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 2006-03-28

00:00:29kclaflinuxstb : did u see that my issue with tagcache was coming from voiceui ?
00:00:48kclafi disabled * voice things, and it's ok now
00:00:52linuxstbkclaf: Yes, but I can't think why that could happen.
00:01:46 Join miner49er [0] (
00:02:04miner49ercan anyone help me with a coding issue?
00:02:33herz42preglow: here is a patch for the backlight bug. I hope it also builds for targets with remote etc.
00:03:27preglowahh, lookie there, yeah
00:03:29preglowthat was nasty
00:04:05preglowwhy do you need the added line in the ifs?
00:04:13 Quit [IDC]Dragon ()
00:04:22obolinuxstb: if you get a chance, could you look at 4899 again?
00:04:39herz42the ifs?
00:04:53preglowthe < 0 part
00:05:02preglowwhy is that added now when it worked without them before?
00:05:25herz42I assume it didn't work before
00:05:33godzirraHmm.. rebuiling my ipod database with foopod didnt work
00:05:35godzirraoh well.. ply with it tomorrow.
00:05:41preglowi would have thought the only thing you needed to change was adding 'signed'
00:05:53 Quit OPP ()
00:06:23herz42yes, but when the backlight is off, keypresses will never make it through the 'first keypress enables backlight' filter...
00:06:33herz42at leat I think so.
00:07:51 Join tianjing [0] (
00:08:06preglowherz42: but i haven't got time to test it now, someone else needs to
00:08:21herz42no prob. it's in the tracker...
00:08:23 Join [TCK] [0] (
00:11:29miner49eris there an equivilent of GetTickCount in the rockbox API? I need to know how many ms have passed.
00:11:40Bagderthere's a tick counter
00:12:22 Quit lodesi ("Leaving")
00:13:06 Quit tianjing_ (Read error: 110 (Connection timed out))
00:13:10miner49erBadger: That sounds like the ticket. Can I access it from within a plugin?
00:14:19Bagderyes... rb->current_tick
00:14:25Bagderit points to the tick counter
00:15:41Bagdertime to sleeeeep
00:15:46miner49erBadger: Excellent! That's just what I need :-)
00:16:48 Join Doomed [0] (
00:17:06 Nick Doomed is now known as Doomed9 (
00:17:41herz42ahh, I just see that the check for backlight switched off is done at some other place already (but wrong). So I'll try to fix that one to make things more pretty
00:19:18preglowminer49er: keep i mind the tick timer has a resolution of 10ms
00:19:59miner49erpreglow: That's probably okay, it's only for a game so it doesn't need to be mega-accurate.
00:20:46Doomed9question, is there anyway to implement a db function like the one in the iriver 120 firmware?
00:20:58preglowDoomed9: it's already been done
00:21:09preglowcommited a couple of days ago
00:21:32Doomed9is it in a build yet/
00:21:45preglowit's still got some minor glitches, though
00:22:16Doomed9aight i gotta update then
00:23:29 Join imphasing [0] (
00:24:05linuxstbDoomed9: Make sure you get a CVS build - there were a few bugfixes to the tag database today that are not in the last daily build.
00:24:25Doomed9alright, how should i go about updating?...
00:24:33Doomed9just drag the .rockbox folder?
00:25:05linuxstbThe best way is to tell your unzipping program to extract the files directly to your player (and make sure it includes the directories).
00:25:34Doomed9oh, is there anyway to make it exclude wma files?
00:25:57linuxstbIt should automatically exclude any files Rockbox can't play.
00:26:58linuxstbThere is a (brief) wiki page here:
00:27:17 Join ashridah [0] (
00:27:57kclafTracks are not yet sorted by track number within an album.
00:28:10kclafon my ipod, it's reverse order
00:29:05Doomed9hm....are ur id3 tags messed up?
00:30:04 Join arf-arf [0] (
00:31:49 Quit [TCK] (Read error: 110 (Connection timed out))
00:32:41 Quit quobl_ (SendQ exceeded)
00:32:43 Quit Henry43 ("using sirc version 2.211+KSIRC/1.3.12")
00:35:37 Quit ender` (" First things first, but not necessarily in that order.")
00:35:45 Quit mirak (Connection timed out)
00:39:58linuxstbkclaf: I think that's just a co-incidence. On my ipod, some albums are in reverse order, some are just random. But Slasheri has promised to implement track sorting.
00:40:24 Quit webguest35 ("CGI:IRC (Ping timeout)")
00:41:57 Join Paul_The_Nerd [0] (
00:42:07Doomed9did u change the ID3 sorting?
00:42:44 Quit miner49er ("CGI:IRC (Ping timeout)")
00:44:00linuxstbDoomed9: What ID3 sorting?
00:44:13Doomed9its like...V2 then V1
00:44:25Doomed9dont no the difference, im just throwing things out
00:44:44Paul_The_NerdOh, ID3 priority.
00:44:50Paul_The_NerdAlso, what have I walked into the middle of?
00:45:05linuxstbErr, #rockbox?
00:45:33linuxstb:) Just discussing track sorting within an album and the fact it isn't implemented.
00:45:43Paul_The_NerdAh, gotcha
00:45:50linuxstb(in tagcache)
00:45:58arf-arfthe way that the Rio Carbon organised tags was quite clever,
00:46:13arf-arfin that it forced an update when USB was UNplugged
00:46:44arf-arfand compared the db with any new files (somehow, possibly by magic) and then updated the db with only new files
00:47:32Doomed9damn, i guess ill have to delete these wma
00:47:44Doomed9not like i need em, converted to ogg
00:48:37linuxstbarf-arf: I believe that's the plan for tagcache (if it doesn't do it already).
00:49:03ashridahlinuxstb: i thought it already did. a wakeup scan seems to be much shorter...
00:49:04Doomed9it thinks that wma is supported, so it shows :(
00:49:37arf-arfaha, nice one, slasheri
00:49:51linuxstbDoomed9: That should be fixed then IMO.
00:50:26linuxstbAre you sure you're using the latest CVS?
00:50:31arf-arfokay, next request: being able to queue up artists from the tagcache at one time.
00:50:46arf-arfat the moment a long-click only gives the option of Open With
00:50:55Doomed9ogg its funny...not many of my friends no what it is
00:51:13Doomed9arf-arf i think that works...the long click
00:51:27ashridaharf-arf: the ui needs work
00:51:49arf-arfDoomed9, yes, in directory view it does but not (yet) in tagcache
00:52:05 Join JdGordon [0] (
00:52:18arf-arfanother thing; whilst I'm on a roll 8)
00:52:52arf-arfin tagcache, when looking at Genre, artists with "The " are sorted as if the "The " doesn't exist
00:53:03arf-arfBut not in artist
00:53:09Doomed9havent tried that yet
00:53:35Paul_The_Nerdlinuxstb: WMA has been shown as supported erroneously for a while.
00:54:03Doomed9i no that
00:54:11linuxstbPaul_The_Nerd: Yes, they are supported in the file browser (to add to playlists), but they shouldn't play. Looking at the tagcache code, I can't see how they can be added to the database.
00:54:25Doomed9they are
00:54:28 Join webguest46 [0] (
00:54:59 Quit webguest46 (Client Quit)
00:55:09linuxstbDoomed9: How do they appear in the database? Are the artist/title/album tags etc being used?
00:55:22Doomed9gimme one minute
00:55:23Paul_The_Nerdlinuxstb: Ah. Odd. Not knowing how the code works, I can imagine them being just in the "Songs" section, with Filename as their title, if it does that.
00:55:55Doomed9for song it is
00:56:29linuxstbLooking at the code, .wma files should be identifed as AFMT_UNKNOWN and then not added to the database.
00:56:51Doomed9hm, nvm ur right, its not added in
00:57:15Doomed9i was looking at a song thta was put on twice
00:57:37linuxstbNo problem. At least we know that part of the code is working.
00:57:55Doomed9and yes, the sort is reversed
00:58:43*preglow is anxious to see the new wma codec code
00:58:52Doomed9they have one?
00:59:10*linuxstb is even more anxious to see ffmpeg's new AAC codec finished and converted to fixed point
00:59:16ashridahsomeone's been working on it
00:59:20preglowlinuxstb: sweet lord, yes, two good things
00:59:29preglowlinuxstb: we'll probably have only ffmpeg code in the end
00:59:40preglowthey've got a vorbis decoder too...
00:59:52Doomed9isnt there a problem with the ffmpeg wma codec
01:00:03preglowno idea
01:00:14linuxstbYou mean the fact that it's floating point?
01:00:43Doomed9yeah ..still dont no what that means
01:00:46 Join rkostynu [0] (
01:01:12linuxstbSomeone is currently working on fixing that - to make it work in Rockbox.
01:01:27Paul_The_NerdDoomed9: 1.234 11.154 0.01767 <−−- Floating point
01:01:43Doomed9opposed to what
01:01:48JdGordonmorning all
01:02:05arf-arfso fixed point would be a set amount of decimal places??
01:04:26*linuxstb wonders if he should commit a 220x176 DancePuffDuo WPS
01:04:32 Nick Spida_ is now known as Spida (
01:04:52Paul_The_Nerdarf-arf: Internally, it generally means, don't use floating point variables.
01:05:17Doomed9linuxstb whats that
01:05:25preglowarf-arf: fixed point means the decimal point is fixed
01:05:33preglowarf-arf: floating point moves the decimal point as it sees fit
01:05:56preglowarf-arf: the first can be done with ordinary integer math, the second cannot
01:06:15arf-arfaha ->
01:06:23preglowwell, of course it _can_, but it'll be so slow it's not even fun
01:06:58Doomed9well, the tag things needs some updating, like updating after coming out of usb mode
01:07:08JdGordonBg3r: did u end up looking at the text editor?
01:07:10arf-arfI said that!
01:07:14Doomed9i no
01:07:19linuxstbDoomed9: A version of the DancePuffDuo theme (currently designed for the H120) adapted to work on the H300 and iPod Photo.
01:07:28Paul_The_NerdDoomed9: I'm kinda surprised it doesn't, but then I use the Bootloader for USB so I haven't encountered that either way.
01:07:48preglowit's not quite polished yet
01:08:17preglowi'm looking forward to seeing how we'll handle the case of modified metadata
01:08:23preglowit'll probably be dog-slow
01:08:40Paul_The_NerdModified metadata?
01:08:47 Join Daishi [0] (
01:08:53linuxstbpreglow: What are you talking about?
01:09:03preglowPaul_The_Nerd: if you scan the tags, then modify the id3 name of a file, it won't catch it
01:09:17 Join webguest35 [0] (
01:09:26linuxstbAh, tagcache dealing with tags being changed...
01:09:31Paul_The_Nerdpreglow: Ah, I had wondered what would happen if I changed my tags on my PC, then I immediately thought "Wait, I don't do that."
01:09:37preglowweren't we talking about tagcache? :>
01:09:41arf-arfI'm such a twunt. I didn't realise that you have to set the Max files in dir browser to all your songs otherwise the tagcache cuts out 1/2way through...
01:09:59herz42maybe the filedate should be stored for each file -> check only new files
01:10:16linuxstbpreglow: I thought we were talking about fixed/floating point...
01:10:20preglowarf-arf: yes, that's a nice artifact stemming from the tagcache not being too polished yet
01:10:27herz42hmm, maybe there will be problems with RTC-less targets
01:11:04preglowherz42: mwell, you can't modify the data from within rockbox yet
01:11:12preglowherz42: so that might be one way of dealing with it, i guess
01:11:30preglowjust need to store the old file date
01:11:33herz42yes, but tagcache could remember the date of the file when it is scanned
01:11:35preglowbut it'll still be slow, heh
01:11:41***Saving seen data "./dancer.seen"
01:11:49herz42scanning the file tree though
01:12:04preglowthe best thing will probably be having the user initiate the scan manually
01:12:15Paul_The_Nerdpreglow: Really simple checksums for each file, and when you do a Forced Rescan, it compares them, and if they've changed it then updates the tags in the database? So, while changed tags aren't detected on the quick rescan, they can be on a forced one without it taking 6 hours?
01:12:20preglowor it'll use a bunch of battery every time you connect usb just to check if you changed anything
01:12:29herz42lol checksums :)
01:12:29Paul_The_Nerdpreglow: Or file date. Sure, go and be simple.
01:12:35preglowPaul_The_Nerd: checking all the really simple checksum of all your files will take _LOTS_ of time
01:12:40herz42parsing 60GB...
01:12:42preglowPaul_The_Nerd: almsot as much as doing a scan
01:13:24preglowhell, just opening and closing all the files on the drive will take loads of time
01:13:32Paul_The_NerdWell then that tells me there's no simple enough checksum in existence yet.
01:13:39 Quit muesli__ ("ich will Kühe!!!")
01:13:41 Quit Moos ("Glory to Rockbox !!!")
01:13:50Paul_The_NerdAh well, I have bad ideas more often than good ones anyway
01:14:02herz42the problem with checksum is that you have to read the whole file to calculate it
01:14:25linuxstbIf only the tags had a revision number...
01:14:29herz42of course it is more easy than reading two files and comparing byte by byte
01:14:41herz42just depends on the purpose
01:14:56 Join ^BeN^ [0] (
01:14:56Paul_The_Nerdherz42: I suppose that's true, since you can't checksum the tags themselves without reading to the end of the file either.
01:15:00 Nick ^BeN^ is now known as Paprica (
01:15:36preglowlike i said, i believe the most elegant solution (and it really isn't very elegant), is having the user select when to do an incremental tagcache update
01:15:47preglowthere's no way around the fact that it'll be slow and use a lot of battery
01:16:07preglowi wonder how hard/easy it'll be to make a computer based updater
01:16:08ashridahwhat about taking checksums of the directories themselves? isn't there a 'last modified' time in fat32?
01:16:25kclafor like archos does, incremental tagcache update on usb unplug
01:16:44herz42preglow: a pc based updater should be simple
01:16:47preglowkclaf: that's what i'm saying, doing that will be slow as hell
01:17:08preglowherz42: yeah, looks like reusing metadata.c shouldn't be too hard
01:17:16kclafhmm on my gmini400 it was quite fine
01:17:23preglowkclaf: flash
01:17:25Paul_The_Nerdpreglow: I have nothing against a PC based updater, for people who don't want to waste batteries on it
01:17:33kclaf20 GB HDD
01:17:37herz42so add that to the umount sript and it's somehow automatic
01:17:38preglowgmini with hd?
01:17:50preglowPaul_The_Nerd: yup
01:17:58linuxstbkclaf: Could you just drag and drop files, or did you have to use special software to transfer files?
01:18:09kclafdrag and drop
01:18:16kclafarchos is UMS
01:18:40kclafbut it uses some internal tag db as rb does
01:18:55kclafit's called ARCLib
01:19:10preglowbut i need to go to bed
01:19:11Paul_The_Nerdpreglow: I do think there needs to be a "(Re)build TagCache" option that builds the whole cache from scratch, for a forced (emergency) update, as well as the other update that just adds on.j
01:19:31preglowwe haven't got much time to figure it out however we want to do it :)
01:19:35preglownot long until feature freeze
01:19:54preglowoh, all the ironing out of bugs to be done
01:20:22Paul_The_NerdMy *first* thought was that there needs to be an "Add to Tagcache" or "Update in Tagcache" context menu item for adding folders or files, or updating their entries the hard way
01:20:36preglowi can't ever imagine using that
01:20:39Paul_The_NerdBecause I *really* wanted to just add my Music folder to it, and not my Audiobook and Spoken Word folders
01:20:40preglowsounds cumbersome
01:20:47preglowlike that
01:20:54preglowyeah, that'd be nice
01:21:00preglowsome exclusion list, at least
01:21:09preglowbut yeah
01:21:10preglowi need sleep
01:21:11Paul_The_NerdAnd that would allow you, if you update the tags on just one album, to force an update to those album's entries and no others, maybe.
01:21:16dpassen1Anyone else think Playlists should be selectable in the tagcache?
01:21:28Paul_The_Nerddpassen1: Aye, at some point.
01:21:43dpassen1Glad to hear it
01:22:21herz42I would prefer the other way around to start with
01:22:43herz42I want to add whole Albums to the Playlists (as recursive insert)
01:22:46 Part obo
01:23:11herz42It seems atm it's only possible to add single songs
01:23:42Paul_The_Nerdherz42: In tagcache, or in general?
01:24:04kclaflinuxstb : do you still have that url describing ipod's protocol to deal with accessories ? can't find it in ipl's wiki
01:24:12herz42in tagcache
01:24:44 Quit nudel ("At Argon, we're working to keep your money.")
01:28:13 Quit ashridah ("uni (*sigh*)")
01:30:13 Join enzyme [0] (
01:31:40enzymecan g5 ipod rockbox potentially play avi or divx files without converting the files
01:32:31Paul_The_Nerdenzyme: If anyone answered that, it'd probably be entirely a guess
01:33:13midkayit's likely possible..
01:33:19enzymedo you know of any one working on the video part
01:33:22midkaywhen it will happen, or even if, is a different story..
01:33:27 Join RotAtoR [0] (n=e@rockbox/developer/rotator)
01:33:29midkayand nobody's touched video yet.
01:33:59herz42a question regarding playlists: how do I remove all files from the actual one?
01:34:24Paul_The_Nerdenzyme: Rockbox is an Audio Player firmware, so video's a fairly low priority overall, if it has any.
01:34:36Paul_The_Nerdherz42: "the actual one"?
01:35:21Paul_The_Nerdherz42: If you mean the in-RAM active playlist, stop playback, and then insert a file/folder, and it should clear the playlist and create a new one.
01:35:22herz42...that one where you queue songs into
01:35:43Doomed9anywhere to dl a binary ffmpeg?
01:36:12herz42Paul_The_Nerd: thanks - I knew it would be easy :)
01:36:36enzymeis there any way to have the lcd light on at a lower brightness to save battery
01:36:56Doomed9check ur options
01:37:41herz42that depends - dimming requires cpu power which in turn takes power
01:37:41*Paul_The_Nerd wonders how much a lower LCD brightness actually affects battery life.
01:38:01herz42where "cpu power" is speed in MHz
01:38:37midkayPaul_The_Nerd, there isn't dimming in CVS is there?
01:39:32Paul_The_Nerdmidkay: There's "fading" which could presumably be adapted into Dimming, if you really felt the need, right?
01:39:39herz42ipod lcd is about 30mA at 12V, so 360mW
01:39:49midkayPaul_The_Nerd, right - asked about that earlier..
01:40:08herz42Paul_The_Nerd: The problem as I heared is that CPU must be boosted for this
01:40:10midkaydimming is entirely possible; the irivers need a cpu boost to handle a constant dim level..
01:40:18midkayipods can handle it without a boost. that's what i was told.
01:40:26Paul_The_Nerdherz42: The CPU must very definitely be boosted for any LCD that doesn't natively support dimming.
01:40:30herz42ahh, so let's do it for ipods
01:40:38Doomed9iriver supports it right
01:40:48Paul_The_Nerdmidkay: Not all iPods. At least, on the Nano I'm pretty sure it's handled the same way as iRivers.
01:40:48 Quit Febs ("CGI:IRC (EOF)")
01:41:02midkayPaul_The_Nerd, hm, could be.
01:41:06Paul_The_Nerdmidkay: And I think the H300 can't do it at all, just the H100, if I recall
01:42:18Paul_The_NerdDoomed9: Basically, it's faked. You can't really set a brightness, but you can flicker the backlight on the iRiver H100 series so that it LOOKS like it's fading out.
01:42:37Doomed9is that what it does on the native firmware?
01:42:49midkayPaul_The_Nerd, i don't think that's how it works.
01:43:19Paul_The_Nerdmidkay: It's not really how it works, no. But it's a similar idea.
01:43:34Paul_The_NerdAt least, to my understanding
01:43:41 Quit maeck ("CGI:IRC (Ping timeout)")
01:43:47herz42compared to the power the cpu boosting takes, it might make sense to dim. And the backlight is usually not active very long. The drawback that no one knows might be backlight lifetime decreasing
01:44:35enzymeif rockbox is audio format only how come people work on getting gameboy games on there
01:45:16herz42maybe they are bored so much
01:45:37midkayPaul_The_Nerd, i can't vouch for the irivers, but the ipod even at 10 second fadeout has a very smooth fade - definitely doesn't look flickered..
01:45:39Paul_The_Nerdenzyme: Because that's a plugin made by someone who wanted to contribute. The focus of the *core* developers is "What is already there, and audio support"
01:45:56linuxstbenzyme: Rockbox is whatever people want it to be. It just so happens that most developers are interested in audio.
01:45:58Paul_The_Nerdmidkay: Watch very carefully with long fade times, and you can see it on the Nano at least
01:46:02Paul_The_Nerdmidkay: Especially near the end
01:46:11midkayPaul_The_Nerd, yes, you can see "stepping"..
01:46:27midkaywhich to me looks like somehow they're just decreasing the power to it with visible steps.
01:46:36midkayeach step looks fine, it's just that you can see the steps..
01:46:55Paul_The_Nerdmidkay: It's really done with PWM, but it's simpler to explain flickering of the actual LCD than to talk about that.
01:47:20midkayno idea how it works, so never mind. :)
01:48:17 Quit Doomed9 (Read error: 104 (Connection reset by peer))
01:48:34Paul_The_Nerdenzyme: For example, there's also the Doom patch, which will be in CVS, but that was done almost entirely by one person who wanted Doom on his iRiver. You could have video on your 5G if you were determined enough, as well.
01:49:08sharpei'm back!
01:50:23 Quit rkostynu ("Chatzilla 0.9.72 [Firefox]")
01:50:35enzymewell i prolly would but i dont know how to do that kind of stuff lol
01:51:02Paul_The_Nerdenzyme: Well, then this sounds like an opportunity for self improvement! :)
01:53:16 Quit elinenbe (Read error: 110 (Connection timed out))
01:53:36sharpeyay motivation speeches
01:56:03 Quit webguest35 ("CGI:IRC (EOF)")
01:59:15 Part enzyme
02:02:28 Join damaki__ [0] (
02:09:41 Join TCK [0] (
02:12:14 Quit tianjing (Remote closed the connection)
02:18:36 Join Doomed9 [0] (
02:18:37 Quit damaki_ (Read error: 110 (Connection timed out))
02:35:25 Quit Doomed9 ("( :: NoNameScript 4.01 :: )")
02:37:10 Quit herz42 ("Und wech")
02:39:47biffheroI turned rockbox (on my 5G) off by holding down the play button for n seconds. How do I power it up?
02:40:05Paul_The_NerdOr Select
02:40:35biffherohrmm... ok. so that wasn't a deep sleep I put it into, it was an 'off'
02:41:09 Join XavierGr [0] (
02:41:12biffherobut the 'resume on startup' is there...
02:45:23Paul_The_NerdWhy wouldn't the resume on startup be there?
02:46:14biffheroI didn't see it before. I was unaware that I was shutting rockbox off, thought it was only sleeping.
02:47:04Paul_The_NerdRockbox doesn't really have a sleep
02:47:36Paul_The_NerdThere's been talk of implementing one since iPod users are used to it, but last I heard the talks didn't go anywhere
02:47:51*biffhero nods
02:53:36 Quit TCK (Read error: 110 (Connection timed out))
02:55:01 Quit Kohlrabi (Nick collision from services.)
02:55:10 Join Kohlriba [0] (
03:01:25 Part Paul_The_Nerd
03:04:54midkaybiffhero, 'resume on startup' means audio playback
03:08:35 Join Paul_The_Nerd [0] (
03:11:42***Saving seen data "./dancer.seen"
03:12:11 Join maeck [0] (
03:16:41 Join webguest06 [0] (
03:18:30 Quit webguest06 (Client Quit)
03:22:49midkayanyone here with an H100?
03:24:10midkayPaul_The_Nerd, do you know if the display flip works on it?
03:24:23Paul_The_NerdIll check
03:24:31XavierGryes it works
03:24:49Paul_The_NerdUpside Down?
03:24:56Paul_The_NerdIndeed, it does.
03:25:42midkayyeah, that. thanks.
03:26:48 Join linuxstb_ [0] (
03:27:57 Join jd|uni [0] (
03:29:05 Quit linuxstb (Read error: 110 (Connection timed out))
03:30:05 Part Paul_The_Nerd
03:37:38kkurbjunBagder, I have the latest version of the doom code on the tracker. It compiles cleanly for the sim and H300, I think it's ready for cvs unless anyone has any other problems that need to be resolved before inclusion
03:38:03midkaykkurbjun, no ipod yet i assume?
03:38:52kkurbjunmidkay, no, not yet, linuxstb was going to look into it once the code was commited to cvs
03:39:32midkaykkurbjun, gotcha
03:41:08 Join muesli__ [0] (n=muesli_t@
03:49:55jd|unikkurbjun: awesome :D
03:51:25midkayanyone know the rules for ifdeffing in the menus? in addition to just ifdeffing the function and the display entries, do you also go and ifdef the associated functions?
03:51:48 Part biffhero
03:52:08midkayah, i guess they don't really exist anyways..
03:52:49midkayjd|uni, if in the main menu, say i #ifdef CONFIG_SWCODEC around sound_menu..
03:53:13midkaythat disables it for showing up on the hardware codec targets, but afaik the source still compiles for sound_menu on hardware codec targets.
03:53:23midkayi was just wondering if sound_menu should also be ifdeffed out.
03:55:26jd|uniumm.. im a bit confused... and i gtg to a lecture
03:55:40 Quit jd|uni ("CGI:IRC (EOF)")
03:59:40 Quit Kohlriba ("Fast alle Menschen sind Regenwürmer")
04:05:34 Quit muesli__ ("ich will Kühe!!!")
04:09:31 Join actionshrimp [0] (
04:16:32 Join mymomthelush [0] (
04:17:37 Part mymomthelush
04:26:00 Quit midkay (Read error: 104 (Connection reset by peer))
04:26:09 Join midkay_ [0] (n=midkay@
04:29:01 Join TCK [0] (
04:30:41 Join Febs [0] (
04:31:26 Quit Ironhand ("Changing server")
04:36:47 Join Ironhand [0] (
04:42:54 Join yeahx [0] (
04:51:10 Quit dpro (Read error: 110 (Connection timed out))
04:57:14 Quit kkurbjun ("Leaving")
05:00:31 Quit thegeek_ (Read error: 104 (Connection reset by peer))
05:00:41 Join thegeek [0] (
05:01:28 Join jd|uni [0] (
05:01:40jd|uniwell that lecture was boring as all hell!
05:02:07midkay_jd|uni, haha.
05:02:24jd|uniget your problem sorted?
05:02:40midkay_jd|uni, in a way. i just committed it how i thought it would work. can always change it later..
05:02:44midkay_thanks for asking :)
05:03:22jd|uniclose enough is good enough :D
05:04:21midkay_haha. and the reason jd|uni didn't have commit access became suddently crystal clear.
05:05:32midkay_btw. know anything about the lang files?
05:07:12 Quit Daishi ("Client exiting...")
05:07:23 Join carini [0] (
05:08:43 Join Sinbios [0] (n=Sinbios@
05:09:28jd|uninot really... what in particular?
05:09:35midkay_whoops, my bad.
05:09:52midkay_i changed a number of things and couldn't figure out why the changes weren't taking place with a new rockbox build..
05:09:58 Quit XavierGr (Read error: 110 (Connection timed out))
05:10:05midkay_then i realized i needed english.lang.
05:11:45***Saving seen data "./dancer.seen"
05:12:39jd|unichanged? or added new lang strings?
05:12:57midkay_apparently it was that i needed to change the.. um..
05:13:03midkay_version numbers in binlang and language.h?
05:13:10jd|unidont u need to run configure again after changing it so it redoes lang.h?
05:13:25midkay_didn't help until i changed the version number, so..
05:13:35midkay_but it's at "2" right now, which worries me. :)
05:13:41midkay_you'd think there'd be more changes..
05:14:33jd|uni.. i dunno
05:14:48 Quit ShyK (Read error: 113 (No route to host))
05:18:00midkay_any core devs around?
05:18:16jd|unithem all in bed still i would assume..
05:18:36jd|uniits like 6.20am there..
05:19:20 Quit TCK (Read error: 104 (Connection reset by peer))
05:19:44 Join DBordello [0] (
05:20:59 Part carini
05:23:02sharpegoodnight rockbox people.
05:23:41midkay_night sharpe
05:23:47 Nick midkay_ is now known as midkay (n=midkay@
05:24:54DBordelloIs the interface supposed to feature 'some' color? I just installed rockbox and the interface is all black and whitre on my ipod photo, wonder if i used a wrong file
05:25:14midkayDBordello, you can customize colors in the menu.
05:25:28midkayby default, it's a light blue background, i thought..
05:26:03DBordelloappears to be just the backlight, i may be wrong
05:26:06DBordelloor wihte
05:26:16DBordellobut i am impressed, listening to flac is sick
05:27:12midkayDBordello, rockbox isn't based around the concept of "pretty".. it's intended to be functional. :)
05:27:39DBordellothat is fine with me.
05:28:07DBordelloalthough, there are themes or something right? The default interface leaves a little to be desired
05:28:11DBordellobut if i can play flac, i am happy
05:28:24DBordelloi need to read the docs on how to use it, or just poke around for a while
05:29:07midkayDBordello, there are 'themes', yes, but they won't do too much.
05:29:14midkayyou can choose a custom font and customize your While Playing screen..
05:29:19 Join Doomed9 [0] (
05:29:29midkaybut for the most part it's solid colors. you can also load a backdrop (like a wallpaper), though.
05:29:36midkayno gradient cursors or anything, though..
05:29:38DBordelloah, i just need to play around with it. Overall i am really impressed, very featurful
05:32:53 Quit mikearthur ("Konversation terminated!")
05:34:42jd|uniany1 any good with shell scripting?
05:37:06jd|unifark this.. im going home
05:37:36 Quit jd|uni ("CGI:IRC (EOF)")
05:39:10DBordellosweet, changed the fonts around and the backgorund colors and i am happy
05:45:39DBordellowhen i am browsings WPS's, do i need to do something to get them to redraw? They aren't displaying properly
05:46:17midkayDBordello, what? they constantly redraw..
05:46:31DBordellohmm, they look odd, maybe they aren't for the ipod
05:48:34 Quit Vertigo_t (Read error: 110 (Connection timed out))
05:55:11 Quit yeahx (Read error: 110 (Connection timed out))
05:57:14 Quit Rick ("I… don't need to be here.")
06:00:44DBordelloDoes anybody know what wps files i should use with an ipod photo? Ones for the H300 or the ipod 5g?
06:01:42 Join mymomthelush [0] (
06:01:53 Part mymomthelush
06:03:54 Join Rob2222_ [0] (
06:06:32 Quit Sinbios (Read error: 104 (Connection reset by peer))
06:07:11 Join Rick [0] (
06:11:14JdGordonyay :) home
06:14:21midkaywb JdGordon
06:14:34Doomed9hello all
06:14:49midkayhey Doomed9
06:14:53JdGordon's good to be back :)
06:15:09arf-arfah, jdgordan, the perfect person to ask!
06:15:13midkay \/home sweet ~ !!
06:15:20midkay /home*
06:15:27JdGordon~ sweet ~ ?
06:15:37arf-arfwhat does your recent patch do: "FS #4929 — add playback_control() to plugins "
06:15:51arf-arf(in words of one syllable for a thickie like me)
06:16:04 Join Gibbed [0] (
06:16:08 Quit Rick (Nick collision from services.)
06:16:19 Nick Gibbed is now known as Rick (
06:16:32JdGordonplayback_control() loads up a menu with stuff like next/prev track and vol+/- and suff.. so u can control the playback without having to exit plugins
06:16:43JdGordonso, that patch just adds the call to a few plugins
06:17:28arf-arfThat's exactly what I hoped you'd say. Yay! Now I can play jeweled to my heart's content while shuffle all is on. And still be able to skip tracks!
06:17:43JdGordonyup.. alsmost
06:17:57*arf-arf does little dance
06:18:13Doomed9how do i get that?
06:18:38 Quit midkay ("Leaving")
06:18:43 Join midkay [0] (n=midkay@rockbox/developer/midkay)
06:18:48arf-arfDoomed9, you'll have to patch
06:19:02 Quit Rob2222 (Read error: 113 (No route to host))
06:19:09midkayJdGordon, referring to linux's ~ variable being the home for that user.... :)
06:19:18Doomed9can i get help on that?
06:19:27arf-arfor if you can hold on 10/15 mins, I'll send you the build I'm about to do 8)
06:19:32 Quit maeck ("Leaving")
06:19:40JdGordonmidkay: ye, i know.. i was wondering why the first was /home instead of ~
06:19:50midkaybut.. BUT..
06:23:16 Join maeck [0] (
06:34:43arf-arfJdGordon: any advice for the correct -p option when patching/order of the 2 patches?
06:34:44JdGordonim so pathetic... im doing a bloody survey in the hope i might win an ipoo :p
06:34:52JdGordon-p0 should work
06:35:01JdGordoneither first.. they dont effect eachother
06:35:16arf-arfI get a hunk failed with current cvs
06:35:20DBordellodoes anybody know of anything like foo_pod (to build a itunes database from the directory structure on the ipod) for linux?
06:35:26arf-arfupdating again in case I arsed up!
06:36:58JdGordondamn these tupid phscological surveys.. they ask the same q at least 3 times always changing the logic of the q..
06:37:09JdGordon:'( on page 3/6
06:37:13JdGordondamn i hate being bored
06:37:16 Quit midkay ("Leaving")
06:37:31 Join Shadowarrior13 [0] (
06:37:40arf-arfhow do you copy+paste text from within the debian dev environment?
06:38:15JdGordonur on vmware image?
06:38:29JdGordonwhat text editor u in? or do u mean from host to the dev env?
06:38:50arf-arffrom dev env back to windows (so I can c&p the error to you)
06:38:58arf-arfI'm kind like that
06:39:05 Quit Shadowarrior13 (Client Quit)
06:39:10JdGordonno idea.. im sure vmware has an option to do it.. but dunno
06:39:22 Join Shadowarrior13 [0] (
06:39:44arf-arf"Hunk #1 failed at 18" when using -p0 on plugins.patch
06:40:00DBordelloi believe you need vmware tools installed to copy and paste between enviroments
06:40:52JdGordonarf-arf: it should have made a .rej file of the file that stuffed up.. which file was it?
06:41:29 Join XavierGr [0] (
06:41:43 Quit HuntermicAtWork ("CGI:IRC (EOF)")
06:41:43 Quit c0utta ("CGI:IRC (EOF)")
06:42:19arf-arfbut it's only tiny. I'll see if I can manually do it. Fingers crossed
06:43:09JdGordonshuoldnt be hard to do manually
06:43:31arf-arfnah just had to c+p #include "../lib/playback_control.h" into the file. even *I* can manage that much!
06:43:56JdGordonhmm.. it didnt do it autoamtically??? i must have screwed the patch up then..
06:45:16arf-arfyeah, it looked like that was what should have happened, but having looked at the .rej file, perhaps the patch was one line out?
06:45:35arf-arfSaying 18,25 not 18,26
06:45:55arf-arf(btw, I know no C at all so that was literally a wild guess)
06:45:56JdGordonmaybe.. i chopped up the patch to remove nonesence changes.. might have slipped up.. no matter
06:49:54JdGordonwhat text editor does the vmware image use?
06:51:27arf-arfprob vi
06:52:03 Quit XavierGr (Read error: 104 (Connection reset by peer))
06:52:33arf-arfyep, vi
06:54:42JdGordonhow long should i set the slideshow mode timer to for the jpeg viewer? 5s?
06:54:57arf-arfmaybe an option?
06:55:07JdGordonwell.. ye.. but untill then...
06:55:20arf-arf5s sounds good to me, then
06:55:43 Quit Shadowarrior13 ("( :: NoNameScript 4.01 :: )")
06:57:01arf-arfDoomed9: still around?
06:57:13JdGordonarf-arf: what is vmwares mem usage while its running?
06:57:24arf-arfin Task Manager?
06:58:01arf-arf+ 267MB!
06:58:11JdGordonhaha, that sux :D
06:58:25Doomed9i dont want to do it right now
06:58:41JdGordoncolinux is using ~5mb and ~6mb while compiling!
06:58:41arf-arfvmplayer.exe = 20MB and vmware-vmx.exe is 267MB
06:59:05 Quit maeck ("Leaving")
07:04:12arf-arfJdGordon: can't get the playback menu to work in Bubbles, but it's great in jeweled/chessbox
07:04:28JdGordonwhats the problem with bubbles?
07:04:36JdGordon.. which target?
07:04:43arf-arfrec doesn't bring up a menu (h120)
07:04:52JdGordonodd... it should...
07:05:10lostlogicprobably because of the special way bubbles handles buttons at 1230812308291 places in the code
07:05:18lostlogicI'm not here.
07:05:29arf-arfmove on, nothing to see here
07:05:40 Quit DBordello (Read error: 110 (Connection timed out))
07:05:56*JdGordon checking the code...
07:06:35 Join DBordello [0] (
07:07:04JdGordonare u sure it patches bubles.c correctly? i tested it and it works on h300.. and it uses the same code on both iriver targets
07:07:18arf-arfJDGordon: it works fine in the menu (before you start the game) but not in-game
07:08:03JdGordonAH.. its not setup to work in the game....
07:08:15JdGordonminor oversight by yours-truly
07:08:19arf-arfAnd that would be why then!
07:08:37arf-arfprobably because of the special way bubbles handles buttons at 1230812308291 places in the code?
07:09:01JdGordonback soon... playing taxi driver
07:09:31arf-arfDoomed9, I've compiled a build with JdGordon's playback patch if you want me to send it to you?
07:09:56 Join B4gder [0] (
07:10:35 Quit cannard ("2.0 Build 3515")
07:11:49***Saving seen data "./dancer.seen"
07:15:15 Part Aditya
07:20:19 Quit Doomed9 ("( :: NoNameScript 4.01 :: )")
07:20:36JdGordonHZ*5 is 5 seconds right?
07:20:39 Quit actionshrimp (Read error: 110 (Connection timed out))
07:21:09B4gderyes, if you mean for sleep etc
07:21:33JdGordonye.. button_get_w_tmo() reutrns button_none after the timeout ye?
07:21:49JdGordonwell then something is fucked!
07:22:28 Join nave7693 [0] (
07:23:03lostlogicnot SYS_TIMEOUT?
07:23:16lostlogicoh, sys_timeout is onqueie wait, button_get_w_tmo changes it
07:24:42 Quit Lynx_ (Read error: 104 (Connection reset by peer))
07:25:08 Join Lynx_ [0] (
07:25:54 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
07:26:36JdGordon*que sunrise music* GOOD morning sweden!
07:26:57*B4gder waves on the behalf of Sweden ;-)
07:29:48 Join Vertigo_t [0] (
07:30:42B4gderit seems the cvs builds really improved with t0mas' fixes last night
07:30:52Bg3rmorning :)
07:33:16 Join tianjing [0] (
07:33:27nave7693hi what players will play the most formats with rockbox besides the ipods
07:33:27JdGordonok, ive changed all 3 rb->button_get() to _w_tmo() and added case BUTTON_NONE: with debug and it still never gets triggered! ?
07:34:10Bg3rJdGordon huh ?
07:34:24B4gdernave7693: all the other swcodec ones: iriver h1x0, h3x0 and iaudio x5
07:34:46JdGordonim trying to get jpeg.rock to work in slideshow mode.. so after a 5 sec timeout it should jum,p tot he next pic.. but the button timeout isnt workign.. and its annoying me
07:34:48B4gder"will" as the x5 has no sound yet
07:35:17Bg3rJdGordon ah, u're working on this ?
07:35:29JdGordonye.. unless some1 beat me to it?
07:35:52Bg3ri _planned_ ...
07:35:57Bg3rbut won't have time soon :)
07:39:53JdGordon:'( oh ffs... i was putting jpeg.rock in rocks/ not viewers/ ..
07:39:59*JdGordon hits self
07:40:54JdGordonit works :)
07:41:16 Join midkay [0] (n=midkay@rockbox/developer/midkay)
07:41:42midkayyay, B4gder's here :)
07:42:27midkayum. i changed a number of strings in the lang files.. what exactly needs to be done to have them compiled? bump the version number in language.h and binlang?
07:42:32Bg3rJdGordon good one ;)
07:42:52Bg3rmidkay huh ?
07:43:15Bg3rwhat does this "i changed a number of strings in the lang files" mean ?
07:43:21midkayBg3r, what don't you get? i changed some stuff, it only seems to show up when i raise the version numbers in language.h and binlang.
07:43:30midkayi changed... some .. strings.
07:43:38Bg3ru haven't added ?
07:43:44Bg3rbut changed existing ones ?
07:44:22Bg3rin the english.lang ?
07:44:29Bg3ror just translations
07:44:47midkayand the eng: part of translations..
07:45:19 Join Gibbed [0] (
07:45:24Bg3rit shouldn't need raising ..
07:45:24 Quit Rick (Nick collision from services.)
07:45:33Bg3rdid you try binlang ?
07:45:35 Nick Gibbed is now known as Rick (
07:45:48midkayBg3r, "try" binlang?
07:45:55nave7693wow, how can the MAS play WAV now...
07:46:11midkayBg3r, "try" binlang?
07:47:26Bg3rrockbox-devel $ tools/binlang apps/lang/english.lang apps/lang/english.lang english.lng
07:47:55*Bg3r greps his hda1, because he is silly ...
07:48:00midkayBg3r, it's not that. it..
07:48:09midkayi mean, compiling it "into" rockbox.
07:48:16 Join linuxstb__ [0] (
07:48:24B4gderthat's done by genlang
07:48:29midkayB4gder, i mean..
07:48:31B4gderthe convertion to a .c file I mean
07:48:43Bg3rB4gder is the man
07:48:45midkayi changed a lot of strings.. and after numerous compiles it didn't show up in rockbox.ipod..
07:48:49B4gderand no, it doesn't care about the version field
07:48:55midkayso i bumped version from 2 to 3 in language.h and binlang.
07:48:58midkaynext compile. it worked.
07:49:08Mikachudid you make clean first?
07:49:08B4gderpossibly it is a makefile flaw
07:49:22midkayMikachu, started a new build directory, make verycleaned, reconfigured...
07:50:10 Quit linuxstb_ (Read error: 110 (Connection timed out))
07:50:23B4gderstill, genlang has no notion of that version
07:50:39midkayB4gder, hmm.
07:50:39 Join Arrogant [0] (
07:50:44B4gderthat's only used for the binary files
07:51:10midkayso just committing all the lang files and nothing else, if i've changed a bunch of stuff, will work fine? nothing else needs to be modified to tell it that strings have changed or anything?
07:51:31B4gderonly the .lang files should be necessary to edit
07:52:15midkayalright, thanks.. hmm.. any way to commit all the lang files at once? :)
07:52:31B4gder"cvs ci" in the lang dir?
07:52:47midkaywon't commit anything but the lang files that are already there, right?
07:52:59B4gderit'll commit all modified files in that dir
07:53:10midkayexcellent, thanks again :)
07:53:13 Quit RotAtoR ("zzzzzzz")
07:53:18B4gderor you can just "cvs ci lang1 lang2 lang3 ..."
07:53:25midkayis it like.. cvs ci -m "comment" ?
07:53:29midkaythat's hard though :)
07:53:34midkayall.. like.. 20 of them.. :)
07:53:52B4gderso what have you modified in the translations?
07:54:03B4gderI doubt you speak 20 languages ;-)
07:54:04midkayjust the eng: fields.
07:54:17midkaydoesn't need to be changed? what?
07:54:35B4gderno, they're supposed to remain what was used when the translation was made
07:54:50B4gdertranslators then use 'uplang' to find out changes
07:54:58B4gderand it'll show the new eng and the old eng
07:55:08midkayalright, that simplifies things. just committing english, i guess :)
07:55:11midkaycool, alright
07:59:10midkaybtw, what does "Repeat Shuffle" mode do? when the playlist reaches the end, it reshuffles and starts over?
07:59:24JdGordonhow well (or not at all) would using button_repeat with another work to toggle a setting?
07:59:46B4gderrepeat with another?
08:00:02Bg3rhaha "grep: memory exhausted"
08:00:17midkayJdGordon, if there's a free key or combo available, it'd probably be nicer.. depends on what you're doing, i guess..
08:00:22JdGordonanother button.. i meant in a combo.. or would it trigger too wuickly?
08:00:33JdGordonmidkay: there isnt.. which causes the problem
08:00:42midkayJdGordon, i guess you have no choice then. :)
08:01:04JdGordonill see how it is.. i guess i can check if the last button was the same then ignore it..
08:04:35 Quit Rick (Read error: 104 (Connection reset by peer))
08:05:06Bg3rmorning, amiconn
08:05:12amiconnmidkay: What did you change in english.lang?
08:05:34*amiconn asks because he still has is english.lang cleanup pending
08:06:53Bg3ramiconn why not just commit it ?
08:07:06JdGordonwhats the a-b button in the sim?
08:07:15amiconnIt's not yet done
08:07:25Bg3rJdGordon see uisimulator/sdl/button.c
08:07:37JdGordon(y) :(
08:08:53Bg3r case SDLK_KP_PERIOD:
08:08:53Bg3r case SDLK_INSERT:
08:08:53Bg3r#ifdef BUTTON_MENU
08:08:53Bg3r new_btn = BUTTON_MENU;
08:08:53Bg3r#elif defined(BUTTON_MODE)
08:08:53***Alert Mode level 1
08:08:53Bg3r new_btn = BUTTON_MODE;
08:09:23 Quit thegeek (Read error: 113 (No route to host))
08:10:42 Join thegeek [0] (
08:11:54JdGordonmm.. goody... rb->slash doesnt work well with jpeg... whats a better way to show that slideshow mode was toggled?
08:12:31 Join Rick [0] (
08:12:38Bg3rJdGordon is this on a h1x0 ?
08:12:45JdGordonh300, but ye
08:13:10Bg3rbecause on h1x0 u should stop the grayscale first, but obviously this isn't the case
08:14:52amiconnJdGordon: When using the grayscale lib (e.g. jpeg viewer on archos, h1x0, hopefully grayscale iPods soon) you *must* not use anything that changes the lcd driectly while the grayscale lib is running
08:15:03 Quit DreamTactix291 ()
08:15:07midkayamiconn, a large number of strings.. mostly just fixing case on some menu items.. clarifying a couple things..
08:15:17amiconnmidkay: Argh!
08:15:21Bg3rJdGordon you can use the get_image(); it will either return u the already decoded image zoom (which should be your case), or will run jpeg_decode...
08:15:23midkayamiconn, big deal?
08:15:27midkaywhat are you up to?
08:15:50amiconnIf you commit your changes, I'll basically have to redo my whole makeover, which is almost complete now
08:15:57midkayamiconn, i simply won't then.. :)
08:16:05 Join RoC_MM [0] (
08:16:30amiconnI have sorted the strings into groups, removed deprecated stuff, combined strings that were just split because we had no splash() back then...
08:16:42Bg3rJdGordon do you get my idea ?
08:16:47midkayamiconn, sounds cool. any ETA?
08:16:48JdGordoni tihnk so
08:17:05Bg3rJdGordon the problem with splash is that it remains on the screen, yep ?
08:17:17Bg3rso this should solve it ;)
08:17:44amiconnmidkay: I am bound to complete it asap, we need it as a precondition for switching to v2 (which means a second cleanup cycle), and want it for 3.0
08:17:58midkayamiconn, sounds good. brb.
08:18:15Bg3rgood amiconn :)
08:18:54***Alert Mode OFF
08:19:08Bg3rbtw, what about changing all switch()es with ifs and using vars/defines to make iriver remotes work correctly in 3.0 ?
08:19:26B4gderthat's not planned for 3.0
08:19:29Bg3ri mean the button switches() in the tree/menus
08:19:45amiconnBg3r: We need a more clever way
08:19:55Bg3ramiconn yes, i know ...
08:20:14amiconnChanging to vars for platforms which don't need it means code bloat - and that we don't want
08:20:35Bg3ramiconn no, my idea was vars for irivers, defines for all other
08:20:43amiconnJdGordon: Using functions that change the lcd directly while the grayscale lib is running means there are two processes sending commands to the lcd, possibly mixing up.
08:21:01amiconnThe grayscale lib controls the lcd from a timer isr
08:21:23amiconnThat means all drawing has to be done using the grayscale lib.
08:21:48JdGordonok, so turn it off, show the splash, turn it back on and redraw the image?
08:22:33amiconnEither that, or don't use splash() but the appropriate *_putsxy() function
08:22:54amiconnAh, no, you can't in jpeg.c
08:23:27amiconn...because it uses the grayscale lib in unbuffered mode which only offers a subset of the drawing functions
08:23:50amiconnSo you have to switch it off before splash() and switch it back on afterwards
08:24:08Bg3ryep, i forgot again about this ...
08:24:38B4gderthat's what we love with rockbox, these little magic gems ;-)
08:24:40JdGordonok, so after turning it back on, what the best way to show the image as it was? (grey or othersie) ?
08:25:15amiconnThe grayscale lib doesn't loose its framebuffer if you switch it off
08:25:15Bg3rB4gder what can i use for searching a string in /dev/hda1, because the grep returns me memory exhausted
08:25:25Bg3reven with -F
08:25:33JdGordonamiconn: and the colour screens?
08:25:42Bg3rJdGordon the color screens don't need it
08:25:49Bg3rjust ifdef GRAYSCALELIB
08:25:54Bg3ror how was it called
08:25:56JdGordonno, but ethey have the same prob with the splash staying
08:26:14Bg3r#ifdef USEGSLIB
08:26:23JdGordon.. dnoe that...
08:26:27B4gderBg3r: you mean on the actual device or just the files on the device ? "find / -type f | xargs grep foo"?
08:26:31Bg3rfor them i suggest u to use my idea
08:26:44Bg3rB4gder searching for deleted text file...
08:27:03B4gderI know LinusN used some fancy tool for that
08:27:08B4gderwhat fs?
08:27:34Bg3rbut i'll probably just forget about it :)
08:27:52amiconnBg3r: Imho the 'extendable action translation' idea was the best one so far. It allows to continue using switch() in the individual event loops, and these switch()es may even get smaller if the action values are close to each other, because then gcc starts using jumptables
08:27:59B4gder has two tools in the bottom
08:28:15Bg3rB4gder yep, 10x :)
08:28:19Bg3ri remember about this
08:28:24B4gderI know Linus used them to recover files from the "Wiki Delete Incidence"
08:28:31Bg3ramiconn hm, yep
08:28:50Bg3ri was talking about a temporary solution for ver 3.0
08:30:09amiconnIn fact the extendable translation idea might work very well for iriver, if we design it to allow more than one translation extender to register
08:30:22*Bg3r goes to read about jump tables in wikipedia
08:30:29 Join XavierGr [0] (
08:30:40Bg3rhaha very big topic on that ...
08:31:09Bg3rXavierGr morning
08:31:25amiconnFor targets with constant button assignments, the core translator would do all the work. For iriver, we would have 3 additional translators. Plugging a remote would register the appropriate translator for the remote type plugged in
08:31:36amiconnPlugins could register their own translator
08:32:30Bg3ramiconn sounds very good
08:32:32amiconnSo on iriver we'd need the ability to register 2 variable translators
08:35:19Bg3rwhat do you think about having different translators for navigation, wps, recording...
08:35:21 Quit Rick (Read error: 104 (Connection reset by peer))
08:36:48amiconnBg3r: The idea is to unify actions as much as possible.
08:37:08Bg3ramiconn but the problem is there are different actions in the wps and in the navigation for example
08:37:22amiconnHaving separate translators would be the same as having individual event loops, just with additional call overhead
08:37:28amiconnYes, and?
08:37:35Bg3rand ...
08:37:41amiconnThe translator would send events like TREE_UP
08:37:44JdGordonamiconn: ok, the splash works great with greyscale now.. how do i reshow the image on colour targets after the splash?
08:37:49Bg3ramiconn yep
08:37:59Bg3rbut TREE_UP is sometimes also WPS_VOL_UP
08:38:04Bg3rbut sometimes not
08:38:07JdGordon.. also, should the jpeg viewer stop the backlight timout?
08:38:09amiconnJdGordon: You need to redraw
08:38:15JdGordonbest way to do it?
08:39:44amiconnBg3r: Hmm, I think in case where a button has different actions in different places, the translator would just send all of them
08:40:02Bg3ramiconn ah, yep
08:40:26Bg3rbut wouldn't it be simpler just to have 2 diff translators ?
08:40:51amiconnThen what's the difference between that hand the current way?
08:41:08amiconnCurrently each app does have its own translator
08:42:08 Quit darkless (Client Quit)
08:42:12amiconnBtw, TREE_UP and WPS_VOL_UP are *not* necessarily the same button
08:42:32amiconnE.g. the gigabeat (and the archos remote!) have separate volume buttons...
08:42:46Bg3ramiconn yep
08:42:58Bg3rthat's why i said this
08:43:20 Join Rick [0] (
08:44:46amiconnThe advantage of having one place to translate button into events forces us to keep the number of different actions low, helping to get the UI more consistent
08:45:09amiconnLike, TREE_UP is (or should be) the same as MENU_UP
08:45:13Bg3ramiconn yep
08:45:19amiconnThese 2 will become just one action
08:45:31Bg3ryes, that is very clear
08:45:40Bg3rJdGordon ?
08:45:48midkayamiconn, there should be like.. MENU_UP and VALUE_UP.. that's the only time i think they'll be different. you want to scroll back to go up in a menu, but to change something, scrolling back should decrease.. or something.
08:46:00JdGordonbest way to redraw the pic after the splash on colour screens?
08:46:13amiconnJdGordon: Just... redraw
08:46:15Bg3rJdGordon just use this get_image and ...
08:46:46Bg3rsee yuv_bitmap_part(...
08:46:53amiconnmidkay: Yes, VALUE_UP can be different from TREE_UP and therefore stays a separate action
08:46:55Bg3r MYLCD_UPDATE();
08:47:18midkayamiconn, right.. but i don't think you need too many others (i'm not quite sure what you've got in mind, though).
08:47:29Slasherii am just going to add open() to check from dircache the file startcluster to make opens *much* faster and even tagcache building way faster :)
08:47:34midkayin fact, that's all as far as i can think of..
08:47:36amiconnOn targets with the standard left/right/up/down buttons they are the same, but e.g. on player and iPod they aren't
08:47:46midkayamiconn, right.
08:47:47JdGordonoh ffs.. i keep forgetting to copy the recompil;ed rock to the archos folder... grrr
08:47:58B4gderJdGordon: make install does that
08:48:14midkayit also takes like.. like.... more than a few seconds!!
08:48:28B4gderat least on cygwin I bet
08:48:36midkaytru dat.
08:48:39JdGordonok, well MYLCD_UPDATE(); doesnt get rid of the splash...
08:49:07amiconnJdGordon: Of course not. You have to reissue yuv_bitmap_part(..) before
08:49:20amiconnsplash() draws to the lcd framebuffer
08:49:24JdGordonwel thats just silly :p
08:49:32amiconnIt has to
08:49:56amiconnIn fact the grayscale targets make it easier for you, because the grayscale lib has its own framebuffer
08:49:58Bg3rJdGordon all lower level lcd funcs just change the framebuffer, but don't send it to the lcd
08:50:23Bg3rso, u draw all u want on the screen, and then do lcd_update
08:51:23 Join nudel [0] (
08:53:48preglowSlasheri: faster tagcache is always welcome. is open really that slow?
08:58:08JdGordonok, i must be just plain stupid or something.. i copied the yuv_bitmap_part() and MYLCD_UPDATE() stuff from the scroll code, so at the very worst it should show get rid of the splash and scroll a bit.. but nothing happens :'(
08:58:43JdGordonand im definatly copying the .rock to the correct folder after make
08:59:17 Join actionshrimp [0] (
09:00:30 Join jbsd [0] (n=admin@
09:01:18 Part jbsd
09:02:30 Quit JdGordon (Read error: 104 (Connection reset by peer))
09:04:41 Join DJ_Dooms_Day [0] (
09:04:43 Join JdGordon [0] (
09:07:54 Join ender` [0] (i=ychat@
09:08:10amiconnWhy do the iPods have no lcd flip? Lack of controller docs?
09:08:59 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
09:10:28Bg3rJdGordon can i see the code ?
09:11:52***Saving seen data "./dancer.seen"
09:12:10Bg3r <= this is oooold, but brutal...
09:12:16JdGordondirect c+p from a bit further up.. so it should work...
09:12:23Slasheripreglow: yes, currently open() is very slow. I will try if i can make it faster with the dircache
09:12:33JdGordonye :) "we" smote them!
09:13:14JdGordonBg3r: they wernt the only ones also
09:13:52Bg3rJdGordon i know ...
09:14:48Bg3rJdGordon see the loop beginning with "/* loop the image prepare and decoding when zoomed */"
09:15:37preglowwe might even want to enable the dircache by default soon
09:15:45preglowbut i believe there are a couple of bugs left that needs fixiing
09:15:57midkayamiconn, something like that.. it should be a hardware function?
09:16:52amiconnWell, almost all lcd controllers have a feature to switch orientation, in order to support flexible mounting of the panel
09:18:07*amiconn still wonders why people want to waste ram on dirache
09:18:26amiconnWell, it really depends on usage
09:18:29Bg3ramiconn because i don't want to wait for the disk spinning up ?
09:18:38preglowhaha, do you think so...
09:18:43Bg3rand because i prefer the disk to not spin when i just browse ...
09:18:45*JdGordon gone
09:18:46midkayamiconn, hmm, makes sense. well, either way - it's not working atm. :)
09:18:58amiconnBg3r: Only for plain browsing. As soon as file read is involved, the disk has to spin up anyway
09:19:18preglowi happily "waste" ram for dircache so i don't have to worry about the disk spinning up and down all the time while i browse
09:19:23midkayamiconn, longer battery life (what's a few hundred kilobytes out of a 32MB buffer)... and especially no pauses when browsing.. insta-access to your files..
09:19:24amiconn...e.g. when browsing with directory voicing set to use directory clips
09:19:29Bg3ri'm with preglow here
09:19:45B4gderI've not used dir cache yet ;-)
09:19:53amiconnmidkay: I expect longer battery life without dircache with my standard usage pattern
09:19:58preglowi used to have to skip in and out of directories to not make it spin down if i noticed i used a long time
09:19:58amiconnMore audio buffer...
09:20:02midkayamiconn, thus the reason it's optional, i guess. :)
09:20:16preglowso for my usage pattern, dircache is almost needed to not annoy the hell out of me
09:20:18Bg3ramiconn 1/29 more buffer
09:20:35amiconnYes? That's >3% ...
09:21:10amiconnI usually go to an album folder and start playing. The next couple of on/off cycles it's just power on and resume
09:21:21amiconnNo browsing involved
09:21:27preglowi usually select an album, start playing
09:21:35preglowthen before it's done, start browsing for another one that fits my mood
09:21:41preglowwhich might take a little while
09:21:42midkayamiconn, the disk needs to spin up anyways.. what's the difference between now or in two seconds without dircache, is my view..
09:22:16amiconnmidkay: The disk has to spin up less often with a larger audio buffer
09:22:31midkayit's kind of a tradeoff i guess - spin up _slightly_ more often during playback and don't spin during browsing.. or have a few more seconds of buffer for playback but need to spin up for browsing..
09:22:44preglowfor my usage it's definitely a net gain
09:22:51preglowno question about it
09:22:58midkayi like it because i tend to not listen to whole albums at once, i guess. i can browse freely to see what i want to hear next without worrying about spinups/hard drive..
09:22:59 Join bobTHC [0] (n=bobTHC@
09:23:03preglowbut anyway, it does of course depend heavily on use
09:23:07bobTHChi folks !
09:23:10preglowbobTHC: hi
09:23:15amiconnThe 'don't spinup during browsing' only holds as long as directory voicing isn't set to using clips. I'm using that in the car
09:23:38preglowamiconn: we really should start looking at porting a speech synth
09:23:39amiconnPlus, by not using dirache I avoid the asociated bugs...
09:23:44preglowthough i guess a database for that would be huge...
09:23:54midkayamiconn, ah, i am indeed looking at that..
09:23:55preglowdircache really doesn't have lots of bugs anymore
09:23:58FebsSlightly off-topic, but does dircache cache all of the files on the player?
09:23:59preglowi even use it while developing
09:24:02preglowFebs: yes
09:24:06preglowFebs: ehh
09:24:09preglowFebs: at least the contents
09:24:16preglowFebs: not the file contents, that is
09:24:18preglowjust the structure
09:24:20*JdGordon back
09:24:21*preglow needs coffee
09:24:35amiconnpreglow: I ran into a dirache bug at devcon withn <1 min after powerup of Linus' H120 (which had dirache enabled)
09:24:37midkayhaha. "i am indeed looking at that" i just said. i meant to say "ah, i'm not using that" or something. i was reading another line while typing, annoying :)
09:25:19Bg3ramiconn what kind of bug ??
09:25:20preglowamiconn: well, tough luck, i've used it for months and months now even when developing and haven't seen any bugs
09:25:24preglowbut there are some left, that i know
09:25:37preglowand i think they can be solved by just not using the cache for accesses until the cache has been fully updated
09:25:39FebsOK. I was just responding to a thread on Mistic River about shuffling the contents of the drive without first building a playlist, and was wondering if dircache could be adapted to, for example, maintain an up-to-date root.m3u playlist.
09:25:40Bg3rsomething with SUB connect ?
09:25:59amiconnMy test plugin created a file. Invisible in the browser after leaving the plugin with dircache on...
09:26:03preglowwouldn't know, i don't use playlists
09:26:06Febs(Not that I find it to be a big deal to generate a playlist, but it seems to be an oft-requested feature.)
09:26:23amiconn...even after entering a subdir and returning
09:26:31preglowamiconn: while it was scanning, then?
09:26:38amiconnDunno, maybe
09:26:52preglowi think it only does stuff like that when scanning, but haven't seen it myself for yonks
09:27:00preglowit did it more often before
09:27:15preglowSlasheri: is the dircache used at all before it's completely updated?
09:27:31JdGordonwooooot... its wortking :D
09:27:42Bg3rJdGordon :) what was the problem ?
09:28:00JdGordongrr... was c+p error or somethgin
09:28:05JdGordonanyway, havta go.. dinner time
09:28:12XavierGrJdGordon I had plans to implement a slideshow but you got me.
09:28:23amiconnpreglow: Thing is that you shouldn't have to care about that, should you?
09:28:34preglowamiconn: of course not, it's a bug
09:28:56preglowi shouldn't have to care about rockbox hanging either, but it still happens
09:29:13Slasheripreglow: no
09:29:49Slasheriand now i will test that open() implementation.. it should be pretty fast now, but used only when reading files
09:30:52 Join petur [0] (
09:30:56Bg3rpetur :)
09:32:52 Quit Febs ("Time to sleep.")
09:33:06amiconndirchache is imho a tradeoff of ram vs. browsing speed. It might not save any battery
09:33:32JdGordonsure it would save batt.. not having to spin up the disk every time the filetree is opened
09:33:33Bg3rit depends on how much do you browse ..
09:33:38JdGordonXavierGr: :D
09:33:42Bg3rpetur was up :)
09:33:51amiconnIt has to scan the disk at boot. That means accessing more dirs than any browsing I am doing by hand later...
09:33:52 Quit midkay ("Leaving")
09:34:40peturBg3r: nothing ;) no free time atm...
09:35:23Slasheripreglow: now the scanning of tagcache was over 50% faster (about 4 minutes, and before that 10 minutes)
09:35:33preglowSlasheri: bodes well
09:36:15preglowamiconn: yeah, but it'll very probably consume less battery than the browsing you would do later
09:37:06preglowand it'll spin up more often, etc, etc
09:37:19XavierGrwell again it depends on the usage.
09:37:32XavierGrI can't live without dircache
09:37:35XavierGrI am spoiled
09:37:49preglowi wouldn't exactly kill myself without it, but rockbox would be more annoying by far, yes
09:37:49XavierGrit is sweet and immediate
09:39:48 Join aliask [0] (
09:40:35JdGordonjpeg slideshow patch is ready.. cept it needs button defines for the slideshow toggle button for all targets except iriver...
09:46:50Bg3rgood :)
09:47:08 Quit safetydan ("Leaving")
09:51:21LinusNJdGordon: really cool
09:51:36LinusNnow all that is left is the button mappings for the other targets
09:51:57JdGordonexcept.. im not sure if its needed in the other button loop or not.. im not 100% sure why it goes to it..
09:52:23LinusNi have a suggestion for the slideshow enable button
09:52:36LinusNhow about NEXT|REPEAT?
09:52:55LinusNhold the next button to enable slideshow mode
09:53:18JdGordonas a general rule for all of them?
09:53:31amiconnNow, lets see whether correctly applying the recording 'blackmagic' fixes the bitshift problem...
09:53:41*amiconn just started a loong test recording
09:54:09LinusNamiconn: you think you can fix the bitshift problem?
09:54:24JdGordonthat could work...
09:54:35LinusNi was under the impression that it was a MAS bug
09:54:42amiconnYes, but...
09:54:57amiconn..there's the 'blackmagic' (as you called it in a comment) that has to be applied to the MAS3587 for recording
09:55:23amiconnThis blackmagic wasn't applied correctly (all the time!) because of the bug in mas_writereg()
09:55:30LinusNi see
09:55:45amiconnMaybe this 'blackmagic' changes some internal priorities...
09:55:57LinusNhehe, i called it black magic because the data sheet doesn't tell what it does
09:57:27XavierGrJdGordon: why use #ifdef JPEG_TOGGLE_SLIDESHOW?
09:57:34 Join markun_ [0] (
09:57:39XavierGrI thought it will be for all targets.
09:57:45amiconnDatasheet page 28,, step 3.
09:57:48JdGordonXavierGr: coz its not defined on all targets yet
09:57:58XavierGrah so you will remove it
09:58:26 Quit markun (
09:58:50amiconnThe 2 writes to register 0xa3 were influenced by the bug. The one to register 0x94 was not
10:06:09 Quit RoC_MM ("Leaving")
10:09:40 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
10:21:06JdGordoncan u do button_menu|button_select|button_rel ?
10:24:10 Quit quobl__ (Remote closed the connection)
10:25:02Bg3rJdGordon hmm yep, but this will react when one of the 2 buttons is released
10:25:09Bg3rw8 to look at button.c :)
10:25:44Bg3rJdGordon this is possible only on h300 ...
10:26:17JdGordonok, im gonna need to do button defines for all the targets..
10:26:41Bg3rJdGordon if u're in doubt about such combinations
10:26:51JdGordonok, ta
10:27:05Bg3rlook at firmware/drivers/button.c:button_read()
10:28:18Bg3rif the btn |= something, than u can use something in combination with other buttons
10:30:15Bg3rthis is of course target specific
10:30:34B4gderbobTHC: indeed an interesting read
10:34:08 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
10:37:37 Join midkay [0] (n=midkay@rockbox/developer/midkay)
10:38:13bobTHCpoor centos project managers
10:38:29 Join Zagor [0] (n=bjst@rockbox/developer/Zagor)
10:38:32JdGordonpoor powertripping city official :)))
10:38:52bobTHC22 y of experience in IT :
10:39:20peturkind of confirms that some/most politicians are idiots ;)
10:39:23bobTHCand dont understand apache default page, whaoo
10:42:50 Quit safetydan ("Leaving")
10:43:11 Quit Mikachu (Read error: 104 (Connection reset by peer))
10:44:13bobTHCor perhaps IT is the acronym for Idiot Talks ;)
10:44:25 Quit phaedrus961 (Read error: 110 (Connection timed out))
10:44:29bobTHC22 years of experience :)
10:48:52 Quit tianjing (Remote closed the connection)
10:49:15 Join tempel [0] (n=tempi@unaffiliated/tempi)
10:49:26 Join tianjing [0] (
10:49:28 Join Mikachu [0] (
10:50:06 Quit damaki__ (Remote closed the connection)
10:50:19 Join damaki__ [0] (
10:51:37tempelHowdy. Question for the Rockbox developers: does the rockbin.bin file for iPods have a unique signature in the first 512 bytes? I like to improve the new iPodLoader2 to detect those images automatically. Please reply to me privately or into #ipodlinux-dev, i do not monitor this channels. Thanks.
10:51:50 Quit amiconn (Read error: 110 (Connection timed out))
10:51:50 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
10:52:03tempel(and I've forgotten the name of the msg-server for leaving msgs)
10:52:06 Part tempel ("Leaving")
10:54:29B4gderI find that slightly odd
10:54:46B4gderhe could've stayed to get the answer
10:55:19linuxstb__Yes, or if he had looked in the source for ipodloader2, he would have seen the code checking for the header...
10:55:28 Nick linuxstb__ is now known as linuxstb (
10:56:02 Join austriancoder [0] (
10:59:49austriancoderi need to expand the codec api so that its possible for me to output YUV4MPEG to the main-lcd
11:00:55Bg3raustriancoder then just do it :)
11:01:06B4gderwhat do you need to expand for that?
11:02:36markun_austriancoder: which video codec are you working with?
11:02:52austriancoderi dont know whats allowed to...
11:02:54Bg3rtheora afaik
11:03:03austriancoderyep theora
11:03:14JdGordonwhats theora?'s codec
11:03:28Bg3rJdGordon mpeg4 video codec
11:03:50austriancoderi got libtheora compiling and i have also a ready codec.. i need only a way to put YUV4MPEG data to main-lcd
11:03:51amiconnThe core should not deal with the plethora of video formats, just rgb
11:04:11amiconnI.e., I think codecs should render into the framebuffer
11:04:17Bg3rthis is 3rd guy to try a video codec ...
11:04:19amiconns/codecs/video codecs/
11:04:21markun_austriancoder: does it take less CPU to decode theora than xvid for example?
11:04:47austriancodermakrun_ i dont know.. i havent run such tests
11:05:05austriancoderBg3r: what video codes were tryed before
11:05:17amiconn*if* there are yuv formats used by more than one video codec, the conversion function should be part of the codeclib,
11:05:29amiconnmuch like the pluginlib does for plugins
11:06:01Bg3raustriancoder xvid & mpeg2 ...
11:06:11XavierGrBg3r: Except from marcun who was the previous person that tried?
11:06:23austriancoderamiconn: okay.. i will convert yuv to rgb..
11:06:26Bg3rwasn't it muesli ?
11:06:35XavierGrmuesli? Are you sure?
11:06:50austriancoderBg3r: and what were the results?
11:07:00markun_austriancoder: and
11:07:06XavierGrmirak maybe but certainly not muesli
11:07:47markun_B4gder: could you add a 'video playback' category to the patch tracker?
11:08:34markun_It was mirak
11:09:22Bg3rah, yes, sorry
11:09:44austriancoderso firstly we need a common interface for video playback in the codec lib
11:09:48Bg3raustriancoder iirc 2fps for xvid with natural resolution on h300 (without sound at all)
11:10:04Bg3r(without any asm opts)
11:10:07B4gdermarkun_: done
11:10:23markun_B4gder: that was quick, thanks!
11:10:54XavierGrI guess iriver did a damn fine job with the video player.
11:10:56Bg3raustriancoder why did you decide to focus on theora ?
11:11:04austriancoderBg3r: asm opt is needed
11:11:17Bg3raustriancoder for sure
11:11:24austriancoderBg3r: i have looked for interger based video codecs and found it
11:11:54***Saving seen data "./dancer.seen"
11:12:14aegrayaustriancoder: which one?
11:12:55peturXavierGr: and then they removed it again :D
11:13:05Bg3rpetur what ??
11:13:18peturvideo was removed in 1.30
11:13:23Bg3rreally ?
11:13:26markun_aegray: theora of course
11:13:32B4gderwasn't that only removed from the eu version or something?
11:13:45peturcould be, don't know
11:13:46bobTHCBg3r >2fps with fixed or floated point code ?
11:13:53aegraymarkun_: what's performance like?
11:13:53XavierGrpetur, only the icon was changed not the video function AFAIK
11:13:58austriancoderaegray: i try it with theora
11:14:01 Quit Arrogant ("Leaving")
11:14:22aegraywhat is performance on h300 like compared to an ipod?
11:14:30*aegray only has ipods. lots and lots of ipods...
11:14:43Bg3rhaha my iriver runs the orig fw only for charging
11:14:44markun_aegray: do you want to sell some?
11:14:55aegraymarkun_: what are you looking to buy?
11:15:02peturXavierGr: well it happened after I switched to RB, so I didn't bother to load it
11:15:06markun_b/w g4 for example
11:15:18aegrayi might be willing to sell one of those
11:16:02bobTHCor simply donate one to the project, if u have a lot of ;)
11:16:57amiconnWee, my mini should arrive today, or maybe tomorrow :)
11:17:13Bg3ramiconn u're going for ipod too ?
11:17:16XavierGrpetur so they removed the video completely? I just heard they changed the icon they used.
11:17:50XavierGrsob amiconn went to the dark side too.
11:17:56XavierGror more probably to the white side
11:18:00B4gderamiconn is on all sides ;-)
11:19:07Bg3rPoor self, nobody likes him/her. He/She was attacked 2 times.
11:19:07Bg3rFor example, like this:
11:19:07Bg3r * josh_ smacks self on head
11:19:08Bg3rhimself seems to be unliked too. He/She got beaten 2 times.
11:19:08 Join tempel [0] (n=tempi@unaffiliated/tempi)
11:19:35B4gdertempel: you checked the sources now ;-)
11:19:44tempelyes, and found that it does not work well
11:19:48 Join quobl [0] (n=quobl@tor/session/x-cf5d665658082b96)
11:20:05tempelhere's the problem: i cannot detect the rockbox file by name, ok? thats just not how we want it
11:20:12tempelin ipodloader2, that is
11:20:17linuxstbYes - it's called "rockbox.ipod".
11:20:24JdGordonoh goody, my jpeg.patch didnt work :p
11:20:26tempelnope, not on a fat32 partition
11:20:36XavierGrBg3r: Are these info from stats?
11:20:39B4gderyes it is
11:20:42linuxstbYes - it's just that ipodloader2 doesn't have support for long names.
11:20:52linuxstb(which is a bug that someone should fix)
11:21:07tempelour wiki even suggests that the file may be renamed to whatever the user likes.
11:21:11linuxstbs/bug/missing feature/
11:21:28Bg3rXavierGr yep
11:21:33tempeland the loader currently does not pass the name info to the lower level - it's long forgotten at that point when the user sleects to load it
11:21:40tempelso, let's stop this part, please
11:21:47B4gdertempel: then that is due to how your loader works, not how Rockbox loader works
11:22:05tempeli try to make a point here, would you please listen
11:22:08B4gderour loader assumes a name
11:22:19B4gderyou seem to try to lecture
11:22:45tempelno, i say that loader2 can't look at the name to tell what the img contains. ok?
11:23:05B4gderthen look at the 4-letter string and model code
11:23:12B4gdera bit more painful though
11:23:16tempelmy point is: it would be helpful if the header of the rockbox file would contain a _single_ unique ID that we can use to tell that it's a rb file
11:23:25tempelcurrently, i have to check for a set of multiple ids
11:23:34B4gderI don't see that happening in short term
11:23:48tempeland that's awful because i have to have this check in several places, making it prone to bugs
11:23:57tempelok, that's an answer i can live wiht
11:24:05B4gderyou can fix your loader to use our name ;-)
11:24:27tempelno way, we WANT to allow the user to rename it, as it's the same with the linux kernel and the apple soft
11:24:43B4gderso be it
11:24:44tempelsee, this is an Apple devices :)
11:25:09B4gderwhat does that have to do with anything?
11:25:28tempeluser expectance of easy interface :)
11:25:37JdGordoni need to #define half a sec.. is it better to do it as HZ/2 or HZ>>1 ?
11:25:49MikachuJdGordon: doesn't matter, it's calculated at compile time
11:25:51B4gdertempel: those users hardly use Rockbox in the first place ;-)
11:26:00linuxstbtempel: Rockbox _does_ have a single ID string. You need to set a variable (once, at the start of your loader) to contain the ID string for the model you are running on, and then you can check for that _single_ string.
11:26:01JdGordonMikachu: ok
11:26:07MikachuJdGordon: so /2 is probably better
11:27:06 Join muesli__ [0] (n=muesli_t@
11:28:04 Quit ep0ch (Read error: 110 (Connection timed out))
11:28:32tempellinuxstb - right, but then i can't give the user a good error message. then i can only say: invalid file. if i knew it's a rb file but the id does not match, i can say: "wrong rockbox bversion for this ipod"
11:28:45LinusNtempel: how does the loader know which file too load?
11:28:55tempelhey, does someone know how to avoid this warning in gcc: warning: multi-character character constant
11:29:06*amiconn thinks it's silly to allow renaming of the firmware file
11:29:11LinusNyes, don't use multicharacter constants .-)
11:29:31Mikachuwhat is a multicharacter character constant? like 'hi'?
11:29:47LinusNie you can only use 1 char in the 'x' construct
11:30:11tempelno, i can use 'abcd', it compiles fine
11:30:14Mikachuwhat does 'hi' evaluate to, since it isn't an error?
11:30:28LinusNthe warning is there because it can give endianness problems
11:30:28tempeldepends on your endianness
11:30:34tempelright :)
11:30:53Mikachu -Wno-multichar
11:30:53Mikachu Do not warn if a multicharacter constant ('FOOF') is used. Usually they indi- cate a typo in the user's code, as they have implementation-defined values, and should not be used in portable code.
11:30:57tempel'ab' can result in 0x6162 or 0x6261
11:31:06tempelah, thanks!
11:31:07B4gder-Wno-multichar is the option you're looking for
11:31:12*B4gder is late
11:31:20LinusNand i do suggest not to disable it
11:31:30B4gderI suggest don't use such things in code
11:31:30 Join Kohlrabi [0] (
11:32:00B4gderit is easy enough to use a macro to get the number
11:32:09tempeli decided to use strings instead now
11:32:58tempeland make a central array of the names, so that when one extends the list, both routines will use the new longer list
11:33:07tempelwhat a hack... :)
11:33:14 Quit austriancoder ("CGI:IRC (EOF)")
11:33:53LinusNtempel: how does the loader know which file to load?
11:34:24tempeltwo ways:
11:34:50tempel1. the user creates a config file containing the choices of names for a menu along with file names associated with them.
11:35:25tempel2. if there is no config file, a list of possible paths will be checked - if a file exists there, it gets added to the menu
11:35:29tempelthen the user chooses from the menu
11:35:47 Join austriancoder [0] (
11:35:50tempelso, in the 2nd way, there's only a list of files, no info on their type
11:36:09LinusNif the user renames rockbox.ipod to something else, the automatic firmware update check in rockbox will fail
11:36:15tempelQuestion: is this the current complete list of IDs: "ipco","nano","ipvd","ip3g","ip4g"
11:36:37tempelah, good to know. we should add that to the wiki then
11:36:40tempelwill do
11:36:44B4gder"mini" too
11:37:00tempelah! "mini" gets added...
11:37:13LinusNalso, the user will probably be confused when he/she installs a new version of rockbox
11:37:25LinusNand the loader still loads the renamed file
11:37:35tempelcan you also tell me under which case in loader2's loader.c this "mini" falls?
11:37:40B4gderand he'll run into incompatible plugins and codecs
11:38:02austriancoderLinusN: any progress in the x5 audio driver?
11:38:03LinusNso renaming the file will only cause trouble for the user
11:38:09LinusNaustriancoder: no
11:38:18bobTHC have u already see that ?
11:38:19LinusNi haven't worked on it
11:38:35 Join damaki_ [0] (
11:38:37B4gdertempel: you mentioned a list of strings we use in the rockbox.ipod file, I just mentioned one more
11:38:45linuxstbtempel: The port to the mini isn't working yet, so I haven't submitted a patch to add support to ipodloader2 yet.
11:39:10austriancoderLinusN: i have seen there are only 2 GIOP's for the tlv320.. correct?
11:39:17tempeli see. ok. i will add it to the list, but will then give an error msg in the loading code
11:40:06linuxstbWe are also undecided about whether to have one build for both mini generations, or two separate builds. So everything about the mini could change.
11:40:19B4gderlunch time
11:40:41tempelB4dger - you're in the UK?
11:40:51tempelB4gder, that is.
11:41:59 Join c0utta [0] (
11:42:17LinusNaustriancoder: scl, sda for communication, plus a bunch for the audio signals
11:42:37JdGordonis 1 sec steps good enough for the slideshow timeout?
11:42:43bobTHC"GPAC is officially back at ENST, hence this new numbering jump. The most important new feature is the support of fixed-point in the whole framework"
11:42:56 Join dpro [0] (
11:43:27tempelbtw, the update function can't work on the ipod anyways, or do you support IP-over-Firewire on the iPod?
11:44:20LinusNno, we don't
11:45:40tempeli noticied that there was also no Rockbox info on the ipodlinux wiki. Added one:
11:46:20JdGordonthe whole iPl/ rb thing is too lovey-dovey... it should be a brutal rivelry!!
11:46:22LinusNtempel: don't link to the wiki, link to the main page
11:46:34tempelwhic is?
11:46:35aegrayJdGordon is right. FUCK YOU ALL
11:46:48JdGordon:D thats the spirit
11:46:51*aegray hides
11:47:11tempelok, check again
11:47:31LinusNtempel: muuuuch better :-)
11:48:14tempeland i updated the rockbox info here as well:
11:48:19tempelplease check it
11:48:29tempelwhoops, wait
11:48:45tempelmade a mistake
11:48:59 Quit austriancoder ("CGI:IRC")
11:50:00 Join ashridah [0] (
11:50:20 Join austriancoder [0] (
11:51:42 Quit ze (Read error: 113 (No route to host))
11:53:53austriancoderLinusN: 2wire mode for tlv320 okay?
11:54:09LinusNyes, seems to work for me
11:54:18tempelok, i'm leaving for lunch now too. thanks all!
11:54:24 Quit damaki__ (Read error: 110 (Connection timed out))
11:54:39LinusNtempel: enjoy your meal
11:57:12*JdGordon has jpeg doing slideshow with user configurable timeout :D
11:57:17 Quit DBordello (Read error: 110 (Connection timed out))
11:57:43 Quit austriancoder ("CGI:IRC (EOF)")
11:58:53 Join Moos [0] (
11:59:04MoosHello there
11:59:11 Join austriancoder [0] (
11:59:25 Quit austriancoder (Client Quit)
11:59:34 Join austriancoder [0] (
11:59:51 Quit austriancoder (Client Quit)
11:59:58 Join austriancoder [0] (
11:59:58JdGordonLinusN: im trying to get the longpress next on all targets to switch to slideshow mode.. what do i do where the target has jpeg_next defined with 2 buttons already?
12:00:34LinusNyou mean a combination?
12:00:52LinusNshould still work, shouldn't it?
12:01:09bobTHCyo moos
12:01:49austriancoderLinusN: how can i set SCLK and SDI to high?
12:01:52JdGordon(button_power | button_right | button_rel) for next, (button_power | button_right ) for _pre,(button_power | button_right | button_repeat) for slideshow, that would work?
12:02:27linuxstbIs there a menu in the jpeg plugin yet?
12:02:34Bg3rno, there isn't
12:02:46LinusNaustriancoder: why would you?
12:03:00amiconnaustriancoder: Iiuc you need to write an i2c driver for the tlv
12:03:16amiconnThis should work similar to all other bitbanging i2c driver
12:03:25LinusNJdGordon: seems correct
12:03:59JdGordonjust looks wierd :p
12:04:12LinusNamiconn: unless he managed to get the existing coldfire i2c driver working
12:04:13Bg3rJdGordon isn't this for gigabeat ?
12:04:20Bg3rah, no, it's for iaudio
12:04:38JdGordonBg3r: that 1 was, but all but 2 targets use more than 1 button for _next
12:04:46linuxstbJdGordon: I think a menu would be much easier for users to deal with.
12:05:00linuxstb(and you can also attach your playback menu...)
12:05:17JdGordonsame problem.. still need a button for the menu...
12:05:32JdGordonbut ye.. forgot about the playback menu
12:05:46austriancoderamiconn: the x5 i2c driver should be similar to h1xx-i2c driver
12:07:14austriancoderi mean with the i2c-coldfire
12:07:47linuxstbJdGordon: Can you move another function from a button to the menu?
12:08:27JdGordonno, all the buttons are zoom, scroll and next/prev file
12:08:48linuxstbYou could possibly move "quit" to the menu (making quit the first menu option).
12:09:15*LinusN just made a half baked commit by mistake
12:09:17JdGordoni guess that could work..
12:09:28Mikachuon what platform is this shortage of buttons?
12:09:38linuxstbAlways the ipod...
12:09:40JdGordonpretty much all of em
12:09:45Mikachuis select+play anything in jpeg?
12:10:56Mikachubut i think using select+menu for the menu and having a quit entry would be consistent with other plugins
12:11:06austriancoderLinusN: to get i2c working on x5 we need to extend void i2c_init(void) in i2c-coldfire.c
12:11:08JdGordonmenu it is then..
12:12:00*amiconn expects a complementing 'oops' commit from LinusN
12:12:22JdGordoni can then also move the timeout setting to that and allow up/down scrolling back when slideshow is on
12:12:42 Quit petur ("CGI:IRC 0.5.7 (2005/06/19)")
12:12:46 Part tempel ("Leaving")
12:12:50LinusNhehe, i realized that the %b[x] tag wasn't necessarily the best tag to use, since it is related to battery
12:13:36linuxstbWas there any discussion at devcon about WPSs/themes for the 3.0 release?
12:13:37XavierGrpetur: I will be damned. Well you were right
12:13:40LinusNitoh, i can't find a much better place to have it
12:14:36 Join DBordello [0] (
12:14:38LinusNwell, i can change the description of the %b category :-)
12:16:00LinusNmaybe %b could be more "system" related stuff and not only battery
12:16:54 Join darkless [0] (
12:17:08LinusNany other suggestions?
12:18:18Bg3rwhat's the time char ?
12:19:09linuxstbLinusN: "power-related features" seems appropriate.
12:19:22LinusNlinuxstb: yes...
12:19:27LinusNnot that bad
12:19:41LinusNi'll go with that
12:24:46 Join [IDC]Dragon [0] (
12:25:28 Join webguest13 [0] (
12:25:49*[IDC]Dragon got a new mp3 player today
12:26:01Bg3rwhich is ?
12:26:07ashridahheh. how long till rockbox was on it? :)
12:26:21Bg3r1 minute ?
12:26:28[IDC]Dragonthe cheapest I've seen, for 9 €
12:26:59[IDC]Dragonan SD/MMC stick which can play
12:27:09Bg3rah, so
12:27:19[IDC]Dragondefinitely no rockbox target
12:27:32B4gderah come on, don't let us down!
12:27:36[IDC]Dragonno display, only one button
12:27:46Bg3rhaha 1 button ?
12:27:53[IDC]Dragonplays only mp3 from root, no subdirs
12:28:12[IDC]Dragon(did I mention it was cheap?)
12:28:12Bg3rokay, u can take at least 5-6 buttons from it ... 1 sec, 2 sec... :D
12:28:14Kohlrabiit just measures the time you press teh key?
12:28:18Kohlrabior Morse code?
12:28:33crashdany bets you could get rockboy on there.
12:28:34ashridahBg3r: implement double-click :)
12:28:35linuxstb[IDC]Dragon: Any idea if it can play MP2?
12:28:44[IDC]Dragonit's a rocker switch, which can additionally press down
12:29:07Bg3rso, 3 buttons ?
12:29:07[IDC]Dragonso they get 5 functions from it
12:29:46[IDC]Dragonthis thing has no brand, so I can't look for firmware upgrades ;-)
12:31:05Bg3rwho wants fw upgrades if you have rb ? :P
12:31:16[IDC]Dragonlinuxstb: no idea, it mentiones only mp3
12:31:35[IDC]Dragonrb is a fw upgrade
12:31:48[IDC]Dragonno upgradeable fw, no rb
12:32:07austriancoderwhat are the hex values for gpio1 41 and 42?
12:33:01[IDC]Dragonthis is the one
12:33:21 Join mikearthur [0] (
12:33:22[IDC]Dragonit has an AAA cell sized hunchback
12:33:38[IDC]Dragonon the backside
12:33:41ashridah[IDC]Dragon: it's probably based on a VERY common single chipset used in a majority of those cheap players
12:33:52 Quit dpro ("switching eth interfaces")
12:34:16austriancoderBg3r: can you explain me how you have calculated these values?
12:34:19ashridahthe propsect of getting rockbox onto it would be nearly impossible, since it's probably just mp3-player-on-a-chip
12:34:23[IDC]Dragonashridah: most likely, yes
12:34:27 Join tucoz [0] (
12:34:33LinusNaustriancoder: read the 5250 manual
12:34:34ashridahpity, really :)
12:34:46[IDC]Dragonand it'll have no leftover resources
12:35:01ashridah[IDC]Dragon: even if it did, it's probably mass-manufacturered and unchangable
12:35:05JdGordondoes the recorder have button_f1?
12:35:12tucozLinusN. what will the sleeptimer tag show if the sleeptimer isn't set?
12:35:21austriancoderLinusN: i will have a look into the manual
12:35:23LinusNtucoz: nothing
12:35:42tucozok, just adding that to the manual
12:35:47[IDC]DragonI'll find out when I get closer to a tiny philips screwdriver
12:40:24Lynx_[IDC]Dragon: where did you find this player for 9 euro?
12:40:51 Quit mikearthur (Remote closed the connection)
12:41:09 Join mikearthur [0] (
12:42:31preglowi want more nano wpses :/
12:43:02[IDC]DragonLynx_ : see the link above
12:43:30[IDC]Dragonit even includes headphones, haha
12:43:59preglowi love this
12:44:14aegraywhat are wpses?
12:44:32preglowthe screen that displays while you play
12:44:35aegray<insert more stupid questions here>?
12:45:31preglowa patch correcting typos including "typo's" in the patch name
12:45:33preglowi wonder if it's ironic
12:45:52aegrayhe's posessive
12:46:48aliaskShould the WpsGallery wiki page be split? It's getting pretty long...
12:47:25B4gderI vote for a split
12:48:10*linuxstb votes split
12:48:29aliaskHrm, come to think of it, how do you move an attachment to a new page?
12:48:31*Bg3r joins in voting
12:52:05 Join nls [0] (
12:53:59nlsanother plugin question, will iriverify ship with the release (and be added to the manual)?
12:54:08nlsoh and split!
12:54:23preglownow someone go split it! :]
12:54:35aliaskI'm happy to once I know how to move the attachments.
12:54:59nlsand anyone using the alpine CDC thingy?
12:55:18nlsthe wiki doesn't have a lot of info
12:56:17 Join Jungti1234 [0] (n=jungti12@
12:57:10[IDC]Dragonnls: I wrote tha alpine plugin
12:57:17[IDC]Dragon that
12:57:46nlsI was going to add it to the manual but couldn't find any info on it.
12:58:03nlscould you perhaps update the wiki page
12:58:14[IDC]Dragonis there one?
12:59:17[IDC]Dragonindeed, not much int there
12:59:25nlsnope :-)
12:59:26 Join drspoon [0] (
13:00:18linuxstbaliask: I think that moving all the attachments will be a nightmare - AFAIK, you need to click on "Attach" at the bottom of the page, and then the "manage" link next to the file. then the "move attachment" link at the bottom of that page. About 200 times....
13:01:09preglowcan't we just let the bigger part of the page remain and just rename it?
13:01:09 Quit austriancoder ("CGI:IRC (EOF)")
13:01:11 Join Paul_The_Nerd [0] (
13:01:18 Join Mikaelh [0] (
13:01:40drspoonhas anyone tried wormlet on h300?
13:01:58 Quit Mikachu (Remote closed the connection)
13:02:08drspoonit works fine on ipod video - but on the h300 after a while no more food appears
13:02:36aliasklinuxstb: I value splitting the pages, but not enough to do that.
13:03:06webguest13Script it with curl!
13:03:11 Nick Mikaelh is now known as Mikachu (
13:03:40JdGordonnew jpeg.patch is ready :D
13:03:52linuxstbaliask: Another option would be to leave the attachments where they are, and just change the links.
13:04:07linuxstb(a search and replace...)
13:05:22aliaskI'll start now (probably won't be able to finish tonight though)
13:05:41 Join ender1 [0] (i=ychat@
13:08:10tucozlinuxstb, you didn't get a response regarding the WPS/Themes question? I think that we should ship a larger default font, and a wps adapted to higher resolutions on 3.0 release
13:08:43tucoznot that it answers your question, just an opinion
13:10:33nlstucoz, I think I'm gonna leave the database related plugins and dict, wavplay, midi2wav etc. for now
13:11:06amiconn[IDC]Dragon: I bet it's a sigmatel chip (in your mp3 stick)
13:11:33tucoznls, sounds good. Are you going to rip out hello world etc as well?
13:11:46nlsthink so..
13:11:56***Saving seen data "./dancer.seen"
13:12:23 Join petur [0] (
13:12:29amiconnHelloworld and grayscale can be removed from the manual
13:12:29linuxstbtucoz: I agree. I'm also thinking we should change the wps build script so that WPSs can specify both a minimum and maximum LCD size they work on. So WPSs which look bad on larger LCDs are excluded.
13:12:55tucozand not shipped? That would be good
13:12:59amiconnI also think that PluginIndex should have the developer-only plugins marked
13:15:31 Join webguest31 [0] (
13:15:36[IDC]Dragonnls: better now?
13:15:39 Quit Mikachu (Remote closed the connection)
13:16:10 Join dpro [0] (
13:16:20 Join Mikachu [0] (
13:16:31[IDC]Dragonamiconn: perhaps, yes
13:16:42nls[IDC]Dragon, nice
13:17:03 Quit ender` (Read error: 110 (Connection timed out))
13:17:07*nls starts making a manual page
13:17:17[IDC]Dragonnls: I don't know about TOC and stuff
13:17:32tucozShould the alpine plugin be mentioned in the users manual at all? Isn't a hardware intervention needed for that to work?
13:18:01[IDC]Dragonno hardware besides a proper cable (external)
13:18:10tucozoh, ok. Then all is good
13:18:17[IDC]Dragonbut you don't need to mention it
13:18:31tucozno, of course that should be mentioned.
13:18:33[IDC]Dragonthere's perhaps 3 people using it, including me
13:18:46tucozIf it is shipped with a standard build that is
13:19:05linuxstbMaybe no-one is using it because it's not well documented and advertised?
13:19:35[IDC]Dragonor few people still have M-Bus radios and Archos
13:20:32 Join Febs [0] (
13:20:34 Quit webguest31 ("CGI:IRC (Ping timeout)")
13:20:43[IDC]Dragonactualy, this M-Bus thingy is what I bought my Archos for, firstplace
13:20:52tucozLet's give those 3 people some space in the manual I would say, and maybe some more will start to use it
13:21:05linuxstbGoing back to themes/WPSs, what do people think about a competition to design the default theme/WPS that will ship with Rockbox 3.0?
13:21:39tucozlinuxstb. sounds good. But maybe the default font should be decided here?
13:21:52tucozand design a wps around that
13:21:58amiconnThe builtin font won't be changed for 3.0
13:22:18linuxstbBut the *default* font could be changed?
13:22:25amiconnIt's too much work going through all places that use it and rely on its size
13:22:27Paul_The_Nerdamiconn: It's possible that there should be scaled versions of the default font, so that different target screens display the same number of lines...
13:22:33tucozamiconn, true
13:22:51linuxstbamiconn: Yes, I agree the built-in font is too much work to change for 3.0
13:22:55amiconnIt's a necessary change, but it's post-3.0
13:23:15Paul_The_NerdWell, that's certainly fair.
13:23:19tucozbut a theme could solve this for now
13:23:43 Quit dpro ("Leaving")
13:23:55 Join dpro [0] (
13:24:42Paul_The_NerdI like the idea of a theme contest, with a requirement that the theme have versions for all bitmapped screens. (I'm not really sure what you do for WPS on archos player, other than different variations of alternating text....)
13:25:21 Quit nave7693 (Read error: 110 (Connection timed out))
13:26:35linuxstbOf course, any entry that meets the criteria could be included in the release anyway. But hopefully a competition with a prize - being the default theme - could generate more WPSs which work on all targets.
13:27:58 Quit Moos ("Glory to Rockbox !!!")
13:28:59tucozsounds good. Maybe set up some rules/restrictions for such a contest. Like, should a default wps include bitmaps?
13:30:05Paul_The_NerdI think simplicity is a benefit in a WPS, but bitmaps look good if applied well. That's my personal opinion, at least.
13:30:07amiconnNot with the current loader, imho
13:30:13 Quit actionshrimp ("( :: NoNameScript 4.01 :: )")
13:30:17amiconnLoading a bitmapped wps takes ages...
13:30:59 Quit Kohlrabi ("Fast alle Menschen sind Regenwürmer")
13:30:59JdGordon.. but u only do it once.. so no big deal
13:31:30amiconnNo, it's done on every boot
13:31:38tucozloading a bitmapped wps with lots of bitmaps yes. But not with just a few.
13:31:47amiconn...and booting rockbox on iriver already takes quite long
13:31:59JdGordonis it?? does it?? i dont consider 8sec long
13:32:09JdGordonnot long enouugh to untangle the headphones...
13:32:12Paul_The_NerdMaaaybe you should wait on the WPS contest idea until a single-file or reduced-file WPS concept is in CVS?
13:32:28tucozhehe, I would not say booting the iriver takes long
13:32:42tucozmaybe compared to a flashed recorder
13:33:18amiconnIt's too long for plug iriver to car stereo - boot iriver & start engine - press Play to resume
13:33:30 Quit Zagor ("Client exiting")
13:33:31*amiconn slaps petur
13:33:55webguest13why not boot iriver, plug to car stereo & start engine - press play to resume :)
13:33:58amiconnI always have to wait for the iriver to complete booting
13:34:13amiconnwebguest13: Because plugging after boot causes a pop
13:34:54tucozhehe, I guess that year and a half with iriver fw gave me some patience
13:34:55linuxstbIMHO, the default theme should show what Rockbox is visually capable of. Users who don't like the slower boot time can always change to a text-only theme.
13:35:18tucozlinuxstb: I agree. Let us brag
13:36:01tucozThe current first meeting with rockbox isn't really visually appealing
13:36:10Paul_The_Nerdlinuxstb: I think it'd still be best to wait for at least the .tar'ed wps idea, or something similar, so that rockbox can also show off the speed well.
13:36:36webguest13The hordes of "Rockbox looks like a bag of sh*t" comments agree
13:36:52tucozgot to go. later
13:36:56 Part tucoz
13:36:58drspoonis anyone actively working on tar wpses
13:37:21Paul_The_Nerdwebguest13: My personal concern is with the people who think it's not working because their screen just shows the really basic text. I fielded a lot of that question in the iPod forum 'fore the FAQ link was stickied there.
13:37:41XavierGrhehe the "Bage of shit" comment seems historical, and more and more people get to know with it :)
13:37:42linuxstbThe number of bitmaps in the design could be one of the criteria. So if someone designs both a good-looking and fast-loading WPS, then it would be favourite for the prize.
13:37:56*JdGordon is using ajant blue atm, and rekon that should be the new default (with some minor changes)
13:37:59webguest13single-bitmap wpses seem like a saner approach than tar-wps to me
13:39:01*Paul_The_Nerd is unfortunately a fan of rockbox_default, with the WPS removed.
13:39:18Paul_The_Nerderr peakmeter, not WPS
13:39:59webguest13removing the wps seems a little drastic
13:40:55Paul_The_NerdI do like to see the track name if I look.
13:41:17*webguest13 is not a big fan of alternating lines
13:41:24webguest13make me wait to see the info I want
13:42:18 Part Paul_The_Nerd
13:42:28drspoonhow fast does my server need to be if its worth making it a rockbox build server?
13:42:39preglowwhat is it?
13:42:48B4gderdrspoon: something more than a p3 preferably
13:42:51XavierGrI agree with the others. The WPS must be refined for the new targets. Imagine how the default WPS seems into the H300 screen, or more likely to iPod 5g
13:43:12drspoonan old linux box doing a bit of file serving in my house 133mhz or something phenominally pathetic
13:43:20preglowdrspoon: then far from it
13:43:24preglowit needs to be quite fast
13:43:33B4gderwe hand out complete builds to each server
13:43:35preglowor all it'll do is slow down the build
13:44:09drspoonif i get a new one i'll certainly set it up
13:45:56 Quit drspoon ("Chatzilla 0.9.72 [Firefox]")
13:48:19CassandraQuite fast is relative. I'm running builds on an Althon 1400, which doesn't really count for much these days.
13:48:49JdGordonyippe... my patch crashes the player :D
13:48:52JdGordoni mean :'(
13:49:28*aegray to increase project animosity
13:49:38aegraymissed a few words there
13:50:43*B4gder starts watching his inbox
13:51:19*JdGordon sends lots of pr0n B4gder's way
13:51:41Slasherieh, cvs commit crashed
13:52:11Slasheri|Generating notification message...
13:52:11Slasheri|Traceback (most recent call last): File "/usr/bin/syncmail", line 458, in ?
13:52:12 Join ze [0] (
13:52:36preglowwhat the hell
13:52:38ashridahhmm. the cvs log mailout script
13:52:41B4gderbloody python ;-)
13:53:01preglowpython ://
13:53:08Slasherithat was first time i have ever seen this to happen
13:53:17B4gderline 458:
13:53:22B4gder main()
13:53:28*B4gder giggles
13:53:41Slasherithe last line was socket.error: (111, 'Connection refused')
13:53:44ashridahSlasheri: uh, that's the top of the stack trace
13:53:48*preglow has a big dislike for python
13:53:49ashridahmail server down?
13:54:13muesli__preglow dont u speak parseltongue?
13:54:17JdGordonany1 wanna have a guess why rb->set_option("Toggle Slideshow", &slideshow_enabled, INT,slideshow, 2, NULL); crashes on the target?
13:54:21B4gdernot down now at least
13:54:54preglowmuesli__: apparently not
13:55:07ashridahSlasheri: chances are, the commit probably worked, just the notification log failed
13:55:12linuxstbI've seen those type of syncmail errors on my CVS server before. I think it coincides with mail server problems, but I never investigated very much.
13:55:22Slasheriashridah: yep :)
13:55:24muesli__hum...ask dumbledore for a privat lecture ;)
13:55:26*ashridah has had syncmail do weird things before and the commit survives, since it's run well after the commit's finished
13:56:07 Quit darkless (Client Quit)
13:56:28preglowred builds
13:56:37Jungti1234wow.. E10..
13:56:40JdGordonwould the fact that slideshow_enabled is bool not int matter?
13:56:49B4gderand Slasheri goes for hhiiiiighscore!
13:57:19B4gderbut it built fast!
13:57:22B4gder136 seconds
13:57:23preglowB4gder: we need a highscore table, goddamnit! i will not sit idly by watching my all time highscore go forgotten!
13:58:00B4gderI don't remember what you're talking about!
13:58:04*B4gder ducks
13:58:06*amiconn wonders how Slasheri managed *that*
13:58:06SlasheriBagder: i don't understand why the build failed..
13:58:11Slasheriall files were committed
13:58:21Slasheriso it must be a build system error :)
13:59:00Slasheriin fact the cvs commit jammed for a while before the dircache.h were committed.. so maybe build started without that file :D
13:59:07LinusNmaybe you made two commits?
13:59:18Slasherino, i made only one.. but the commit crashed
14:00:14B4gderwell, if someone makes another commit we'll find out if that was the case
14:00:16preglowB4gder: you got a 'force recompile builds' function to try out now? :)
14:01:01B4gderI guess I should make one for cases like this
14:01:08linuxstbIs tagcache included in the Archos builds?
14:01:23 Quit DJ_Dooms_Day ("Trillian (")
14:01:56Slasherilinuxstb: yes
14:02:33JdGordonok, is it a bug that set_option() will crash on targets if given a (bool*) for the vairable param?
14:05:16amiconnDo you use BOOL for type?
14:05:33JdGordonno, bool
14:05:53amiconnI mean the type parameter of set_option()
14:06:01 Quit JdGordon (Read error: 104 (Connection reset by peer))
14:06:23 Join JdGordon [0] (
14:06:30petur[13:54] <JdGordon> any1 wanna have a guess why rb->set_option("Toggle Slideshow", &slideshow_enabled, INT,slideshow, 2, NULL); crashes on the target?
14:07:07amiconnOf course
14:07:26amiconnif slideshow_enabled is a bool, you have to use BOOL instead of INT
14:07:29 Join austriancoder [0] (
14:07:44aliaskOk, I've split the WpsGallery page, but it's far from finished, the H300, iPod and iAudio pages have all got broken links, but I need sleep. I can either fix it at school tomorrow at some point if no one fixes it before then HINT HINT. :)
14:07:47*JdGordon slaps self again
14:07:48amiconnGuess why there is a type parameter?
14:07:49dprowhat's up with BUTTON_SCROLL_TOUCH ? I'd really like to use touch / no touch without actually moving on the clickwheel
14:07:50JdGordonduh! i knew that :p
14:07:58JdGordonstupid c+p errors
14:08:38aliaskG'night all.
14:08:44 Quit aliask ("sleeeeep")
14:11:12 Nick ender1 is now known as ender` (i=ychat@
14:11:39 Join Genre9mp3 [0] (
14:13:40 Quit JdGordon (Read error: 104 (Connection reset by peer))
14:14:15 Join JdGordon [0] (
14:15:33Genre9mp3Hi there! I was thinking about this, this morning, and I'd like to know your opinion. Would it be possible to use target definitions in the wps code? This would allow us to load different font sizes or even different sized bmps and so on... for every target.
14:15:39JdGordonsome1 commit so i dont keep finsing little things in it to fiddle with :p
14:16:03Genre9mp3So, with a simple .wps file we could have a wps for all targets...
14:16:40pregloweverything is possible
14:16:52preglowbut those wpses would be so incredibly huge and ugly, i don't think it's a good idea
14:17:11preglowmake a scalable wps format instead :>
14:17:34Genre9mp3Would make load times bigger and use more buffer?
14:17:49 Quit JdGordon (Read error: 104 (Connection reset by peer))
14:18:07 Join JdGordon [0] (
14:19:05webguest13Maybe the build-system could run wpses through a preprocessor? :)
14:19:24preglowi actually think that's a semi-neat option
14:19:36preglowcompress a wps into one file while also converting the bitmap format
14:19:45preglowit's not like a wps works across devices now anyway
14:19:48preglowat least not very many of them
14:19:54CassandraGood news for Archos owners
14:20:16Cassandra(Well, sort of.)
14:20:20[IDC]Dragonoh, we need to partition
14:20:25webguest13Hm, why would you want PVR functionality in a laptop?
14:20:41[IDC]DragonSATA :-(
14:21:30Genre9mp3A scalable wps format would be harder to implement though, or am I wrong?
14:21:34nlsoops think I made a commit without message :(
14:21:35[IDC]Dragonthe picture looks like 3 platters
14:21:40CassandraOh. Bugger. Didn't notice that.
14:21:46[IDC]Dragonso it would be 12mm, too
14:22:51[IDC]Dragoneven 160 GB drives are ridiculously expensive today
14:23:10*Cassandra proposes introducing floggings for people who spell the word "extreme" without a leading 'e'.
14:23:20SlasheriLinusN: hmm, do you think we should offer a rombox version also along with the 3.0 release for iriver?
14:23:22 Join B4gd3r [0] (
14:23:33preglowi think that sounds a bit too soon
14:23:39LinusNme too
14:23:50Slasheriok, probably true :)
14:23:57preglowi will be very hard-pressed to make the release as it is
14:23:58CassandraActually let's just flog anyone that uses the word "extreme" to mean "high performance for lots of money" however they spell it.
14:24:09preglowwe don't need more features, especially not big ones requiring lots of testing like that
14:24:43CassandraYeah, it's going to be tricky enough to make that featureset as is.
14:24:44*amiconn wants iriver rombox
14:25:07 Quit B4gder (Nick collision from services.)
14:25:14 Nick B4gd3r is now known as B4gder (
14:25:15Cassandraamiconn, why? It's not as if load time is enormous, or the extra buffer would be that beneficial.
14:26:47Slasheriat least i would like to add the eeprom routine to the bootloader to set a flag if the disk is dirty. only issue is to find a correct location in eeprom so that it wont affect the iriver firmware in any way
14:26:54*LinusN just sent a "gentlemen" mail :-)
14:27:02B4gderyay, there it is!
14:27:17B4gderlet the x5 flood begin
14:27:35Slasheriooh, sound on x5? =)
14:27:51Slasherihehe, cool
14:27:56nlscongrats LinusN
14:28:12amiconn...but certainly the swcodec playback needs to get stable
14:28:21austriancoderLinusN: thanks
14:28:36CassandraWell done, Linus.
14:28:42LinusNi think i deserve a cup of coffee now...
14:28:43Cassandra(Now work on 3.0 features.)
14:28:49Slasheriamiconn: currently it's getting to the other direction for a while..
14:28:50*Cassandra cracks the whip.
14:28:57amiconnCassandra: Faster boot time, plus we could flash ourselves. No need to use the original firmware for that
14:29:03*Cassandra hits herself in the mouth with it.
14:29:32[IDC]Dragon\o/ \o/ \o/ \o/ \o/ \o/ \o/ \o/ \o/ \o/ \o/ \o/ \o/ \o/ \o/
14:29:38 Join Moos [0] (
14:29:41[IDC]Dragon(big crowd)
14:30:09CassandraI am being incompetent at the moment. I managed to slice off half my thumbnail with my fantasic new Swedish cheese slicer.
14:30:39[IDC]Dragonoh, that breakfast inspired you?
14:30:41MoosLinusN: already song in iaudios? :-)
14:30:52preglowCassandra: an everyday problem in scandinavia
14:31:25CassandraPresumably they're less idiotic than me about how they use it.
14:31:46[IDC]Dragonis there no warning sticker attached to it?
14:31:47ashridahi need to migrate out of this god forsaken country
14:32:07ashridahthey're lobbying to introduce ISP-based smut filtering, EVEN THO they know it doesn't work, and screws performance
14:32:14Cassandra[IDC]Dragon, I said I thought they were cool. (And the lovely piece of cat art I saw in the Old City didn't have a price on it, nor was the shop open - possibly a good thing.)
14:32:23JdGordonashridah: it will never pass
14:32:48 Quit Moos (Client Quit)
14:33:06ashridahJdGordon: the problem is, it's getting attention from both camps, and lets not forget that one of the groups pushing it is family first, a right-wing conservative
14:33:07Cassandra[IDC]Dragon, no thank the goddess. Nor do I propose to sue anyone for not warning me that an implement used to cut cheese would be sharp.
14:33:10 Join Moos [0] (
14:33:11[IDC]DragonWarning: chopping off your thumb with it may cause injuries
14:33:21*JdGordon wants to kill family first!
14:33:39preglowi've done my thumb over a couple of times with the bastards
14:33:58JdGordonashridah: na, its not gonna happen.. specially if johnny can finnaly sell of telt$ra for good, then there is no way they would implement it..
14:34:01[IDC]Dragonso grown up swedes have no thumbs?
14:34:05B4gdermaybe you guys should consult an adult before using sharp tools! ;-)
14:34:40ashridahi want to know who told family first that "it could be done without adding more than 10ms to every query"
14:34:56ashridahbecause whoever they are, they're on crack
14:34:57preglowi think it was a norwegian who invented the things
14:35:14CassandraWhereabouts do you live, preglow?
14:35:15 Quit Moos (Client Quit)
14:35:25 Quit Genre9mp3 ()
14:35:25 Join Moos [0] (
14:35:31ashridahbut, nevetheless, this isn't the first time this tosh has been pushed through
14:35:47preglowCassandra: oslo
14:36:04ashridahalston did something similar, and fortunately, that fell flat on its ass (and actually only harmed hosting as a business here, since some .au smut sites moved to the us where it was CHEAPER to host)
14:36:27CassandraCool. I think I like Scandinavia. People there are nice. And cars stop to let pedestrians across the road (which freaked me out a bit).
14:36:41MikachuCassandra: that's a law in sweden
14:36:50preglowCassandra: i actually tend to like british people better than scandinavians...
14:37:04MikachuCassandra: stopping at crossings, not being nice :)
14:37:08preglowat least norwegians
14:37:09B4gderand I like working with machines instead of humans
14:37:12preglowdanes are cooler
14:37:13Cassandrapreglow, obviously you don't know enough British people.
14:37:16preglowwith swedes someewhere inbetween
14:37:25preglowCassandra: i think i've experienced my fair share of them
14:37:31CassandraMikachu, no, not at crossings. At random bits of road.
14:37:33 Quit austriancoder (Read error: 104 (Connection reset by peer))
14:37:38preglowhalf of my family being british and all
14:38:03Cassandrapreglow, hey, I wouldn't want to live anywhere other than Oxford.
14:38:28CassandraBut Sweden kicks Frances and Germany's arse as far as I'm concerned.
14:38:30 Quit Moos (Client Quit)
14:38:49preglowthat it might do, yeah
14:38:52bobTHCnot dure
14:38:54JdGordonit doesnt take much to kick frances ass :D
14:38:55preglowhaven't been to either of those for a long time
14:38:59 Join webguest24 [0] (
14:39:04Cassandra(Despite the fact I speak far more French and German.)
14:39:21preglowwhat, so you've picked up a bit of swedish? :P
14:39:31CassandraJdGordon, you have to catch them in order to kick it. Those Frenchies run fast.
14:39:37preglowenglish people doing scandinavian language always cracks me up
14:39:38Cassandrapreglow, nope, sadly.
14:40:06bobTHCrun fast and are not toopid
14:40:54 Join austriancoder [0] (
14:41:20*[IDC]Dragon spots its "Ladies and gentlemen," these days
14:41:41CassandraAlthough AFAIK I'm the only woman on the list.
14:41:53[IDC]Dragonnah, those gentlemen emails aren't what they used to be
14:42:07Cassandra(but I bitched about it enough, I guess. :) )
14:42:09preglowthere have been several women in here, though
14:42:16[IDC]Dragonfallen for political correctness, this is so 90ies
14:42:38Cassandra[IDC]Dragon, you want I should come over there and kick your arse too?
14:43:05[IDC]Dragondo so
14:43:28CassandraNah, I have people for that sort of thing.
14:43:36CassandraI'll send amiconn - he's nearer.
14:44:00 Quit webguest24 ("CGI:IRC (Ping timeout)")
14:44:51preglowCassandra: plenty of people in dire need of an arse-kicking around these parts
14:45:04XavierGrhey I don't see the message on the dev-ml!
14:45:06*LinusN hides
14:45:10preglowif you think you can do the job convincingly, i'll pay your fare
14:45:35 Quit muesli__ (Read error: 110 (Connection timed out))
14:45:39LinusNXavierGr: i sent it to the main list
14:45:49JdGordonthats a silly place to send it!
14:45:57XavierGrdamn I am not subscribed there. Too many mails!
14:46:05XavierGrAnway congrats Linus!
14:46:11B4gderyou can't get too many mails!
14:46:13ashridahtoo many mails?
14:46:15Cassandrapreglow, somehow I doubt it. Anyway, where's your sense of chivalry? You should be kicking his arse for me.
14:46:16*ashridah blinks
14:46:16[IDC]DragonI'll bet Rocker will comment on it
14:46:17B4gdermroe more moooore!
14:46:22ashridaheven after the signing off flamefest?
14:46:26*XavierGr dances the "Gentlement-we-have-sound-dance"!!!
14:46:59ashridahso, when do we start inserting hidden code that's designed to hypnotise every rockbox user and send them on a rampage against <insert favorite rockbox issue here>
14:47:06CassandraDo you have to do that in public. It's very cruel to the badgers (and bagders).
14:48:06*Bg3r joins to XavierGr's sirtaki
14:48:38preglowCassandra: i don't fancy going to germany, find someone else
14:48:51Bg3rso, this is the first "gentlemen-...sound" on a unit still in production ?
14:49:10Bg3rah, yes ...
14:49:19preglowLinusN: won't that gigantic lut in the lcd driver slow stuff down no end?
14:49:22Bg3rcoldfire unite
14:49:51LinusNpreglow: not really
14:50:21B4gdermemory lane
14:51:03preglowLinusN: btw, i think i'll wait with moving the ram definition stuff to config-xxxx.h until after 3.0
14:51:24XavierGrLinus: If you want to know, the 4 seconds of music was.... Does that mean that it crashed after that :D !!!!
14:51:54LinusNXavierGr: it was a 64k DMA buffer chunk
14:52:11*[IDC]Dragon sees a pattern in those emails
14:53:18CassandraWe had one for Nano and 5g, I believe.
14:54:10 Quit B4gder ("time to say moo")
14:54:52XavierGrwow! I just spotted that they started selling new H120s on a Big Electronic Store in my country.
14:54:57preglowwould replacing the crossfeed be frowned upon after the feature freeze? :)
14:55:23XavierGrstrange this is that it started selling iPods for months and has labeled it as a new arrival!!!
14:56:27XavierGrH120 a new arrival. I know we are back in tecnology and stuff but 3 years is waaaayyy back....
14:57:36Cassandrapreglow, somewhat, yeah, although the swcodec related area is somewhat more fluid than the rest.
14:57:40LinusNpreglow: no, i consider the low gain a bug
14:57:59SpidaXavierGr: where is that?
14:58:33CassandraYou'd probably want to co-ordinate with lostlogic since I think he's planning extensive swcodec changes.
14:58:53preglowLinusN: i've got a new crossfeed ready for inclusion right now, i'll see if a can summon enough courage to do the settings part over the next few days
14:59:05LinusNafaik, lostlogic isn't working on the dsp stuff
14:59:07pregloweither that or make safetydan do it for me :-)
14:59:14*Cassandra considers crossfade a bug, but some weirdos seme to like it.
14:59:33CassandraLinusN, no, but if the underlying interface changes that might confuse things.
14:59:39*JdGordon agrees with Cassandra
15:00:02CassandraHe certainly seemed to expect crossfade etc. to need some fixing as a result of his work.
15:00:21LinusNCassandra: cross*feed*, not cross*fade*
15:00:28CassandraAh, right.
15:00:43preglowcrossfade i don't care about at all
15:00:44XavierGrIn Greece.
15:00:49preglowcrossfeed is at least slightly useful
15:00:52CassandraI just consider that to be an implementation of stereo width and thus somewhat redundant.
15:01:07preglowthat's a pretty bad case of simplification
15:01:28 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
15:01:32preglowstereo width setting doesn't really make hard panned stereo sound too good
15:01:35t0mas* Cassandra considers crossfade a bug, but some weirdos seme to like it. <−− hm? I like the gap cutting in wrong encoded songs...
15:01:46*JdGordon agreed with Cassandra crossFADE, that sux!
15:01:53t0maslike fade out/in for 500 mb
15:02:07preglowi consider the current crossfade settings a bug
15:02:13preglowi don't understand a thing about them
15:02:16CassandraWell, not precisely redundant. Just ought to be part of the stereo width implementation or something.
15:02:41preglowCassandra: yeah, jlo agrees with you, however, stereo width alone is so cheap that i don't want to burden people who only want that with the added complexity of crossfeed
15:03:08preglowit'll be _much_ heavier on the cpu usage, and the new crossfeed will have at least three/four new settings
15:03:46preglowwhich is another point i'm worrying about
15:03:47Cassandrat0mas, I don't think we should be compensating for bugs in other people's software in general. But I'm happy for people to have different opinions and I'm not suggesting the removal of crossfade. Too many people like it. I just only like my music to be mixed when it's done by a DJ.
15:03:54[IDC]DragonRockers' comment to the "Ladies" email is on the list
15:03:59 Quit ashridah ("sleep")
15:04:30preglow"ladies" email
15:04:38Cassandrapreglow, how do the two interact? Badly, I suspect.
15:05:06preglowCassandra: stereo width and crossfeed?
15:05:15LinusNgotta go, cu folx
15:05:17preglowCassandra: in a completely predictable fashion
15:05:18 Part LinusN
15:05:18JdGordonany israelis here?
15:05:22 Join Chamois [0] (
15:05:35preglowboth are linear processes
15:05:39XavierGrmy god they even sell H140s
15:05:46CassandraBut is it what you'd *expect* to happen?
15:05:58CassandraXavierGr, post to the mailing list. I bet they won't for long.
15:07:00XavierGrthe store is in Greece. I don't know if they will deliver outside of Greece.
15:07:13preglowCassandra: yes, sure
15:07:42preglowCassandra: if you do a crossfeed, then some stereo width, you'll get what the crossfeed is supposed to do, then with a stereo image modified as per the stereo width settings
15:08:27CassandraBut stereo width works by mixing left and right channels as well, doesn't it? Or am I completely barking up the wrong tree?
15:08:54preglowyes it does
15:09:09preglowit mixes a bit of left into right, and right into left
15:09:18preglowinverted, sometimes
15:09:29CassandraSo what does crossfeed do that stereo width doesn't - apply delay?
15:09:33preglowand eq
15:09:55preglowwith most parameters configurable
15:10:23CassandraSo crossfeed with delay set to zero would be equivalent to stereo width?
15:11:04preglowif we included a setting that does what stereo width settings does, then yes, but would be considerably slower
15:11:46preglowcrossfeed won't have any way of inverting any of the crossfed signal, and stereo width does that
15:11:58***Saving seen data "./dancer.seen"
15:12:05 Join rockpod [0] (
15:13:41rockpodHello, Is it possible to charge a color ipod with rockbox while it is playing, when I try to charge it now on the adapter it constantly reboots.
15:14:33Cassandrahmm. There's obviously something I'm not getting here. So stereo width gives the impression of wider or nearer speakers by mixing the left and right channels immediately. Crossfeed gives a more "authentic" stereo effect by taking the channel values for delta time ago and mixing that with the current L/R values, giving a sort of echo effect?
15:15:04Cassandrarockpod, that's probably a bug.
15:16:13preglowCassandra: it models the path of the sound from the speakers to the ear, the sound from left speaker to left ear arrives at t = 0, the sound from left speaker to right ear travels a slightly longer path, so is delayed by a factor depending on the diameter of the head. also, the head attenuates the sound a bit on a complex fashion (which we simplify quite a bit), so we apply some filtering to that path
15:17:04preglowcrossfeed really isn't meant to widen or narrow the stereo image, it's meant to give the impression you're listening to a couple of speakers
15:17:17preglowi've never had that effect, but what it does works just like i want it to anyway
15:17:27 Quit tianjing (Read error: 104 (Connection reset by peer))
15:17:29rockpodI read now that holding down the menu button while pluggin it in the adapter should solve the problem?
15:17:35preglowrockpod: it should
15:17:45 Join tianjing [0] (
15:18:00 Quit austriancoder (Read error: 104 (Connection reset by peer))
15:18:03Cassandrapreglow: Right. What did you want it for then?
15:18:17preglowCassandra: making old 60s rock/jazz records bearable
15:18:25preglowi don't particularily want crossfeed, but other people do
15:18:48preglowlistening to a saxophone hard-panned to the left is _really_ tiring
15:19:00preglowcrossfeed fixes that without narrowing the stereo image (which stereo width setting would do)
15:19:06CassandraAh, so mixing stuff with wide stereo seperation like the Beatles, and the Bonzo Dog Band.
15:19:25preglowtry listening to lucy in the sky with diamonds with and without crossfeed once, and you'll understand
15:19:39CassandraI might well do that.
15:19:46preglowit works on the current crossfeed already
15:19:49preglowbut it lowers the volume
15:19:54CassandraOK - now I think I understand why they're seperate functions.
15:20:54preglowgoodie :)
15:22:10preglowthe new crossfeed is done in it's basic c implementation, all that needs doing is making some settings code for it
15:22:18preglowi've got a ton of work on my hands today, but i'll see tomorrow
15:22:22CassandraYou'd get a different sound depending on which of them you applied first, wouldn't you?
15:22:22preglowi also need to do a couple of assembler versions
15:22:38preglowlike i said, they're both linear processes
15:22:47preglowyou can stick them in any order you'd like
15:23:16preglowlemme think on that a bit, btw
15:23:44CassandraOK. I have to go anyway. Take care.
15:24:08rockpodsweet it works now
15:25:29 Join austriancoder [0] (
15:26:44 Join muesli__ [0] (n=muesli_t@
15:31:29 Part austriancoder ("Kopete 0.11 :")
15:32:23preglowno, it does indeed look like the order matters
15:33:46 Quit JdGordon (Read error: 104 (Connection reset by peer))
15:37:37 Quit nls (" HydraIRC -> <- Go on, try it!")
15:38:51 Quit Jungti1234 ()
15:42:18 Join Nico_P [0] (
15:42:37*linuxstb counts 35 flyspray bugs due to be completed by 3.0...
15:44:14linuxstbNico_P: Someone reported on the ipod forums that using your album art patch with tagcache causes problems. Have you tried tagcache?
15:44:36Nico_Pyes and i've had problems with it
15:44:54linuxstbAny idea why?
15:44:57Nico_Pi was going to try to find the patch that caused the problem in my custom build...
15:45:00Nico_Pseems now i know
15:45:53 Join Moos [0] (
15:45:57linuxstbI think the tagcache patch attempts to fill in the tags for tracks using tagcache data. Maybe that part of the code is the issue.
15:46:00Nico_Pon my h300 i get panics whith tagcache
15:47:06Nico_Pi'll have a look
15:47:31*safetydan summons someone with more fixed-point mojo
15:47:50safetydanadding two fixed point numbers so that the value saturates rather than overflows
15:47:54safetydansimplest way?
15:50:45Nico_Pok album art is indeed the cause of the problems... i removed it from my custom build and now it works fine
15:51:16amiconnsafetydan: (asm) Check for overflow, then correct result. (C) No idea...
15:52:10safetydanone of those cases where it's actually easier in assembler... ah well, I'll just make it 64-bit and check if the value is larger than I want
15:56:45*[IDC]Dragon cracked that mp3 stick open
15:57:11[IDC]Dragonthere's a strange chip inside
15:57:19[IDC]Dragon(not sigmatel)
15:57:39[IDC]Dragonfrom Myson Century
15:57:41rockpodIs it normal that the traks are scrambled when using the tagcache-database fuction when one selects an album?
15:57:52petursafetydan: or check if max-val1 < val2, or check if the result is smaller then one of the values...
15:58:10safetydanpreglow, actually that's probably better
15:58:33safetydanthis is for the summing of replaygain plus eq precut
16:01:25linuxstbrockpod: Yes - see here for the current status of tagcache:
16:04:13rockpodthanks, if that is fixed i can compeltely dump the apple firmware
16:04:51*amiconn plans to not install itunes or a replacement for that at all
16:05:20*amiconn hopes that someone with wm* knowledge will help him getting sound on his mini
16:05:21*petur finds himself writing bugs in irc too :(
16:05:37 Join [TCK] [0] (
16:06:06[IDC]DragonI can't find technical info on that MTM809 chip
16:06:31[IDC]Dragononly for the 808, this has an 8051, 8KB, and an mp3 decoder
16:06:45[IDC]Dragon(bad target)
16:06:51petur8051, no kidding?
16:07:04XavierGryou paid 9 euros what you expect?
16:07:20XavierGrexactly :)
16:08:18[IDC]DragonI've never seriously considered this
16:08:26[IDC]Dragonam just curious
16:08:36safetydanpreglow, apparently you have some ui code for me to write :)
16:09:40[IDC]Dragonthat chip has 128 pins!
16:10:08[IDC]Dragonand has use for perhaps 20
16:10:35*HCl has too little time these days :<
16:10:46HCleven if i want to work on rockbox i simply don't get any free time for it
16:10:46HCl :/
16:11:33XavierGrstop whining and start coding :P :D
16:11:42HCl :P
16:11:47HClneed to unwind from college
16:11:58HClspent the entire day coding a sysv7 filesystem implementation already
16:12:00HCl :/
16:12:20bobTHCfor fun ? ;)
16:12:26HClno not really
16:12:30HClbut it's sort of rewarding
16:12:41HClwhen the implementation starts to be able to read directories and files
16:12:48HCland you finally start to be able to read the filesystem
16:12:54[IDC]Dragonfound the chip:
16:13:24[IDC]Dragon64KB mask ROM
16:13:32[IDC]Dragonveeery bad
16:14:01MoosHCl: did you see the initial Tagcache engine of Slasheri?
16:14:17bobTHC[IDC]Dragon > indeeeeeeeeeed
16:14:47[IDC]Dragonand it truely is an 8051
16:14:50HClnot yet
16:14:58[IDC]Dragonhas ATA interface, too
16:15:15MoosHCl: it's on CVS now, you can start to use it in your iriver
16:15:23HCli'm not gonna look at more C code at the moment, need to unwind from college :/
16:15:33HClmmm, i'll question slasheri about it later
16:15:43HCli still need to start on that vexed port too D:
16:15:48Bg3r100-pin LQFP internal ROM version package and 128-pin LQFP external ROM version package.
16:15:55MoosHCl: hehe, busy man
16:16:01[IDC]DragonI just read that, too
16:16:06HClMoos: i work too hard at college
16:16:13HCli need to stop doing that
16:16:20HClbut it's hard to watch my classmates code slowly
16:16:22[IDC]Dragonso there's hope ;-)
16:16:24HClwhile i already know the solution
16:16:25HClso :/
16:16:29MoosHCl: this is for the right cause
16:16:37bobTHCyour study are the most important HCl
16:16:40HClat the moment, i need to relax >.o
16:16:50Moosrelax is needed too
16:16:55HCltell me about it :/
16:17:04HCli almost have a permanent lack of sleep
16:17:09MoosbobTHC can send you weed :-)
16:17:41*bobTHC sends a weedage with his slingshot to Moos
16:17:54MoosHCl: friends, girlfriends... are needed for your mind
16:18:03MoosbobTHC: hehe :)
16:18:04webminddoes anyone know if rockbox will be able to read iTunesDB ? (not that I like iTunesDB, but it would be nice to be able to use your music with different interfaces)
16:18:40*Moos just remenbering HCl come from Betherlands :-P
16:18:48MoosNetherlands even
16:18:58webmindhm? something with .nl ?
16:19:16Mooswebmind: it will
16:19:31*webmind has to go
16:23:00 Join damaki__ [0] (
16:25:39preglowsafetydan: what you is make sure they never overflow
16:26:01preglowsafetydan: s/what you is/what you are supposed to do is/
16:26:08safetydanSo what's the easiest way to do that? A bunch of if statements seems... inelegant.
16:26:10preglowsafetydan: and yeah, that gui code is your for the grabs
16:26:25preglowsafetydan: i just make sure the numbers never get so big that they overflow
16:28:16safetydanI can do gui work any time you're ready
16:29:37preglowhmm, ok
16:29:47preglowi'm thinking of the following parameters:
16:30:11*safetydan gets his notebook out
16:30:16preglow1) crossfeed enable
16:30:26 Quit [TCK] (Read error: 110 (Connection timed out))
16:30:37preglow2) direct gain, db value, 0 dB max, -20 db or something minimum
16:30:47 Quit Chamois (Read error: 113 (No route to host))
16:31:05preglow3) cross gain, db value, -something db max, -somethinf db min
16:31:34preglow4) cross lf attenuation, 0 db max, -something db min
16:31:53preglow5) cross lf cutoff, some frequency range
16:32:08preglowi just need to think of good parameter ranges
16:32:17safetydanblimey that's a lot of parameters
16:32:20preglowand a granularity of 1 db is probably more than enough for this
16:32:25safetydanDoes lf stand for something?
16:32:35preglowlow frequency
16:32:40preglowit should by hf, btw :-)
16:33:10preglowyeah, it is
16:33:15preglowwe might be able to cut it down to four
16:33:16preglowbut yeah
16:33:20preglownot now
16:33:42safetydanWell I'll just extend the cross feed menu, should be easy
16:33:53preglowis there one?
16:34:00preglowi thought there was just one option
16:34:22safetydanYeah it's just an option, but it's easy enough to extend to a menu
16:34:45*webguest13 wonders how people manage to identify chips so easily
16:37:13 Join ac [0] (
16:38:34 Quit damaki_ (Read error: 110 (Connection timed out))
16:39:29webguest13Probably helps to be able to recognise chipmakers' logos I guess
16:39:48safetydanpreglow, all the dB values are negative right?
16:40:36preglowsafetydan: correct
16:40:53 Join arkascha [0] (
16:41:47 Join Spida_ [0] (
16:41:57 Join Kohlrabi [0] (
16:42:03 Part arkascha ("Konversation terminated!")
16:42:31safetydanpreglow, two more questions, 1) sensible default values? 2) upper and lower limits of attenuation cutoff?
16:42:37preglowsafetydan: as a matter of fact, you can make the direct gain 0db to -12db
16:42:44preglowahh sensible default values
16:42:44preglowgod knows
16:42:46preglowi'll need to test
16:43:10 Join nkh^away [0] (
16:43:23preglowsafetydan: private
16:43:35 Quit webguest13 ("CGI:IRC (EOF)")
16:43:44 Join webguest13-2 [0] (
16:44:34webguest13-2Fun, the chip in my el-cheapo mp3 player appears to be a "MP3/JPEG Controller" heh
16:45:04[IDC]Dragonwhich is it?
16:45:17webguest13-2SiliconMotion 338
16:45:29webguest13-2Completely worthless I'm sure, just did it for the exercise
16:46:02webguest13-2Found the manufacturer etc
16:46:46[IDC]Dragonmine must be more el-cheapo, definitely no JPEG
16:47:04webguest13-2They're not interested in telling anything about it, only their newer chips
16:48:04webguest13-2Not that jpeg helps, the player just has a single led
16:48:21[IDC]Dragonshake it in raster scan
16:49:12 Quit quobl (Remote closed the connection)
16:49:32[IDC]Dragonmay result is such effects:
16:49:39[IDC]Dragon(way cooool)
16:52:44 Quit Spida (Connection timed out)
16:55:17amiconn[IDC]Dragon: 8051 is 8bit?
16:56:58peturamiconn: yes
16:57:17[IDC]Dragonamiconn: yes, you don't know that celebrity?
16:57:48[IDC]Dragonpetur: nice, and cheaper
16:58:41amiconn[IDC]Dragon: Not really. Read something about it, but never dealt with embedded CPUs before rockbox
16:59:00peturamiconn: there exist derived 16bit cores too (I think philips and intel have some)
16:59:33[IDC]Dragonamiconn: perhaps you're too young to have been teased with it ;-)
16:59:51[IDC]Dragon(me too, fortunately)
17:00:01 Quit XavierGr ()
17:00:02*petur stopped doing 8051 about 10 years ago
17:00:34amiconnI know about Z80, even coded some Z80 asm in the 80s
17:01:51webguest13-2petur: Haha, that's "Playa-tested!"
17:02:20*amiconn doubst that he is younger than [IDC]Dragon
17:02:24amiconnAt least not much
17:02:39 Nick Spida_ is now known as Spida (
17:02:53[IDC]DragonII was exaggerating
17:03:28[IDC]DragonZ80 might be similar to rotten old 8051
17:04:18[IDC]Dragonbut the 8051 was (and is) widely used for embedded, so it became immortal
17:04:52[IDC]Dragonprobably you can get it with 100 MHz today...
17:05:35[IDC]Dragonand the VHDL to make one for free
17:09:31amiconnWell, I had no reason to deal with embbeded CPUs myself so far (except rockbox)
17:12:00***Saving seen data "./dancer.seen"
17:13:56[IDC]Dragonamiconn: ok, very well, np
17:14:43 Join Chamois [0] (
17:15:32Lynx_[IDC]Dragon: what does the IDC stand for?
17:15:53 Join Massa [0] (
17:15:59linuxstbHas any H300 owner tried the latest Doom patch kkurbjun posted this morning?
17:16:41MassaNo, what's wrong?
17:17:34KyomiDoom's updated?
17:17:52KyomiIt just crashes if you try to play it more then once
17:18:09KyomiAnd if you delete the .dfg file.. it'll work.. but you'll have no sfx
17:18:41linuxstbKyomi: Is that the latest version (from about 12 hours ago), or an earlier one?
17:18:42KyomiI have a slight question here
17:18:49Mikachuheh, it would be cool if doom played one midi channel on the piezo
17:18:54KyomiIt's probably something I did
17:19:02linuxstbIt would be cool if doom worked at all on the ipods...
17:19:16Mikachuthat too
17:19:53 Quit Massa (Client Quit)
17:19:54KyomiWhen I play music now using JoltAmp.. (after installing the H300 experimental build) the progress bar goes beyond the place that was made to display it
17:20:04linuxstbI've just checked the latest patch, and it compiles cleanly for both the H300 and H300 sim, so I think it's ready to commit (for the h300 only). We can then enable builds for other targets as and when they are working.
17:20:05KyomiBefore that never happened... is that a scroll issue?
17:20:08[IDC]DragonLynx_: for instant death commando, our wanna be loser clan
17:20:33KyomiIs the experimental build updated with the doom patch?
17:20:35Lynx_[IDC]Dragon: ok, nice name ;)
17:20:39[IDC]DragonI created that email account for a lan party registration
17:21:01[IDC]Dragonlater used it as a spam account for all kinds of registrations
17:21:12[IDC]Dragonincluding rockbox
17:21:25Nico_Phow come i can't use gdb ? it should worked now that i've compiled it and did "make install", shouldn't it ?
17:21:31[IDC]Dragonthen I started using it...
17:22:16[IDC]Dragonso this is not really a chosen nickname
17:22:37 Quit rockpod ("CGI:IRC (EOF)")
17:22:58webguest13-2Crikey, the partition table on this cheapo mp3 player is messed up. Strange.
17:23:01linuxstbNico_P: What's the problem? What happens when you try to run it?
17:23:04webguest13-2Won't even mount on Linux
17:23:27Nico_Pbash tells me the gdb command doesn't exist
17:23:41Mikachudid make install actually succeed?
17:23:52Mikachuand you have the target path in your $path?
17:24:48linuxstbAre you using Linux or cygwin?
17:24:59Nico_Pi'm on linux
17:25:11Nico_Phave been for a couple days now
17:25:22Nico_PMikachu: i think i have it
17:25:38Nico_Pis it the same as "/home/linus> export PATH=/home/linus/sh1/bin:$PATH" ?
17:25:49Nico_Pi have that one in my path
17:28:26KyomiAnd this is why I hate linux as a main OS :P
17:28:30amiconnNico_P: You're trying to use gdb for the sim?
17:28:38Nico_Pamiconn: yes
17:28:50amiconnThe target crosscompiler path has nothing to do with that
17:29:15amiconnif a plain 'gdb' fails with 'command not found', you probably don't have gdb installed
17:29:34Nico_Pwell "make install" worked
17:29:38amiconnThere should be a gdb package available for your distro
17:29:50Nico_Pbut there's probably something more i have to do...
17:29:50amiconn'make install' has nothing to do with gdb
17:30:07Nico_Pi mean "make install" just after compiling gdb
17:30:18amiconnYou built a cross-gdb?
17:30:22amiconnFor which target?
17:30:30safetydanAnyone know why splitedit.c does this rb->splash(0,...); rb->button_get(true); rb->button_get(true);
17:30:50Nico_Pi just followed the instructions on the crosscompiler wiki page..
17:30:55Mikachusafetydan: probably for eating a press and release
17:31:01amiconnYou can't use cross-gdb with the coldfire targets, unless you open up your device and connect a bdm
17:31:13safetydanMikachu, good point... now why the 0 tick splash?
17:31:29amiconnIf you want to use gdb with the sim, just install native gdb for your linux distro, or cygwin
17:31:37Mikachuif you're going to wait for a keypress, why wait in the splash?
17:31:40godzirraHowdy folks.
17:31:44godzirraAnyone here use foobar2000 and foo_pod?
17:31:54linuxstbNico_P: Which Linux distribution are you using?
17:31:55*safetydan slaps forehead
17:31:59safetydanMikachu, of course
17:32:05Nico_Plinuxstb: kubuntu
17:32:15safetydanNico_P, apt-get install gdb
17:32:21Nico_Pyeah done
17:32:27amiconnErm, splitedit code style is somewhat strange.
17:32:28 Quit Moos ("Glory to Rockbox !!!")
17:32:44godzirraI tried following the instructions to rebuild my itunes database so I could listen in the ipod firmware as well, but it cant find the files (i.e. it just instantly cycles through all the songs in my playlist, like foo_pod pointed to the wrong place.)
17:32:45Nico_Pi just thought i needed a special gdb for rockbox
17:33:40 Quit Febs (Read error: 110 (Connection timed out))
17:33:51amiconnNico_P: You need a cross-gdb if you want to debug code running on the target.
17:33:54linuxstbgodzirra: Lots of people have reported problems with foo_pod. It seems to need updating to the most recent itunes database format.
17:34:11Nico_Pok it works now... sorry for wasting your time and thanks again :)
17:34:15amiconnThat also requires a gdb stub and a method for target<->host communication
17:34:16safetydansplitedit looks like it will be painful to port to swcodec
17:36:42KyomiDuh... *smacks herself*
17:37:06KyomiI hate it when I'm like "Yeah... I'll leave the cable at home.. dont have much time to mess with it today"
17:37:16 Quit ac ("CGI:IRC (EOF)")
17:40:57 Join muesli- [0] (n=muesli_t@
17:43:46 Quit petur ("gone today, here tomorrow")
17:45:53safetydanAnyone with a MAS based unit want to review a patch to fix task 2387 and 2388?
17:45:54 Join Matze [0] (
17:46:39*linuxstb commits Doom
17:46:47 Join quobl [0] (i=id@tor/session/x-dfacdd953309d23c)
17:46:52*Mikachu watches build times go up
17:47:06safetydanKyomi, the hardware decoder based players like the Ondio
17:48:37 Join Moos [0] (
17:48:38safetydanlinuxstb, breaks the H120 sim :)
17:48:48linuxstbMikachu: Only about 2.4MB of source code...
17:48:52safetydanand h120 target too
17:49:03Mikachulinuxstb: just for doom?
17:49:13safetydanplugin.c:418: error: `load_main_backdrop' undeclared here (not in a function)
17:49:13safetydanplugin.c:418: error: initializer element is not constant
17:49:14linuxstbsafetydan: D'oh.... I know why.
17:49:21*linuxstb fix
17:50:01*safetydan looks at the pretty red builds
17:50:42 Quit muesli__ (Read error: 110 (Connection timed out))
17:50:44 Quit Chamois (Read error: 113 (No route to host))
17:50:48linuxstbDid I beat preglow's high score?
17:50:54safetydanonly 450
17:50:57safetydanso no :)
17:51:03linuxstbI deserved more.
17:51:35linuxstbAnd why did the h300 sim give warnings? It was clean when I built it...
17:52:00linuxstbAh, 64-bit build?
17:52:10safetydanlooks like it
17:52:13amiconn_lame_ errors
17:52:19safetydanyeah it's amiconn's system
17:53:43 Join johnybyku [0] (
17:54:45 Quit DBordello (Success)
17:55:09johnybykuhi everyone. is it possible to make the battery monitoring working properly on ipod nano? it shows 100% all the time..
17:55:19amiconnlinuxstb: Why is load_main_backdrop() exposed on the api, btw?
17:56:12linuxstbamiconn: I was just thinking the same thing. I don't think it should be.
17:56:21linuxstbI'm going to remove that.
17:56:52linuxstbWhen I first read that part of the patch, I didn't realise what function it was - I thought it was the lcd function for setting the backdrop.
17:57:34amiconnBagder: I think breaking multiple targets should cause a higher score then just causing errors in one build
17:58:03amiconnI.e., there should be some additional 'base penalty' per build if there is at least one error
17:58:14godzirralinuxstb: sucky.. so there's no way to use your existing structure in the original firmware currently?
17:58:24godzirra(for the ipod stuff I asked about a while ago... sorry =)
18:00:00 Join Febs [0] (
18:01:10linuxstbgodzirra: No. The only working solution to use both Rockbox and the Apple firmware is to use itunes (or similar) to transfer your music, and then use TagCache to view it in Rockbox.
18:01:12 Join DBordello [0] (
18:01:32johnybykuwhat about the battery monitoring?
18:01:43*linuxstb realises it is far quicker to let the build system build 16 targets than to do it himself....
18:03:21amiconnEven 16 targets, 13 bootloaders and 11 sims ...
18:04:32linuxstbamiconn: Does your server always get assigned the H300 sim?
18:05:02amiconnThe build system is concurrent
18:05:17amiconnMy server can build everything except coldfire targets
18:05:41johnybykulinuxstb: could you please answer my question about battery monitoring on ipod nano?
18:06:07linuxstbWhat's the question?
18:06:32johnybykuis it possible to make it working properly?
18:06:53johnybykuright now it shows 100% all the time
18:07:00linuxstbYes, but it needs someone with a Nano to do it. That's not me.
18:07:06 Join kkurbjun [0] (
18:07:20johnybykuthere is a patch submited in the bugs section that should fix it
18:07:38linuxstbHave you tried that patch?
18:08:10kkurbjunamiconn, I exposed the lcd backdrop as it allowed me to use that feature in the doom startup menu's, I was thinking for the plugins that have menu's it would add to them quite a bit.
18:08:23johnybykunot really.. i not good at it to be honest
18:08:33kkurbjunlinuxstb, thanks for commiting that
18:09:25linuxstbThank you for spending 2 months working on it....
18:09:32linuxstb(or however long it has taken)
18:09:43kkurbjun:), yeah it's been a while
18:13:34 Quit johnybyku ("[BX] Mr. Rogers uses BitchX. Won't you be my neighbor?")
18:14:54amiconnlinuxstb: Still red builds...
18:15:20amiconn...and this time my server got the iPod color sim,
18:15:59kkurbjunamiconn, why don't you think that plugins should be able to use the backdrop?
18:16:23linuxstbkkurbjun: They can use backdrops, but not that function.
18:16:38linuxstbThat function is specifically for loading the backdrop for the main part of Rockbox.
18:17:02kkurbjunlinuxstb, what function should be used for plugins?
18:17:05godzirralinuxstb: what is TagCache?
18:17:23*Mikachu hands linuxstb another 2 keyboards
18:17:28godzirralinuxstb: and to do that I assume I'd have to copy all my music off of my ipod and reinstall it all using itunes?
18:17:34linuxstbamiconn: Don't worry, I'm not going anywhere until the build table is clean again :)
18:18:11 Join TCK [0] (
18:18:27kkurbjunlinuxstb, I see what you did, so does lcd_set_backdrop need to be exposed in the plugin api?
18:19:14 Join drspoon [0] (
18:19:36linuxstbkkurbjun: Yes. And maybe the bmp loading function as well if you want to store the bitmap in a file on the disk. But it would be better to link the bmp file into the .rock (assuming there is space)
18:19:46 Join OPP [0] (
18:20:06OPPjust wanted to say nice job on the x5 work
18:20:17drspoonnow that doom's been commited perhaps pacman romdir should be /games/pacman
18:20:35linuxstbgodzirra: (and yes, you probably will have to recopy your music). But test tagcache first.
18:21:02OPPwhere can i download the rockbox for X5 with audio support or has that version not been released yet
18:21:04kkurbjunlinuxstb, there may be space but it would be pretty tight
18:21:06linuxstbdrspoon: I was thinking about that. Also, we should probably distribute prboom.wad as part of Rockbox (it appears to be covered by prboom's GPL)
18:21:36linuxstbMaybe prboom.wad should be somewhere inside .rockbox/ and the user's WADs will remain in /games/doom/
18:21:55 Join lodesi [0] (
18:22:22kkurbjunlinuxstb, you could do just addon wads in that directory also
18:22:38kkurbjunand keep all the main wads in .rockbox/doom
18:24:37godzirraDoes an ipod have weak bass in general?
18:24:40godzirrathats mostly what I was trying to test.
18:24:41MoosOPP: no audio code runing yet, let's LinusN finish first
18:24:55OPPi thought he said he got it
18:25:18 Join Chamois [0] (
18:25:24OPPon iaudiophile it says he got it
18:25:39 Quit tianjing (Read error: 104 (Connection reset by peer))
18:25:41Moosnot finished yet, when that will be ready, you will know it
18:25:41 Join tianjing [0] (
18:26:06OPPahh ok
18:26:33MoosOPP: he made already a big part but need som more works, but hopefully that will come soonish
18:26:41 Join linuxstb_ [0] (
18:26:43OPPya thats cool
18:26:57 Quit linuxstb (Nick collision from services.)
18:27:08 Nick linuxstb_ is now known as linuxstb (
18:27:16OPPis there any reason why i should get the x5 port yet
18:27:43MoosOOP: curiosity I guess
18:27:58OPPi will then :-D
18:28:14Moosyou can already use plugins (games...) for exemple
18:28:34OPPbut you cant switch frimwares so id have to have a backup firmware on there , correct
18:28:40safetydangodzirra, if it has weak bass, you can alwasy improve it with the equalizer
18:29:07MoosOPP: no dual boot like in irivers yet
18:29:21OPPis there USB?
18:31:35OPPdoes iriver have id3 tag databasing yet?
18:31:38OPPthe h320
18:31:54Mooslinuxstn: wee!! green again
18:32:10MoosOPP: yes already here
18:32:22linuxstbMoos: Yes, as long as amiconn's pesky 64-bit server isn't given any sims to build...
18:32:44MoosGood point, forgot 64-bit
18:32:48 Quit drspoon ("Chatzilla 0.9.72 [Firefox]")
18:32:50amiconnlinuxstb: Need a way to test?
18:34:09linuxstbYou're not volunteering to fix the warnings? :)
18:34:24linuxstbI'm guessing they are all casting pointers to int.
18:34:25amiconn*I* didn't cause them... ;)
18:34:32Mooshehe :)
18:35:09*linuxstb wonders why he volunteered for this job.
18:35:39amiconnWell, there's an el cheapo fix for the warnings... but then doom might still not work on the 64bit sim
18:35:58Mooslinuxstb: cause if you didn't do it, who made it?
18:35:59amiconnYou can double-cast... if you know what yer doing...
18:36:30*amiconn wonders why people are casting pointers to int...
18:36:50*amiconn disappears for a while
18:37:08linuxstbMoos: I guess it's originally the prboom authors. Then kkurbjun ported it to Rockbox, and then I committed to CVS.
18:37:14[IDC]Dragonshould be size_t, right?
18:37:36Mooslinuxstb: and thanks for it, no one seems to was interested to commit it
18:38:01 Join b00st4 [0] (
18:38:37linuxstb[IDC]Dragon: amiconn was casting them to long I think.
18:39:14linuxstbSeems 99% of the errors are caused by a single macro being used in lots of places. So a nice easy fix.
18:39:21[IDC]Dragonthat works, but I don't think it's the right way
18:39:36linuxstbI think there's an "intptr_t", but not in Rockbox.
18:40:30[IDC]Dragongotta go, cu
18:40:37 Part [IDC]Dragon
18:46:29*linuxstb commits his 64-bit fixes and hopes amiconn's server gets a simulator to build...
18:47:48 Quit Matze ("Miranda IM! Smaller, Faster, Easier.")
18:48:50godzirrasafetydan: I tried that. I'm not expecting massive ammounts of bass though, but even with the bass turned all the way up in the equalizer, it seems like there's almost none at all.
18:48:55godzirraWasnt sure if it was my ipod or my headphones.
18:52:10 Join haypo [0] (
18:52:23haypolodesi: heya!
18:52:52safetydangodzirra, which band were you using to increase the bass?
18:52:53lodesihaypo: english only channel
18:53:05haypolodesi: it was in english !
18:53:14haypolodesi: so was is your plugin ?
18:53:36haypowhat is the url?
18:53:48haypodid you developed it yourself?
18:54:42haypooops, "was is your plugin" => "what is your plugin exactly?"
18:54:46haypocan i see screenshots?
18:55:24 Join scootscat [0] (
18:55:30linuxstbamiconn: If you get a chance, can you try compiling a sim and let me know if there are any warnings remaining?
18:56:12haypolodesi ?
18:56:26 Join obo [0] (
18:56:58haypolodesi: i was right, alex. belloni (a friend) hacked the rockbox
18:57:44 Join austriancoder [0] (n=austrian@
18:58:08austriancoderx5 has too less buttons for doom.. hmmm
18:59:33linuxstbIt also requires sound - so that may cause problems at the moment.
18:59:49linuxstbBut I'm sure the X5 must have more buttons than the iPods.
19:00:06 Part haypo ("Konversation terminated!")
19:01:39godzirrasafetydan: which band? I was using the bass setting... I didnt know there was a different equalizer?
19:01:58safetydangodzirra, look under sound settings and you should see "Equalizer"
19:02:02austriancoderlinuxstb: joystick with press, play, rec, power with hold
19:02:36godzirrayou would be my new hero if this works.
19:02:37safetydangodzirra, as far as I know the bass and treble setting currently do nothing on iPod targets
19:02:40austriancoderi will reuse a button twice
19:03:09safetydangodzirra, they're supposed to control the hardware amp settings, but I don't know if anyone's figure out how to control them yet
19:03:25 Quit OPP (Read error: 110 (Connection timed out))
19:03:27safetydangodzirra, the equalizer is all in software so works regardless
19:03:32linuxstbsafetydan: They work on the 4g and Nano. The 5g doesn't have bass/treble settings, but does have a 5-band hardware eq.
19:03:49safetydanso godzirra's got a 5g then
19:04:17godzirraYeah, do have a 5g.
19:04:18preglowto the max!
19:04:25godzirraUhh.. the 5g did have bass treble settings.
19:04:28linuxstbSomeone really needs to hook up the hardware eq to Rockbox - it's not used at the moment. But we have a datasheet.
19:04:28godzirraDo they just not work?
19:04:31safetydanpreglow, that max is missing an x
19:04:39godzirraand which band am I supposed to be changing to add bass?
19:04:44linuxstbgodzirra: Yes, they are there, but are not working.
19:04:50godzirraGotcha. :)
19:05:04godzirraHave I mentioned I love rockbox? Even if I do annoy the piss out of you fellas.
19:05:12godzirraBut which band do I need to change to get more bass?
19:05:35safetydangodzirra, try adding 2db to the 60 Hz band
19:05:59 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
19:06:07godzirraThat didnt seem to amke much of a difference.
19:06:14kclaflinuxstb : when i set bass thing on my 5G, i do notice theres an increase in bass
19:06:49godzirraActually it does make a bit of a difference.
19:06:52kclafi use it sometimes because EQ makes track skip
19:06:59linuxstbkclaf: That's your imagination. The functions are not implemented. But I think the volume adjusts to compensate for non-existent extra bass.
19:07:00safetydangodzirra, this might clip but is the quickest way, try 6db on the 60Hz band and say... 4dB on the 120Hz band
19:07:17linuxstbSo there is a change in volume, but not specifically the bass.
19:07:35*kclaf searches his ipod
19:07:50godzirraI only have a 60 and a 200
19:08:18safetydangodzirra, have you used a parametric equalizer before?
19:08:29godzirraNope. I'm equalizer stupid. :)
19:08:38safetydanwell it's a bit complicated
19:09:25amiconnsafetydan: RED...
19:10:36preglowrockbox doesn't have errno, does it?
19:10:43godzirraI think it may be my headphones. I'm not sure. :/
19:10:48safetydanpreglow, yeah it does
19:10:58kclaflinuxstb : you're right, but i could swear it used to increased the bass with some previous build, BUT you know the code so you must be right ;d
19:11:03safetydanbut plugins can't include errno.h now can they?
19:11:07godzirraTHats what I get for buying cheap jvc inner ear instead of the $50 sony ones I guess.. lol
19:11:09austriancoderwhere do i get needed files to play doom?
19:11:15preglowsafetydan: i wouldn't think so, no
19:11:20preglowthey'd need it in the plugin struct
19:11:36 Quit dpro ("wifi scanning")
19:11:37linuxstbsafetydan: I also didn't think errno was in Rockbox. But even if it was, it would need to be *rb->errno (as preglow just said...)
19:11:52amiconnlinuxstb: 17 64bit warnings remaining
19:12:01***Saving seen data "./dancer.seen"
19:12:17linuxstbProgress then. Could you pastebin the warnings?
19:12:24godzirraIs there a way to rate mp3s within rockbox the way ipod does? the one thing I miss from my ipod firmware is being able to generate smart playlists :)
19:12:32amiconnsafetydan: Why do you check errno at all, btw?
19:12:48amiconnMany many other places just check the return code
19:12:50safetydanamiconn, it was for the error message, the code itself doesn't use it
19:12:54amiconnThat should be sufficient
19:13:06linuxstbgodzirra: Not at the moment, but it's been talked about for the future.
19:13:06 Quit scootscat ("CGI:IRC (Ping timeout)")
19:13:26godzirraAwesome. Thats the only thing I really miss.
19:13:28webguest13-2(it's even been implemented once)
19:13:36godzirrawhy was it taken back out?
19:13:50linuxstbIt needs to be re-implemented/adapted for tagcache.
19:13:56webguest13-2It wasn't very stable
19:14:05godzirra(And on a side question, is there a way to generate random playlists, either in rockbox or using another program while the ipod is connected?)
19:14:06webguest13-2And the approach was abandoned
19:14:08linuxstbamiconn: Thanks.
19:14:37godzirraAnd another side question... is there a way to shuffle through out the whole music list, not just the list in one directory?
19:14:50amiconnThe strange chars result from my linux box using utf-8, and ssh'ing to that using cygwin
19:16:06safetydangodzirra, while in the file browser root go to the menu and select Playlist -> Create Playlist
19:16:30safetydangodzirra, that should create a playlist with all your mp3s, you can then shuffle that list
19:16:49*safetydan really shouldn't code and cook dinner at the same time
19:17:30godzirraHmm.. this is going to take a while.
19:18:06godzirraEh not too bad.. about 15-20 seconds to create a playlist.
19:18:46godzirrahmm.. after loading root.m3u, it tells me >TPPY for the title and is just sitting here.
19:18:57godzirraDoes it take a while to load the whole playlist, or did something horribly wrong happen?
19:19:04godzirraNow it says >FPVC :/
19:19:49godzirraWoo. I'm up to OAGL
19:20:05godzirraAny ideas on what is happening? :/ its not playing any music btw.. and it is advancing a song everytime it changess.
19:20:20godzirraOhhhhhh... you know what I think is happening?
19:20:25safetydangodzirra, what format is your music? Those look like iTunes files
19:20:25godzirramy foo_pod thing that didnt work
19:20:26webguest13-2godzirra: AACs?
19:20:32godzirraI think its grabbin those.
19:20:40godzirrahow do I find out where those are and delete them?
19:20:52*linuxstb scratches head and stares at this line in doom: floor->sector = &sectors[(int)floor->sector];
19:21:51safetydanCan the user who opened a bug reopen it after it's closed?
19:21:53 Quit bobTHC ("Smoke Weed Every Dayz !!!!!!!")
19:22:09preglowi think only admins can reopen
19:22:22 Quit austriancoder ("Kopete 0.11 :")
19:22:48linuxstbI've reopened a bug in the past - so I think all developers can.
19:22:54linuxstbNo idea about users though.
19:23:50safetydanhrmm... so maybe I shouldn't close these then
19:24:45linuxstbAm I right in thinking I should (in general) cast pointers to unsigned long? (for 64-bit safety)
19:25:43 Join dpro [0] (n=x@
19:26:15godzirraAnyone aware that none of the links on this page work:
19:28:27linuxstbgodzirra: Yes. Someone broke it this morning. He's promised to come back and fix it...
19:28:36 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
19:29:07godzirrajust making sure. Should I wait to upload a wps then?
19:30:29linuxstbNo, I think it will be fine to add new WPSs.
19:30:44preglowlinuxstb: yeah, that's right
19:30:57 Join RotAtoR [0] (n=e@rockbox/developer/rotator)
19:33:29linuxstbWhat about the other way around - converting an int to a pointer. Do I need to double-cast somehow?
19:34:14Slasherilinuxstb: i am going to add the tracknumber and year tags to the tagcache soon :)
19:34:18josh_linuxstb: int->ptr is fine with one cast, ptr->int needs two
19:34:34 Join Rondom [0] (n=Rondom@
19:34:59linuxstbjosh_: Thanks. How does that look? p=(void*)(long)i;
19:35:12linuxstbSlasheri: Nice. Will there be a browse by year option?
19:35:36Slasherilinuxstb: yes, if that is coded to the ui =)
19:35:38josh_linuxstb: looks fine... sizeof ptr == sizeof long in most modern cases
19:36:24preglowanyone know how well ccache sits with several compilers on the system? handles it nicely?
19:36:34webguest13-2Slasheri: how about a <all> album? So that I can eg. show all songs on all albums by a specific artist
19:36:49linuxstbpreglow: That's how all the build servers work.
19:37:00 Quit huntermic (Read error: 104 (Connection reset by peer))
19:37:22Slasheriwebguest13-2: should be easy to do, only little modification to the ui
19:37:49preglowperhaps we should add uintptr_t as well?
19:37:50preglowthat is, stdint.h
19:37:55Slasherihmm, in fact please add these feature requests to the TagCache wiki page
19:38:04Slasherithen it would be easy to keep track on those
19:39:22 Join XavierGr [0] (
19:40:50linuxstbSlasheri: You don't want to encourage that :)
19:41:21godzirraok.. I think I got it uploaded right.
19:41:24godzirrahard to test when its broken =)
19:45:23 Join huntermic [0] (
19:45:29 Part huntermic
19:46:37 Join huntermic [0] (
19:47:23 Quit obo ("bye")
19:50:02 Join obo [0] (
19:56:31 Quit aegray (Read error: 110 (Connection timed out))
19:56:40 Join goffa [0] (n=goffa@
19:56:46goffaanyone here?
19:58:47markun_goffa: you have a question?
19:59:18goffathe codecs... are they specific to each model.. or is rockbox rockbox?
19:59:44webguest13-2They're not binary compatible, but the same source is used to compile them
19:59:54goffain other words... i have a lot of musepack files... and was thinking about flashing my iaudio in the near future because i heard that rockbox supported musepack
20:00:13goffabut... i didn't know if they had to develop codec support for each device
20:00:13preglowall rockboxes support musepack
20:00:18goffasweet deal
20:00:22preglowapart from the hardware codec ones
20:00:24webguest13-2(no seeking though)
20:00:24preglowthe archoses
20:00:37goffadon't care about seeking just want to be able to play them
20:00:42preglowthat it will
20:00:48webguest13-2Then you're in luck
20:00:49goffaabout 1/3 of my music collection is musepack
20:00:59preglowat least if they're sv7 files, that is
20:01:17preglowrockbox doesn't currently supports pre-sv7 files, but i'll have it done in a couple of days if you can supply me with files
20:01:45pregloweven the musepack people themselves seem to have some trouble in finding pre-sv7 files, so i think we'll be ok
20:01:58goffayeah.. i have plenty :)
20:02:05webguest13-2Isn't old source available?
20:02:10 Join aegray [0] (
20:02:11goffabut... i'd have to dig through
20:02:20preglowbut yeah
20:02:22preglowthey wont play
20:02:31preglowbut the codec supports them, so it is possible
20:02:32Slasherilinuxstb: hehe, now i have added "feature requests" section to the wiki page
20:02:43goffathey just got sound support for the iaudio x5 yesterday.. i'm pretty excited about that
20:02:59preglowwasn't it today?
20:03:03goffamaybe it was
20:03:12webguest13-2Still not committed, I think
20:03:18preglowno it'snot
20:03:30preglowbut yeah
20:03:40preglowx5 will, just like h120, have excellent support for musepack
20:03:48preglowif anything with no seeking can be called excellent
20:03:59preglowbut it's the best we can do until the musepack people find some clever way to do seeking
20:04:03goffai NEVER use seeking
20:04:08goffaso for me it will be excellent :)
20:04:09preglowthen hooray!
20:04:30godzirrawhats musepack?
20:04:38goffajust a different file format
20:04:49goffalike ogg, mp3, aac, etc
20:05:02preglowbut yeah, i gotta go again
20:05:09goffaabout the same bitrate as mp3... better sound quality.. (arguably)
20:05:13preglowi'm pretty much the musepack guy around here, so let me know if you have any problems some day
20:05:22goffathanks for the help
20:05:36goffaand i might have to send you some samples
20:05:40goffain fact
20:05:43preglowsounds sweet
20:06:08 Join Bger [0] (n=Bager@rockbox/developer/Bger)
20:06:11 Join hyl [0] (
20:10:38Slasherilinuxstb: Hmm, can metadata engine parse track number directly from filename if real tag is missing? Adding that functionality would be good :)
20:10:49 Join rockpod [0] (
20:12:43 Quit tianjing (Read error: 110 (Connection timed out))
20:13:08 Join ze__ [0] (
20:14:43 Quit ze (Read error: 104 (Connection reset by peer))
20:14:43 Nick ze__ is now known as ze (
20:15:52preglowSlasheri: shouldn't be too hard, i guess, but does it belong in the metadata reader?
20:16:32Slasheripreglow: i think that should belong there, at least it would be more logical if metadata reader would assign all tags
20:16:55preglowi guess extrapolating that particular tag wouldn't hurt
20:17:17 Join Matze [0] (
20:17:27 Join PhR3aK [0] (
20:17:54Slasherihmm, it seems these new tags are very easy to add to tagcache. Only little modification is needed to handle these simpler numeric tags and not to create separate index files for these :)
20:18:28 Quit dpro (Connection timed out)
20:19:11DBUGEnqueued KICK PhR3aK
20:19:11PhR3aK(LinusN): Will there be a downloadable version for the x5 soon that supports sound?
20:20:34preglowtime to set the topic again
20:21:41 Join dpro [0] (
20:21:43 Nick hyl is now known as qwm (
20:21:55LinusNPhR3aK: yes, if i find the time
20:22:31 Join SereR0KR [0] (
20:23:39 Join Rob2222 [0] (
20:23:50 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
20:24:00 Join ^BeN^ [0] (
20:24:18 Quit Paprica (Nick collision from services.)
20:24:28 Nick ^BeN^ is now known as Paprica (
20:25:39 Join Gibbed [0] (
20:25:43 Quit Rick (Nick collision from services.)
20:25:55 Nick Gibbed is now known as Rick (
20:28:24safetydanAnyone familiar with the replaygain code?
20:28:55preglowmildly, how?
20:29:17safetydanAny particular reason it keeps a pointer to the string representation?
20:29:30safetydanI know it gets reused in the show id3 screen,
20:29:39safetydanbut not all formats have replaygain values as a strin
20:29:57 Join qwm_ [0] (
20:30:25amiconn_linuxstb: Still one 64bit warning left, and it seems your fixes added another warning...
20:30:28preglowsure, and for those, you print it
20:30:32preglowi do that for musepack
20:30:59safetydanyes but what if you have no where to store the string?
20:31:13amiconn_linuxstb: again...
20:31:52 Join _FireFly_ [0] (
20:33:09amiconn_linuxstb: Also, some of your fixes are looking dangerous, like this one:
20:33:25preglowsafetydan: *shrug*, you usually do
20:33:26amiconn_I assume zonebase is an int, but it must be a long to work properly
20:33:33CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
20:33:33*webguest13-2 learned today that recent LAME versions adds replaygain tags as well.. in the mpeg header
20:33:35preglowsafetydan: i save it in the seek point buffer of mp3, i think
20:33:51amiconn_This code is not only wrong for 64bit, but also for 16bit
20:33:58webguest13-2So each file could have RG tags in 3 places.. mpeg header, apev2 tag, id3v2 tag
20:34:14safetydanjust trying to add rva2 tag parsing and wondering why we keep a string representation if the id3 screen could just regenerate it
20:34:21amiconn_casting int <-> ptr just works by coincidence on 32bit systems...
20:34:53preglowsafetydan: i bet the replaygain format changes between codec header types
20:35:02preglowsafetydan: we coulf of course change it to one internal rep, but yeah
20:35:04preglowno one did it :>
20:35:10safetydanwe do change it to one internal representation
20:35:15safetydans7.24 :)
20:35:43preglowthen hooray!
20:35:53preglowi'd be happy to see that usage of internal buffers go
20:35:58safetydanSo no one will object if I remove the string part then?
20:36:07preglowif you make it work, then hell no, go ahead
20:36:14preglowit just slows down get_metadata
20:38:36webguest13-2No news from ifp port lately?
20:38:52 Quit Rob2222_ (Read error: 110 (Connection timed out))
20:41:15 Quit rockpod ("CGI:IRC (EOF)")
20:42:15 Quit amiconn (Read error: 110 (Connection timed out))
20:42:16 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
20:42:41preglowwebguest13-2: nope
20:43:04 Quit qwm (Connection timed out)
20:43:43 Quit Nico_P (Remote closed the connection)
20:44:33linuxstbamiconn: zonebase is a pointer.
20:46:13linuxstb(it's the return value of plugin_get_audio_buffer())
20:46:52amiconnhmm, ok
20:47:25amiconnThen it wasn't necessary to change the first cast, only the second.
20:47:54amiconnThat's why I thought zonebase might not be a pointer
20:48:11linuxstbI know, but it seemed clearer to me to make all the expression longs, and then cast the result to a pointer.
20:48:45amiconnAlso, I don't understand what's going on in your first fix, this one:
20:49:02amiconnThis function is declared int, but seems to return a bool?
20:49:42linuxstbThe return value was just being used in an if statement - e.g. if (P_Sector_Active(x)) { ...
20:51:58linuxstbThere's some nasty mixing of pointers and integers going on in this code...
20:52:20 Quit _FireFly_ ("Leaving")
20:52:24PhR3aK(LinusN): Is only mp3 playback working? or do other formats also work?
20:53:11webguest13-299,9% sure it's all formats
20:53:25safetydanPhR3aK, I'd imagine if sound is supported then all the formats that rockbox supports will work
20:53:27webguest13-2Rockbox doesn't differentiate between codecs
20:53:52b00st4hmm, any1 could help me with fxp settings?
20:54:10linuxstbSlasheri: The problem with using get_metadata() to infer the track number is that get_metadata() only knows about a single track. The track number could be inferred from that file's alphabetical position in its directory.
20:54:40PhR3aKwill there be a cvs release for the x5 soon that supports sound?
20:54:46safetydanI never really liked guessing of metadata.
20:54:50safetydanToo easy to go wrong.
20:55:04safetydanPhR3aK, there will be. When it's done.
20:56:44webguest13-2PhR3aK: 5 minutes after it's done, you'll be able to download it
20:57:26LinusNPhR3aK: patience, grasshopper
20:58:15 Join Nico_P [0] (
21:00:41amiconnlinuxstb: A properly tagged album track should have the track number as a tag
21:01:53Papricacould someone take a look on this?
21:02:01linuxstbYes - but Slasheri was asking if get_metadata() could fill in that info if there was no tag.
21:02:01Papricaits a scaled album art bitmap
21:02:07Papricabut there is a little problem
21:02:17linuxstbPresumably by checking the filename.
21:02:20Papricaits view half bitmap scaled
21:02:31Papricaand the other half is garbage =\\
21:03:30Chamoiswhere has to be a .wad for Doom ?
21:04:47PhR3aKis rockbox able to open other picture formats than jpg?
21:05:00Chamoisanyone knows ?
21:05:11linuxstbChamois: One moment, and I write a quick wiki page...
21:05:12 Quit muesli- (Read error: 110 (Connection timed out))
21:06:22Nico_Pall people intersted in album art : i've edited my recent comment with a few explanations...
21:06:48Nico_Pi'm getting closer to the system where no memory is wasted
21:07:10Nico_Pi strongly suggest you try it ;)
21:07:30PapricaNico_P, i'm trying to get the album art scaled..
21:07:58Nico_PPaprica: nice :) how's it coming ?
21:08:44Papricai have finish it
21:08:52Papricabut i have some problem
21:09:17Nico_Pi've just noticed your code above
21:09:22Nico_Pwhat sort of problem ?
21:10:41Papricamm i'm compiling the simulator now, i'll take a screenshot
21:11:25Nico_Plooks nice, and it should be quite easy to add to the rest of the code, too
21:11:51linuxstbChamois: The start of some instructions:
21:11:57Nico_Phwo have you combined it whith the rest of the code atm ?
21:12:02***Saving seen data "./dancer.seen"
21:13:24 Quit quobl (Remote closed the connection)
21:13:59webguest13-2linuxstb: Screenshots!
21:14:18webguest13-2Preferably actual photos, because that's just that much cooler
21:14:32amiconnlinuxstb: Should doom work in the sim?
21:14:35Chamoislinuxstb :thx
21:14:39 Join quobl [0] (
21:15:34linuxstbamiconn: Yes, the H300 sim works for me (Debian/x86)
21:16:02amiconnI'll try on amd64 then...
21:16:21amiconnHmm, where's that prboom thingy again?
21:16:29linuxstbRead the wiki page :)
21:16:34 Quit Matze ("Miranda IM! Smaller, Faster, Easier.")
21:16:45linuxstbI think we should distribute it - afaik, it's GPL'd.
21:16:55linuxstb(I mean in
21:17:07linuxstbBut maybe wait until we split the downloads.
21:18:20amiconnIt's not that big. If it compresses well, we could include it
21:18:43linuxstbIt compresses to about 110KB with zip.
21:19:50 Join Scootscat [0] (
21:20:29Nico_Plinuxstb: any chance for you to have a lokk at my album art patch ?
21:20:32webguest13-2That's a quarter of unifont
21:22:00amiconnThe fixed mas_writereg() doesn't fix the recording bitshift, unfortunately :(
21:22:18amiconnNot that it would have been very likely...
21:23:27 Quit Febs ("CGI:IRC (EOF)")
21:25:31godzirraHow do I shuffle an existing playlist?
21:26:39 Join muesli__ [0] (n=muesli_t@
21:27:07 Join webguest26 [0] (
21:27:53safetydangodzirra, load it up and switch play mode to shuffle
21:28:28godzirraThanks ;)
21:29:42 Quit webguest26 (Client Quit)
21:30:00 Quit c0utta ("CGI:IRC")
21:31:32godzirraHmmm... my rockbox now skips everytime I change to a new song. :/
21:31:40linuxstbamiconn: Any luck with doom on amd64?
21:31:43godzirraI'm not sure why.. the song will start, it will stop playing for a second, then start up again.
21:31:54amiconnYes, it's working fine. But how the hell can I fire
21:32:29amiconnOh, and btw, not having forward on 'Up' is really confusing. Why is that?
21:32:37Nico_Pgodzirra: you can also do it without changing the play mode : long press NAVI, playlist > reshuffle
21:32:49Nico_Pthat's on H300
21:33:00linuxstbamiconn: I'm guessing that's the way all the h300 users wanted it. I don't understand either though.
21:34:14amiconnHmm, just found fire. Why that is put on Mode is also beyond me
21:34:17safetydangodzirra, it's probably having the eq enabled, the ipod version of Rockbox isn't fast enough to play mp3 and use the equalizer at the same time
21:34:45amiconnI would put fire on Navi, Forward on Up, change weapon on Mode
21:35:01 Quit TCK (Read error: 110 (Connection timed out))
21:35:33 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
21:35:34godzirrasafetydan: wow.. that sucks. :/
21:35:56amiconnOh, it crashes on exit :(
21:36:01safetydangodzirra, it will get better, but for now, just turn off the eq
21:36:02linuxstbMaybe those keys were chosen to make common combinations easier to press on a real h300. You should ask kkurbjun.
21:36:36linuxstbWas that crash on the sim, or target?
21:36:42linuxstb(the sim exits fine for me)
21:37:04 Join TCK [0] (
21:40:04amiconnlinuxstb: In-game menu: Quit Game, Navi ("press y to..."), Navi (rockbox-blue screen) -> SIGSEGV
21:40:13amiconnThat's with doomu.wad
21:40:34 Join [TCK] [0] (
21:40:34linuxstbI'm using doom1.wad (the shareware wad).
21:40:43amiconnHmm, now it just crashed within the demo...
21:43:25 Quit SereR0KR ("XChat Aqua")
21:43:51Nico_Plinuxstb: any chance for you to have a look at my album art patch ?
21:44:24linuxstbNico_P: Yes, I've just been looking at it.
21:44:44linuxstbIs it working reliably now?
21:45:07 Quit Scootscat ("CGI:IRC (EOF)")
21:46:00 Join DrMoos [0] (
21:46:24Nico_Plooks like it
21:46:37linuxstbIt's good that (as far as I can see) the patch doesn't need any extra memory - storing the images in the audio buffer, and then copying it to the existing wps image buffer before using it.
21:46:45safetydanin struct mp3entry, track_gain and album_gain are amplitude not dB right?
21:46:52Nico_Pi've been playing music on my h300 with it since i updated it, and it never crashed
21:47:42*amiconn actually went back to archos for actual usage
21:48:04linuxstbNico_P: I still think the actual album art loading functions should be in get_metadata() though, and not playback.c
21:48:54 Join stripwax [0] (
21:49:46Nico_Pyes, i'll try to change that... but i initially chose that place because it seemed right : it does the buffering of the actual track...
21:50:01Nico_Pwhere is the metadata written by get_metadata ?
21:50:23Nico_P(i'm not on my computer right now so i can't have a look at the code...)
21:51:05stripwaxIn what order are tracks of an album displayed when using tag view? Doesn't seem to be either alphabetical or track number order
21:51:26stripwaxbtw tagcache is v. v. cool indeed. props to all involved!
21:51:32linuxstbNico_P: get_metadata() is passed a pointer to a track_info struct. It fills the id3 part of that.
21:51:57*linuxstb points stripwax towards Slasheri
21:52:12*stripwax shakes Slasheri warmly by the hand
21:52:22linuxstbstripwax: I think Slasheri is working on that issue now.
21:52:28stripwaxah, neat-o
21:53:09linuxstbNico_P: So you will need to pass more parameters to get_metadata() - so it knows where it can load the bitmap to.
21:53:29amiconnI thin that album art patch is too early
21:53:31Nico_Pyes, like a pointer to the current write position in the file buffer ?
21:54:05amiconnAfter the playback engine cleanup (and the 3.0 release) we're going to put metadata in the main buffer
21:54:18Nico_Plinuxstb and about what you said earlier : it indeed doesn't need any extra memory, but the copying to the WPS image buffer needs to be done at another place... currently it gets done several times per track change and maybe a bit too late
21:55:01 Join webguest99 [0] (
21:55:38 Quit xmixahlx ("blah blah blah")
21:55:42Nico_Pamiconn: once it works correclty, it probably won't be too hard to move the loading code
21:56:12Nico_Pit can even be moved with the rest of the metadata loading
21:56:19 Quit safetydan (Read error: 104 (Connection reset by peer))
21:56:39Nico_Pi don't think it will be much of a problem
21:56:53 Nick qwm_ is now known as qwm (
21:57:00 Quit webguest99 (Client Quit)
21:58:23 Quit TCK (Read error: 110 (Connection timed out))
21:58:57tucozI accidently assigned a patch to me. Is it possible to change that, as I am not sure I will be around when/if an updated version of the patch is posted?
22:00:33tucoznevermind, I was brave and clicked on the 'Edit this task' button. :)
22:01:56 Quit webguest13-2 ("CGI:IRC")
22:03:02Nico_PPaprica: any news on bitmap scaling ?
22:03:08Papricano =\
22:03:28Papricait doesnt work on the simulator dont know why
22:03:34Papricai'll try to do somthing
22:03:38 Quit Moos (Read error: 110 (Connection timed out))
22:03:52 Nick DrMoos is now known as Moos (
22:04:33amiconnWhy oh why did this vm decide to perform an fsck... :/
22:06:24Nico_PPaprica: did you upload a screenshot ?
22:06:34Papricayeah sec
22:08:22Nico_Pso this is the right size ?
22:08:35Papricathis is scaled size
22:08:43Papricathe right size is 100x100
22:08:47Nico_Pyes i meant it's scaled to the right size
22:09:19Papricamm just a sec, im trying to do somthing with it
22:09:36 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
22:10:36safetydangreat... linux locked up twice in a row with no oops or any other log message
22:11:57ender`what did you do?
22:13:01safetydanremote X session from another laptop
22:13:17safetydanI blame the wireless card driver personally, flaky little thing that it is
22:17:16 Nick BoD[sleeping] is now known as BoD[] (
22:17:57BoD[]Hello world!
22:18:31BoD[]hey does anybody know if the tag db is supposed to work with the ipod target?
22:19:01tucoztagdb or tagcache?
22:19:33tucozmaybe the tagdb is removed?
22:19:44BoD[]heeeeeeerm :) I don't know ?
22:20:00BoD[]the db you create with the java program
22:20:12BoD[]and then you say "id3 tags" in file view
22:20:19tucozoh, that. Use the tagcache instead. That is the way of the future :)
22:20:35BoD[]how do I do?
22:20:41BoD[]and what's the difference
22:20:56tucozOn target vs off target
22:21:07BoD[]wow it's on target ?
22:21:08tucoz...building of the db
22:21:22tucozthere is a wiki, let me find it
22:21:44BoD[]thank you very much
22:21:55tucozbeware of bugs though
22:22:17tucozBut it looks really promising and is rather functional at this early stage
22:22:36BoD[]good :)
22:23:16 Join ashridah [0] (
22:26:03BoD[]in other news, I'm having lots of "freezes" on the ipod
22:26:14BoD[]when I press next
22:26:35linuxstbThen don't press next :)
22:26:36 Join qwm_ [0] (
22:26:39BoD[]haha :)
22:26:57linuxstbWhat do you mean by "freeze"? Do you have to reboot?
22:27:17BoD[]it shows the next song in the wps
22:27:27BoD[]but playback doesn't start
22:27:38PapricaNico_P, huh,
22:27:47BoD[]and everything's frosen including the backlight staying on
22:27:50Papricai scale it and rotate it lol
22:28:17Nico_Pis it the right size... i mean scaled correctly ?
22:28:25Nico_Paside from being rotated
22:28:48Papricascaled correctly
22:28:53Nico_Pcould you put the code on pastebin ?
22:29:44amiconnlinuxstb: On 32bit sims, doom doesn't crash on exit (tried both cygwin and debian). So it must be a 64bit issue...
22:29:51PapricaNico_P, just a sec
22:29:57Nico_Pis it really rotated ? or is it just a matter of x and y being inversed ?
22:30:02Papricaim fixing the rotating
22:30:39 Quit ender` (" We know the speed of light... so, what's the speed of darkness?")
22:30:50 Join ender` [0] (i=ychat@
22:31:04amiconnlinuxstb: Doom says this on exit (sim): "There were still: 1 files open"
22:31:45PapricaNico_P, Done!!!
22:31:53 Quit b00st4 (Read error: 110 (Connection timed out))
22:32:31linuxstbamiconn: My sim doesn't say that...
22:32:43Nico_Plinuxstb: what time do you think would be best to do the buffering ?
22:32:55amiconnlinuxstb: That's on debian/x86...
22:32:58linuxstbNico_P: Which buffering?
22:33:01Nico_Psorry i mean the scaling
22:33:26PapricaNico_P, it scaled from 75*75 to 50*50
22:33:38linuxstbYou have two choices - when loading the bitmap, or when copying the bitmap to the WPS image buffer.
22:34:08linuxstb(or I guess when displaying the image, but that seems a bad choice)
22:34:24amiconnlinuxstb: On target, doom hangs on startup :(
22:34:27PapricaNico_P, i did it when the bitmap is loading
22:34:40Papricanot loading
22:34:43linuxstbamiconn: The only message I get from the sim when quitting is We open the real file 'archos/games/doom/default.dfg'
22:34:57safetydanDoom's not ported to grayscale lib is it?
22:35:07amiconnHmm, second time it didn't hang....
22:35:08 Join iJason [0] (
22:35:25linuxstbWell, we have a month to fix it...
22:35:42amiconn'Play Game' now hangs it...
22:36:18*linuxstb summons kkurbjun
22:36:28 Quit iJason (Client Quit)
22:36:41linuxstbsafetydan: No, not yet.
22:36:57amiconnReproducable :(
22:37:16Nico_Plinuxstb: thinking about it, i don't really see how i could scale before writing to the file buffer
22:37:36amiconnThere's some more strangeness:
22:37:50Nico_Pand it's probably better to have the original bitmap in the buffer, and the scaled one only for the WPS
22:37:57Nico_Pseems more logical
22:38:27amiconn(1) In the sim, it immediately enters the 'Game' submenu after start (all of debian 32bit, 64bit and cygwin). On target it stays in the main menu
22:38:37 Quit lodesi ("Leaving")
22:38:43Nico_PPaprica: pastebin ?
22:38:52linuxstbYes, I get that on the sim. On the ipod, it enters the game menu as well.
22:38:55PapricaNico_P, no, patch
22:39:01Paprica3 minutes
22:39:01Nico_Peven better
22:39:03amiconn(2) The 'Game' submenu doesn't work correctly. Using Navi selects and exits, and immediately re-enters
22:39:22Nico_Pi'll probalby update my test version with it
22:39:30amiconnNo luck in starting any game on target. Freezes always.
22:40:26 Join Sinbios [0] (n=Sinbios@
22:40:29PapricaNico_P, try it and tell me what you think
22:40:33 Quit bam__ (Remote closed the connection)
22:40:46amiconnHmm, disabling voice did the trick. That's really really strange...
22:41:09linuxstbYes - voice seems to cause unexpected problems everywhere...
22:41:14Nico_PPaprica: could you send it to flyspray ?
22:41:34Papricammm... but its not all the patch...
22:41:41Nico_Pmy client rejected it.. i changed the setting but i'm not on my computer anyway
22:41:49Papricaok sec
22:41:54 Quit qwm (Connection timed out)
22:41:57linuxstbWould the fact that doom uses the standard menus, and also claims the audio buffer cause the problem?
22:42:12 Quit Nico_P ()
22:42:29linuxstbOr iram...
22:44:34muesli__will there be dual boot for x5 at all one day?
22:44:46LinusNmuesli__: yes
22:44:56muesli__oki :-)
22:44:58MoosLinusN: you are one "God" !
22:45:06Mooshehe :)
22:45:25*Moos si ready for listen music with Rockbox on X5
22:47:05BoD[]:( ... I made a cool bdf font with fontforge, but now I can't read it anymore
22:47:06amiconnlinuxstb: Target just said there were still 3 files open...
22:47:31PhR3aKis the actual cvs release (20:36) supporting sound? and if not, how can i see that sound works? are there no warnings then or what?
22:47:56PhR3aKso no yellow or red boxes
22:48:54BoD[]i'm sad
22:49:08safetydanis not sending emails any more?
22:49:23safetydanor possibly it's my mail server hrmm..
22:49:29BagderI got linus cvs commits a few mins ago
22:49:36Bagderby mail I mean
22:49:56LinusNMoos: the build is ready, go go go!
22:50:17Bagderstill yellow!
22:50:34LinusNyes, i haven't removed the bass/treble settings, for example
22:50:46Bagderone server down I see
22:51:03Chamoismusic on x5
22:51:20Chamoisit works
22:51:22dpassen1How's it sound? Decent?
22:51:25Chamoisi love you Linus
22:52:06LinusNhmm, i still can't reach the menu from the wps
22:52:06*ashridah hmf's
22:52:11PhR3aKlinus, you rule!
22:52:24preglowLinusN: do you also think we shouldn't bother to do a software implementation of the bass and treble boost when we have an eq?
22:52:26ashridahall of these coldfire platforms reaping the benefits of the hard yakka put into the H1xx series port
22:52:35MoosLinusN: dowloading
22:52:46ChamoisMoos : already have it
22:52:55Bagderyakka being the technical term then ;-)
22:53:06amiconnLinusN: volume in sound.c for TLV320 is off...
22:53:18BoD[]is there a fnt to bdf utility?
22:53:19amiconnYou allow -73 to 0, but set a default of +6 ...
22:53:30BoD[]by any crazy chance
22:53:45BagderBoD[]: why would you want that?
22:54:20muesli__XavierGr ?
22:54:22 Join Nico_P [0] (
22:54:38BoD[]Bagder: I made a cool font with fonforge.. I converted it with the utility, it works fine... But now fontforge refuse to load the font ...
22:54:39*amiconn wonders about the small volume ranges all the newer DACs allow
22:54:51BoD[]I don't know why... it must be corrupt I guess
22:54:55CassandraBoDO: Such a thing exists, yes. I used it to convert jackash to bdf format.
22:55:06BoD[]Cassandra: really??
22:55:10muesli__what means hello in greek?
22:55:19CassandraI think so - was a while back.
22:55:26LinusNamiconn: good catch, thanks
22:55:41LinusNwhat does "precut" do?
22:55:48preglowin the eq?
22:55:56preglowallow you do lower the volume before eq processing
22:56:06preglowto prevent clipping if you use any positive gains
22:56:09amiconnLinusN: Btw, you should set a volume default lower than 0 (provided that 0 is similar in actual volume to the other targets)
22:56:10LinusNhas no effect for me on the x5
22:56:17 Join Arrogant [0] (
22:56:26preglowi haven't tried it myself yet
22:56:27LinusNamiconn: check my commit
22:56:50BoD[]Cassandra: do you know where I should look?
22:56:58LinusNpreglow: maybe i'm silly, hang on
22:56:58XavierGrmeusli: You mean if there is a word spelled hello (there isn't), or the word that we use for it?
22:57:01safetydanLinusN, should work... unless you have replaygain on in which case I haven't fully sorted things (but should mostly work)
22:57:24MoosLinusN: you rocks !!
22:57:29CassandraLooks like I used psf as an intermediate stage and a package called psftools.
22:57:40muesli__XavierGr something that says: hello :-) whatever it is
22:57:47sharpea greeting?
22:58:03muesli__beginning of an email
22:58:13safetydanLinusN, you'll also need to enable the equalizer for the precut to have an effect
22:58:19LinusNi know
22:58:22 Nick qwm_ is now known as qwm (
22:58:27CassandraOh, hold on - you mean fon as in the Rockbox internal font.
22:58:29XavierGrWe say Geia (more like Wia or Gia) and come from Ygeia which mean Health. So Geia means have good health
22:58:52CassandraSorry - format confusion. I thought you meant something else.
22:59:03BoD[]Cassandra: oh :(
22:59:24BoD[]I don't know, it's weird, I'm looking at the bdf file, it looks fine
22:59:35BoD[]but fontforge doesn't agree :)
22:59:53preglowLinusN: but yeah, about the question about treble and bass boost, got an opinion?
22:59:58Cassandrafontforge can be a little weird sometimes.
23:00:10LinusNpreglow: the eq is enough imho
23:00:11 Join maeck [0] (n=chatzill@
23:00:25preglowok, then we'll just disable the other ones for x5 then
23:00:27BoD[]Cassandra: yeah :( it crashes often too... well I'm on cygwin, maybe it's not the best use
23:00:49preglowi have some cheap treble and bass boost filters coded ready, but unless they're really, really good, i won't bring this up any further
23:00:51LinusNah, the precut is only updated when i leave the precut setting screen
23:00:54 Join iJason [0] (
23:01:08muesli__thx XavierGr
23:01:30preglowwe don't want that
23:01:37BagderLinusN: perhaps you could put an x5v boot there to see if the concept works as presumed
23:01:46Bagderthere => in the wiki
23:01:49PhR3aKhuhh man i thought ive crashed my x5... some files could not be extracted, so rockbox wont boot, so i wanted to flash back by connecting the x5 to the host port, but he turn off every time because he wanted to load rockbox, the ive tried it over the subpack and it worked
23:01:57PhR3aKhell i had a fear
23:02:02LinusNBagder: can you have a look at the wps menu key?
23:02:11Bagderme check
23:03:33Bagderhm, WPS_MENU not defined for x5...
23:04:01Bagderbut rec is unused
23:05:46 Quit [TCK] (Read error: 110 (Connection timed out))
23:05:48 Join phaedrus961 [0] (n=dfdsfsdf@
23:06:09LinusNBagder: i think we should use the same key in the tree and wps
23:06:39 Join webguest22 [0] (
23:07:24webguest22would any ipod rockbox devs be so kind as to investigate an bug for me?
23:07:28BagderI guess perhaps rec is better for the file tree view anyway
23:08:11Chamoisi guess too
23:08:13 Join Febs [0] (
23:08:38Bagderok, so long press on REC => menu
23:10:52LinusNx5v bootloader is on IaudioBoot
23:11:05PhR3aKis the battery status working? so does it really show how full the battery is?
23:11:44 Join solexx [0] (
23:12:05***Saving seen data "./dancer.seen"
23:12:59LinusNPhR3aK: it should be fairly accurate
23:13:12sharpefairly, is the keyword :D
23:13:20 Join webguest62 [0] (
23:13:25LinusNbut we haven't calibrated the discharge curve and the power consumption estimations
23:13:41webguest62lets go
23:14:32webguest22anyone checked on this bug?
23:14:38 Join linuxstb_ [0] (
23:14:43preglowanyone think that four settings for the crossfeed is too much for one feature?
23:14:54 Quit linuxstb (Nick collision from services.)
23:14:56 Nick linuxstb_ is now known as linuxstb (
23:15:07dpassen1preglow: agreed
23:15:15preglowdpassen1: i didn't say i think so
23:15:23dpassen1well, then i'll be the first
23:15:29safetydanpreglow, I don't think so so long as the defaults are sensible
23:15:34preglowi think it's ok if the options are meaningful
23:15:38preglowand that they are
23:15:58safetydanSo if someone really wants to muck with it, they can, otherwise they just switch it on and off
23:16:22Chamoiswhy not a short press on REC to go in the menu
23:16:24preglowsafetydan: sure, but it's also possible to compress several options into one, but i for one don't really like the loss of flexibility that gives
23:16:29Chamoisa short press too to exit
23:16:33Chamoisfor x5 I mean
23:16:38Bagderrec exits the menu already
23:17:06 Join RedBreva [0] (
23:17:13Chamoisso short rec to go in can be good ?
23:17:26BagderI just think that you don't need to reach the menu in a hurry
23:17:36Bagderso it could be good to save the short press
23:17:46Bagderbut I don't care very much
23:17:53safetydanpreglow, so leave it with all the settings and if too many people get confused, simplify the settings
23:18:03LinusNpreglow: crossfeed is for picky people, and picky people want to be in control
23:18:31safetydanokay, I have precut updating in real-time now
23:18:33preglowthere might be another option as well, if i have enough registers left over :-)
23:18:52ChamoisBadger : so if you want lONG PRESS
23:18:55*Bagder builds custom keys for x5 with langv2 support
23:18:57Chamoisi don't care too
23:19:00PhR3aKaahh great... viewing pictures while listening to music...
23:19:01RedBrevaHi, quick question for anyone with Ondio knowledge... what buttons would you press to duplicate the functionality of F1, F2, F3 on the Recorder?
23:19:58BagderChamois: I'll try the long press and see how it feels
23:20:50Bagdermy god the x5 has a sexy remote
23:20:53Chamoismaybe you can add to the cvs the long press REC for the WPS at least
23:21:00LinusNBagder: indeed
23:21:11preglowsafetydan: updated the manual with precut info, btw? :>
23:21:31 Quit webguest22 ("CGI:IRC")
23:21:36safetydansafetydan, uhh yeah... about that...
23:21:39safetydannext commit? :)
23:21:47Febsput it in the wikimanual if you do...
23:21:59preglowFebs: why?
23:22:00 Quit Chamois (" Like's GUI? Then try HydraIRC -> <-")
23:22:07Bger <= this one ?
23:22:16 Quit Rondom ("I'm leaving on a jetplane, don't know when I'll be back again...")
23:22:37Febsi'm on a plane that's about to take off and I am going to be working on that chapter of the manual en route.
23:22:59Febswow, ironic quit message above my post.
23:23:03Bagder"iAUDIO - Friends & Lovers.mp3"
23:23:41*Febs has become a certified gadget frak.
23:23:54preglow'frak' sounds more gadgety!
23:23:58 Quit solexx_ (Read error: 110 (Connection timed out))
23:24:49FebsNot easy to type really fast on a Treo's keyboard.
23:25:42 Join webguest48 [0] (
23:25:57preglowi believe you
23:27:38tucozRedBreva, write something like "Please correct this key" if you don't know the correct keys for a specific target
23:27:54XavierGrSo to play Doom: 1)I need the doom1.wad in which directory? 2)Anything else?
23:28:00tucozRedBreva, like I've done in a few tables
23:28:14RedBrevaCool - cheers...
23:28:32tucozRedBreva, that is \opt{ONDIO_PAD}{Please correct this key}
23:28:35tucozfor instance
23:28:41*webguest48 suggest using "xxx:" or "fixme:" so as to easier find those
23:28:44Bagdermy x5 sounds fine
23:28:55tucozwebguest48, ...or that, yes
23:29:19PhR3aKsometimes my x5 lags when i browse through the menu
23:29:29PhR3aKsry i mean the sound lags
23:29:51ashridahPhR3aK: skipps you mean?
23:30:16webguest62thank u lot 4 everything youve done for my ipod btw
23:30:18PhR3aKespecially when i change background colour and so on
23:30:32ashridahPhR3aK: there's probably still a bit of optimisation to go
23:30:58 Quit ashridah ("uni")
23:30:59PhR3aKanyway... great work... rockbox rules!!!
23:31:45Bagderehum, my language strings are all mixed up...
23:31:56webguest62give yourselfs a pat on the back
23:31:56BagderI guess I should work on my script some more ;-)
23:32:37t0masBagder: I spot 1 missing server..
23:32:46PhR3aKsomeone should try to set background black and foreground white... looks damn good on the x5 :)
23:33:10t0maswell... everything still works
23:33:21Bagderyeps, very neat
23:33:24t0masbut just out of curiosoty: what happened?
23:33:29 Join Doomed9 [0] (
23:33:30Bagderwith what?
23:33:38t0masthe serve
23:33:45Bagderask Zagor, its his server
23:33:51BagderI have no idea
23:33:54amiconnLinusN: See here and here :-)
23:33:59 Part stripwax
23:34:10webguest48Someone should develop a blocky, wide font and set the background to blue and foreground to light-blue
23:34:15Bagdert0mas: I fixed a little flaw in your sleep => retry thing from last night
23:34:15safetydanpreglow, I found out why we keep the string value for Replaygain... it's because we only store the amplitude and to display dB again we'd need a log() function
23:34:24t0masBagder: hm? what was wrong?
23:34:28preglowsafetydan: sounds sensible
23:34:38Bagdert0mas: you did the goto before the sleep, so it never slept ;-)
23:34:41safetydanSo I've left that alone
23:34:50Doomed9i have a suggestion for the tag cache: a directory option,
23:34:51t0masI changed that after my last test
23:34:55Bagdert0mas: one of them logs got 2.7MB ;-)
23:34:56safetydananyway, night all
23:34:57 Quit safetydan ("Leaving")
23:35:01Doomed9so u can see roms and stuff
23:35:01webguest48Doomed9: a what now?
23:35:20 Quit Febs ("Time to fly")
23:35:24webguest48Wouldn't that be.... not using the tagcache?
23:35:27Doomed9i have a suggestion for the tag cache: a directory option, so you can see roms and stuf
23:35:36t0masBagder: now we just need to find out how to output messages from all threads simultaneous...
23:35:45Doomed9webguest48 thats not the point im trying to make,
23:36:09Bagdert0mas: yes! but when things seem solid we can probably remove that "sleep and retry" output
23:36:24t0maswhy remove it?
23:36:27t0masjust make it silent...
23:36:46Bagderhow is making it silent not removing output ?
23:36:51t0masoh wait
23:36:55t0masit's getting late again
23:36:58LinusNamiconn: nice
23:37:06*t0mas missed the "output" part of it
23:37:29t0masdamn... I've been coding like shit in the past week... and now even chatting is to hard for me :P
23:37:35Doomed9but okay
23:37:48XavierGrDoom was suppposed to have sound, no?
23:37:54amiconnLinusN: If you (or anyone else dealing with such low-level things) finds more of these, please add them
23:38:07XavierGrI remember kkbjurn had this. (though didn't test it)
23:38:09LinusNwill do
23:40:56webguest62is there any chance of the option to default boot into the apple firmware in the pipeline?
23:41:11amiconnpreglow: Is there a programming manual for arm, like for coldfire?
23:42:02 Join mirak [0] (
23:42:20PhR3aKwill dual booting be aviable soon?
23:42:37webguest62is there any chance of the option to default boot into the apple firmware in the pipeline?
23:43:04 Join amx109 [0] (
23:43:05Mikachumaybe it will be in another 2 minutes
23:45:23*t0mas runs off... time to go to bed
23:45:27 Join webguest36 [0] (
23:45:43*LinusN is tired too
23:45:47Nico_Plinuxstb: looks like i'd have to add a whole lot of new parameters to get_metadata() if i want to load album art from there... i'm not sure it's worth it
23:45:49*XavierGr too
23:45:56 Join Daishi [0] (
23:46:00*t0mas has found the reason for his sloppy coding and slowness btw
23:46:05t0masI'm getting old
23:46:15t0mas21 days and 15 minutes form now... I'll be 18 :P
23:46:21t0masplanned my driving lessons this morning
23:46:27Bagderand you don't even mean hex!
23:46:46*Bagder turns 0x24 this year
23:46:46t0masno please.. I'm only 12 in hex :P
23:47:30Bagderok, all kids go to bed now
23:47:31Bagderso so
23:47:33Bagderoff you go
23:47:47t0maseverybody runs... and L*nus is left alone
23:47:51t0mas(avoiding hiligh ;))
23:48:01Mikachui'm going to be 42 this year (in base 5)
23:48:58*LinusN is 10 in base 37
23:49:26BagderI'll be 33 in base 11
23:49:55*Mikachu uses his handy zsh function, anytoany () { echo $(( [#$3] $2#$1 )) | sed 's/.*#//' }
23:50:17t0masI'll go to bed before I get headaches from those calculations
23:50:22*amiconn is 10 in base 34
23:50:28Mikachui think i can trim out the sed if i use some weird shell constructs
23:51:33Bagderecho "obase=N; Y" | bc
23:51:46Bagderreplace Y and Y
23:52:13amiconnLinusN, Bagder: I tried to find something about the vmware tools license. Vmware itself says nothing about it, and there's nothing in the tools itself
23:52:26 Part amx109 ("Leaving")
23:52:27LinusNwhat is vmware tools?
23:52:38amiconnAll I could find is that vmware links to several community-built VM images, and some of them contain vmware tools
23:52:39 Quit webguest36 ("CGI:IRC (EOF)")
23:52:48t0masLinusN: a packet of tools to make vmware pc's display better
23:52:57t0masand have better mouse support etc
23:52:59XavierGrany news on the updated VMware image?
23:53:10LinusNthe vmware display driver is included in later versions of
23:53:35amiconnLinusN: Mouse synchronisation to host, display driver, shared folders, better networking driver
23:53:38maeckHi guys, I am new to this. Can anyone of you tell me whether it is possible to link something like Freetype into a plugin (rock)?
23:54:00LinusNmaeck: no
23:54:12sharpemy right hand is always cold, i don't get it...
23:54:26amiconnThe mouse synchronisation eliminates a major annoyance
23:54:49Mikachumaybe mousing too much stops blood flow somehow
23:55:00LinusNsharpe: your shoulder and neck muscles are probably stiff
23:55:05maeckDo I have to link into the base package and include things like that like the lcd tools?
23:55:07LinusNand stops the blood
23:55:23LinusNget a massage
23:55:32sharpeno one around...
23:55:36Bagdermaeck: to do what?
23:55:38sharpeout doing stuff.
23:55:42 Quit iJason ("CGI:IRC (Ping timeout)")
23:55:53maeckTry to keep the mouse in front of the keyboard instead of the right side. Gives better blood flow through the mousing arm
23:55:57LinusNsharpe: do it to yourself then
23:56:39sharpeheh, linus, what would you think of a 6502 cpu emulator for rockbox?
23:56:53LinusNshould work
23:57:02LinusNwe do need one to play sids
23:57:12Bagderthe cpu itself shouldn't be hard to emulate
23:57:15sharpei mean, interesting idea, bad idea, useless, etc
23:57:19sharpeit isn't...
23:57:28LinusNsid player is a wonderful idea
23:57:30Bagderthe hard part is the interesting c64 chips
23:57:54Mikachuatari .ym files would be fun too :)
23:57:56Mikachu<- optimist
23:57:57 Quit RedBreva ("CGI:IRC (EOF)")
23:58:20sharpe<- somewhat pessimistic
23:58:49LinusNgotta go to sleep now, cu tomorrow

Previous day | Next day