't really need pow. I can just do it the long way. Thanks for your help bluebrother.
22:23:26bluebrotherlibspeex for example has a pow function.
22:23:43bluebrotherit's just not part of the fixedpoint plugin lib.
22:24:12bluebrotheroh, and the calculator plugins seems to have a pow function
22:26:45 Nick Genre9mp3 is now known as XavierLimn (n=yngwiejo@rockbox/contributor/Genre9mp3)
22:27:33XavierLimnDamn it, I just tried to flash an svn bootloader in H300 and I get the same behaviour as my H100 (from SVN again)
22:28:00XavierLimnCurrent bootloader will not manage to boot rockbox. Fortunately the OF works like a charm
22:28:08 Join Buschel [0] (
22:28:51XavierLimnI will update the FS bug report, it seems that something is wrong with all iriver Hxxx bootloaders
22:29:06amiconnXavierLimn: You seem to like it dangerous...
22:29:28*amiconn doesn't even dare to flash the H1x0 prerelease bootloader needed for rombox
22:29:59*petur has the same feeling re h300 bootloader
22:30:43amiconnThe iriver Hxx0's are the most delicate targets regarding bootloaders
22:31:13amiconnAll the others are recoverable one way or another without expensive equipment
22:32:46peturamiconn: re clipcounter strings: you're right, must only be for targets with recording
22:32:54Domonokybluebrother: i have now removed the multiinstaller file, and incorporated it into rbutilqt.cpp
22:32:59amiconnYeha, but it's not that simple
22:33:03bluebrotheroh, nice :)
22:33:07n1schrisjs169: someone decided that since there were no rplacement batteries for it, having a menu for the battery capacity is tupid
22:33:17bluebrotherI started looking into http caching ...
22:33:19amiconnAll bitmap targets have peakmeters, but only recording targets need the clipcounter
22:33:36amiconnSo just taking out the .lang strings isn't enough
22:33:56peturthe lang strings are only used in recording.c
22:34:16 Quit Buschel ()
22:34:36 Quit Lars_G ("Leaving")
22:34:36amiconnThey're used in settings_list.c too
22:34:54amiconnIf I just set the strings to none for non-recording, Ondio SP doesn't compile
22:34:56 Join linuxstb [0] (i=5343d4aa@rockbox/developer/linuxstb)
22:35:05peturhmmm my bad... will fix it
22:35:15XavierLimnamiconn: well I saw the changelog from r12546 (that I had compiled and worked back then), and it seemed pretty safe to flash it.
22:35:34XavierLimnof course I could be horribly wrong and brick it.
22:35:43amiconnAnd there is some code in peakmeter.c that should be #ifdef HAVE_RECORDING
22:36:09n1sXavierLimn: why are you using svn bootloaders?
22:36:25XavierLimnLinus must investigate this, it will just hang on a SET_REG macro...
22:36:26amiconnXavierLimn: I wouldn't flash a Hxx0 bootloader even if I compiled it from the exact same svn revision as the official one
22:36:34peturamiconn: I'll see to it...
22:36:44*petur bows and gets to work
22:37:24XavierLimnn1s: I wanted to test a new feature that was added by linus after r12546. And while I was at it I decided to just upgrade to the latest one seeing that there weren't many changes in
22:37:31amiconnpetur: I got curious when I saw that all the non-recording targets (well, except the player which has no peakmeter at all) got 200+ bytes binsize increase...
22:38:03amiconn(i.e. Ondio SP, iPod G1..G2 and the minis)
22:38:11amiconnErr, G1..G3
22:38:30amiconnI also wonder what's the exact purpose of the clip counter...
22:38:42 Nick XavierLimn is now known as Genre9mp3 (
22:38:48amiconnI know what it does, but it doesn't say much about the recording quality
22:38:48peturfor stealth recording
22:39:24amiconnEven hundreds of slight clips can be unnoticeable when listening, but one big clip might ruin a whole track
22:39:46n1sXavierLimn: I am pretty sure that LinusN or whoever decides to release a new iriver bootloader will fix that when the time comes for another release, keeping it working inbetween releases is IMO not very useful
22:39:55amiconnBut that's something we cannot detect
22:40:19 Quit matsl (Remote closed the connection)
22:41:09n1samiconn: regarding binsize, just reordering some structs can save quite a bit here and there...
22:41:58 Join Domonoky_ [0] (
22:42:10n1sgot about 700 bytes from reordering one in playlist.h don't know if much readability was sacraficed though since everything seemed to be added to the end when needed so most things were in no particular order
22:42:47 Quit merbanan (Remote closed the connection)
22:43:23Domonoky_only three more commits to rbutilqt, and the frontpage will be dominated by dominiks :-)
22:44:22 Quit qweru ("KVIrc 3.2.0 'Realia'")
22:44:24*petur hurries up
22:45:41 Quit tvelocity ("Αποχώρησε")
22:48:29 Quit jhulst (Remote closed the connection)
22:50:15 Nick _pill is now known as pill (
22:57:42*Domonoky_ lets the last two commits for "dominik domination" to bluebrother, and goes sleeping..
22:59:06 Quit Domonoky_ ("Trillian (")
22:59:29 Quit Domonoky (Read error: 110 (Connection timed out))
23:03:28*linuxstb is also curious about why clip counting is helpful
23:05:11peturso you can adjust the gain?
23:06:22linuxstbYes, but why is the number of clips important? Isn't the fact that it's clipped once enough?
23:07:46peturif it clipped more, the gain is way too high, if it only clipped some, it is just a bit too high or somebody yelled in your mics
23:09:05peturother requests in that area are a logfile containing 1) timecodes of clipping 2) timecodes of gain changes
23:09:53n1sthe logfile sounds kind of overkill to me but I'm no recorder really...
23:10:17peturfor me too, but some recorders are real freaks ;)
23:10:45linuxstbBut can't that be done after the fact? (clipping detection)?
23:10:56linuxstb(I'm talking about the log)
23:11:22 Join w-soap [0] (
23:11:23 Quit Robin0800 (Read error: 110 (Connection timed out))
23:11:30peturyes... but gain changes not, and that would be handy for postprocessing
23:11:37amiconnImo the important thing about clipping is whether it's audible. Slight clipping often isn't.
23:11:55amiconnBut the amount of clipping isn't detectable of course
23:12:06linuxstbpetur: Yes. I would agree the gain change log could be useful.
23:12:24amiconnAnd I rather let it clip a bit than changing the gain during recording (for music, obviously)
23:12:45peturamiconn: sometimes slight clipping can sound horrible and strong clipping not...
23:13:20*petur always uses AGC_SAFETY
23:14:18*amiconn wouldn't use agc for music
23:14:36amiconnThat doesn't mean agc is useless - it's useful for voice recording
23:14:42peturyou do not know AGC_SAFETY me thinks...
23:15:08peturit is a wrong name, it is more a clipping protection
23:15:18peturit only lowers the gain
23:15:23amiconnI do know what agc safety does
23:15:37amiconnStill I wouldn't use it for music recordings
23:15:45 Quit freqmod (Remote closed the connection)
23:16:26linuxstbWouldn't a change in gain part-way through the recording be better than future clipping?
23:16:31peturI rather have a slight (and inaudible) gain change then clipping...
23:16:52peturit has saved many of my recordings
23:17:09linuxstbYou could always lower the gain of the first part of the recording anyway, to match the later part...
23:17:12*DerPapst agrees petur
23:17:41peturhi DerPapst, goodmorning ;)
23:17:50DerPapstmorning petur ;)
23:22:56 Quit BHSPitMonkey (Remote closed the connection)
23:24:54 Join BHSPitMonkey [0] (n=stephen@
23:27:38amiconnpetur: agc_safety would cause a huge drop in gain if there's a big spike (be it in original audio or electrically induced)
23:27:48amiconnI'd rather let it clip in such a case...
23:28:11peturit would if there was a long spike...
23:28:27 Part n1s
23:30:26*DerPapst transfers smiley to other window...
23:30:26*amiconn never recorded music with agc
23:31:00*amiconn recorded more than today back in analog times
23:31:01peturand AGC has a time constant so you can avoid having it react to short stuff like a clapper or a yeller
23:31:21*petur records every concert he visits
23:31:37amiconnToday it's not worth recording music, as you can buy it with higher quality on CD
23:32:05peturyes and no
23:32:14 Quit bospaadje (Remote closed the connection)
23:32:54 Join bospaadje [0] (
23:33:01petur1) live CD's are rare 2) some bands are better live than on CD 3) most CD are over-processed and have almost no dynamic range (they even clip!)
23:33:21peturrip a CD to wav and look at the waveform, it is horrible
23:33:33*amiconn doesn't care about (1) and didn'texperience (2)
23:33:45peturand I like live performances more :)
23:35:04amiconnLive performances are nice, but only when attending them live imho
23:35:36peturif I listen to them again, it's like being there again :)
23:35:47*amiconn tries to remember when he last visited a lve concert...
23:36:18amiconnMust have been almost 3 years ago...
23:36:25 Quit pill (Nick collision from services.)
23:36:35amiconnFor normal listening, I much prefer studio recordings
23:36:45 Join _pill [0] (
23:42:57peturamiconn: as I said, some bands are live way better. Two of my fav bands are. And songs evolve and become better because of the live playing
23:44:38 Quit seablue ("life, death, life, death")
23:47:06peturwhat's up with that x5 build... some build-server failed?
23:48:43peturthe zip seems to be available though...
23:48:46amiconnLooks like th etable generation failed
23:49:01amiconnzip is there and has the correct version (according to rockbox-info.txt)
23:50:35peturthere is even delta info (/me hopes amiconn is happy with the last line)
23:51:10amiconncygwin is acting up :(
23:54:40 Join BobShield [0] (
23:55:28 Quit davina (Remote closed the connection)
23:57:56 Quit darkless (Read error: 113 (No route to host))
23:58:21 Join darkless [0] (

Previous day | Next day