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 2006-08-23

00:00:01amiconnThat dreaded TI thing?
00:00:09BagderTMS320 based, yes
00:00:29Bagderbut the rest is an arm9
00:01:27Bagderand it could be a way to reach more tms320-based targets
00:02:17 Quit Arathis ()
00:06:17 Join JdGordon [0] (
00:07:06 Quit charkins ("Leaving")
00:08:51amiconnHuh? lcd_roll() is implemented for X5, I really wonder why...
00:09:14amiconnLinus added it on Jul 31, 2006 (!!)
00:10:05amiconnlcd_roll() is officially gone for quite some time... We don't need it in the core, and it is too hardware specific
00:14:55 Quit ender` (" Anyone who thinks people lack originality should watch them folding roadmaps.")
00:15:59 Quit MarcoPolo (Remote closed the connection)
00:20:05 Join Sacro [0] (n=ben@adsl-83-100-224-218.karoo.KCOM.COM)
00:20:40SacroI04: I11Instr
00:20:46Sacroat 1000C000
00:23:13linuxstbSacro: Are you going to say anything more? What were you doing when that happened?
00:24:00Sacrooh, sorry, its when i try and play an mp3, its an iRiver H340 with latest cvs build
00:25:42 Join petur [0] (n=petur@rockbox/developer/petur)
00:28:29Sacroill try it with last nightly build
00:31:39 Quit fatherfork ()
00:31:53 Quit Kohlrabi ("Quit")
00:33:32mirakI can't find how to pull out data from a adress register ...
00:37:52peturdid you map a variable on it correctly?
00:38:04 Quit murdock (Remote closed the connection)
00:39:04peturI mean, use the volatile keyword?
00:39:08 Part Sacro ("Leaving")
00:39:08 Quit Bagder ("time to say moo!")
00:40:08 Join EbErT [0] (
00:42:03 Join Arathis [0] (
00:42:13preglowlostlogic: didn't you make a patch for measuring codec speed? null output or something?
00:45:12 Quit lightyear (Remote closed the connection)
00:48:00 Quit BHSPitLappy (Read error: 110 (Connection timed out))
00:52:40 Quit oxygen77 ("Leaving")
00:52:53 Quit TeaSea (Remote closed the connection)
00:55:06 Quit petur ("Zzzz")
00:58:37 Quit dan_a (Read error: 110 (Connection timed out))
01:03:14 Join Davide-NYC [0] (
01:03:55 Join CriamosAndy [0] (
01:04:01 Join fatherfork [0] (
01:05:20Davide-NYCAnyone feel like tackling a bug?
01:05:34Davide-NYC(or a feature request, depending on howyou look at it?)
01:08:52 Join twisted` [0] (
01:09:02mirakgot a direct crash
01:13:35mirakif anyone is motivated to test some assembly code ...
01:13:59mirakit crashes. I don't know how I could debug
01:15:07mirakthe algorithm is the on in the "optimising embeded systems" pdf
01:15:17 Join Paul_The_Nerd [0] (
01:16:26Davide-NYCPaul_The_Nerd: the bug I just put up on flyspray is not for all targets. But many.
01:16:31Davide-NYCno matter?
01:16:47Paul_The_NerdDavide-NYC: It's probably for all targets with a Recording Screen, so good enough
01:17:19Paul_The_NerdEither way, there's no "A few" choice in the list. ;)
01:17:43Davide-NYCRadio buttons would be good for that
01:20:47 Quit barrywardell ()
01:20:54Davide-NYCPaul_The_Nerd: If you wanted to suggest that a single pixel gray line be drawn between the LCD and the RLCD in the Simulator, where would you post it?
01:21:09 Quit Criamos (Read error: 110 (Connection timed out))
01:23:05Davide-NYCPaul_The_Nerd: Feel free to delete
01:24:06 Join trougnouf [0] (
01:24:10Paul_The_NerdDavide-NYC: I'll leave it there, because for example, non-Developer programmers don't necessarily look at feature requests, so I think having it show up in the forums is also helpful.
01:24:32Davide-NYCwhat about my other question (see above)
01:24:38Paul_The_NerdDavide-NYC: As for that line, that's a Simulator Feature Request, I'd say
01:24:54Paul_The_NerdThere is a section specifically for simulator-only things. :)
01:25:38 Quit Arathis ()
01:26:44Davide-NYCon flyspray? in the forums? totally not seeing it
01:26:49Paul_The_NerdOn flyspray
01:27:00Davide-NYCfound it (doh)
01:28:04Davide-NYCPaul_The_Nerd: there are a couple of things on flyspray pertaining to recording that should be close (IMO)
01:28:33Davide-NYC5358 mark tracks while recording
01:28:51Davide-NYC5732 reduce ticking while recording
01:29:03Davide-NYCme thinks those be done
01:30:30Paul_The_NerdThey are
01:30:35Paul_The_NerdThey're both closed and "Fixed" now.
01:31:13 Quit mikearthur ("Konversation terminated!")
01:32:52 Part trougnouf
01:34:17midkayPaul_The_Nerd: btw - i saw your post on "badges - you may qualify"..
01:34:44midkayi think that would be cool if it were extended to have a section about each badge there currently is.. and how you can qualify for one..
01:34:55midkaylike, including developer/expert..
01:35:04midkaymaybe mod and admin as well..
01:35:11Paul_The_NerdI was actually planning on unstickying that one in about a week, and replacing it with an explanatory one.
01:35:41midkayi see.. the week being lack of time to write the explanatory one, or..?
01:35:57Paul_The_NerdThe week to give this one time to get read.
01:36:12midkaycouldn't you just keep the same information you have, and extend it to the other badges?
01:36:18Paul_The_NerdThe problem is the topic
01:36:36Paul_The_NerdIf I just titled it "Forum Badges" or something similar, people are less likely to read it expecting just explanation.
01:36:52midkaywhy not keep the current title then?
01:36:58midkayForum Badges - You may qualify?
01:37:21Paul_The_NerdI'd rather separate the discussion / application post from the explanations.
01:38:09*Paul_The_Nerd shrugs
01:38:19Paul_The_NerdMaybe I'll combine the two.
01:38:24midkaythat's what i'm suggesting.
01:38:58midkayhave a section for each badge.. like for example, dev: "Developer Badge - indicates a Rockbox developer with commit access. To qualify for this badge, you must have CVS commit access."
01:39:15Paul_The_NerdSeems fair enough
01:39:24Paul_The_NerdI'll update that post, probably sometime in the next 24 hours then.
01:39:52midkay"Artist badge - to qualify for this badge you must have created one completely original WPS or theme." etc etc.. maybe some more thorough descriptions/qualifications, but.. something like that. :)
01:41:00scorcheand priviledges? or is that better kept behind closed doors?
01:41:20Paul_The_NerdI think that I'll not be public about the fact that Developers and Experts get bonus priveleges.
01:41:32midkaythey're a bonus :)
01:41:48midkayyeah, i bet if you mention that people will apply like crazy..
01:41:55midkay"i know a lot about rockbox. can i have an expert badge?!?!"
01:42:03Paul_The_NerdIt's not a secret or anything, but I don't want people *trying* to get expert status or anything
01:42:11scorcheexcept, expert and developer badges are not able to applied for
01:42:35Paul_The_NerdYeah, but people will still beg for Expert, I'm sure
01:43:17scorchehaha...i wonder if there is a module for spam filtering in your mgith need it then =P
01:43:21***Saving seen data "./dancer.seen"
01:43:38 Join jaebird [0] (n=jae@
01:44:31Davide-NYCAnyone wanna take a stab at a tiny tiny feature request involving the simulator and a single pixel line?
01:45:24*Paul_The_Nerd is currently trying to get help fixing the "Sometimes replies start a new thread" issue we've developed recently
01:45:25amiconnWhy don't you just use rockboxui −−background?
01:45:50midkayPaul_The_Nerd: any idea where that's coming from?
01:46:18*Soap bets somewhere between keyboard and chair.
01:46:19Davide-NYCamiconn: what is that?
01:46:58amiconnIt shows the sim with a target image as background, with the lcd output places properly
01:47:11Paul_The_NerdSoap: There's really not a way to start a new topic that carries over the thread title, without retyping the whole title, so it's really likely it's an actual software bug of some sort
01:47:25Paul_The_Nerdmidkay: No clue, but I'm approaching Simple Machines' support forums for help
01:47:52midkayPaul_The_Nerd: hm, gotcha..
01:47:57Soapok, I'm cynical.
01:48:31Davide-NYCamiconn: holy crap! That's cool. Where should I have looked to know about this?
01:48:44Paul_The_NerdSoap: In most cases I would be. But it's happened 3 or 4 times now, and seems likely to be more than just user stupidity.
01:48:55*Davide-NYC documentation probablt
01:55:29Davide-NYCPaul_The_Nerd: Please destroy this with extreme predjudice.
02:01:41 Quit Rondom ("restarting X")
02:02:48 Join old45 [0] (
02:04:18 Part lorijho ("Kopete 0.11.3 :")
02:05:14old45I just did my first daily build install on my H 120, is there a way to ID the new build on my player? Wanting to know if I did it properly.
02:05:36Paul_The_NerdI don't know what you're asking.
02:06:23old45does the daily build have an ID tag of some sort that I can view to see if it installed?
02:06:50Mikachuthe date should be printed under the logo when booting
02:06:59Paul_The_NerdIt's also in the Rockbox Info thing in the Info menu, right?
02:07:03old45OKO I will check
02:07:07scorcheit is
02:07:08Mikachuyeah that should be the same screen
02:09:56old45Ok I found it and it did install
02:10:11 Quit Kitar|st (Connection timed out)
02:10:34 Join Kitar|st [0] (
02:10:57old45I am interested in the recording feature and see that there is a patch for enhanced recording, anyone use the patch?
02:14:27Paul_The_NerdI believe most of the recording enhancements are now in the official build, with only one or two left.
02:16:15old45that is good news
02:16:51old45Official build would be the most current daily build, correct?
02:17:35Mikachucvs builds are newer than daily builds
02:17:51 Join Rondom [0] (
02:19:19 Quit dwihno (Read error: 104 (Connection reset by peer))
02:20:45 Quit old45 ("CGI:IRC")
02:24:03 Join dwihno [0] (n=dw@
02:25:59 Join [HO]vo|t [0] (
02:39:21 Quit lodesi_ ("leaving")
02:58:39 Quit EbErT (Read error: 104 (Connection reset by peer))
02:58:43 Join EbErT_ [0] (
02:59:10 Quit EbErT_ (Client Quit)
03:03:10 Quit ShadowdogMU ("You know you'll miss me :P")
03:03:48 Join ShadowdogMU [0] (
03:05:02 Join JdGordon__ [0] (
03:06:07 Part pixelma
03:06:17 Quit Rondom ("Ex-Chat")
03:23:01 Quit JdGordon__ ("CGI:IRC")
03:27:18 Quit dwihno (Read error: 104 (Connection reset by peer))
03:27:27 Join dwihno [0] (n=dw@
03:30:04 Quit XavierGr ()
03:30:47 Quit Rudy4Pez (Read error: 110 (Connection timed out))
03:39:44 Join Febs [0] (n=medifebb@
03:43:24***Saving seen data "./dancer.seen"
03:45:35 Join old45 [0] (
03:46:28 Quit Pyromancer ("Leaving")
03:47:03 Join Pyromancer [0] (
03:48:19 Quit Pyromancer (Client Quit)
03:49:01old45I would like to get a handle on the build files for rockbox. The daily build includes the previous days CVS files? What is the point of a patch? I am refering to the enhanced recording patch
03:49:39Paul_The_Nerdold45: You already posted in the forums, and I answered there as well.
03:50:03Paul_The_Nerdold45: The daily build is a CVS build, but made once per day, whereas CVS builds are created every time Rockbox is changed by a developer.
03:50:12old45I still don't see the point of the patch if the files are in the daily build
03:50:17Paul_The_NerdWhat patch?
03:50:38Paul_The_NerdAs I've told you, a few of the featured of the Recording Enhancements patch are not in CVS yet.
03:50:43Mikachuthe point of a patch is to discuss a change before putting it in cvs
03:50:56old45they still have an enhanced recording patch link in that forum
03:51:28Paul_The_NerdIf you read the thread that goes with it, you can see the discussion of what features have made it into CVS and which haven't.
03:51:34Paul_The_NerdPatches become out of date frequently
03:51:58old45OK I see now. Thanks for the explanation
03:52:21old45looks like only two features to add yet
03:55:47 Join Pyromancer [0] (
03:56:27 Quit old45 ("CGI:IRC (EOF)")
04:06:53 Quit sharpe ("Leaving")
04:06:53 Quit dwihno (Read error: 104 (Connection reset by peer))
04:07:58 Quit CriamosAndy ("( :: NoNameScript 4.02 :: )")
04:11:26 Join dwihno [0] (n=dw@
04:11:30 Quit mirak (Remote closed the connection)
04:27:57rotatormmm, beer
04:28:21rotatortime for a buzzed session of coding...
04:31:11fatherforkhave fun.
04:32:08rotatori will :)
04:44:27 Join TCK [0] (
04:46:27 Quit fatherfork ()
04:51:08 Join Falco98 [0] (
04:51:51Falco98hey all
04:59:36 Join BHSPitLappy [0] (
05:02:09FebsI swear I'm going to start just deleting forum posts that don't contain ANY punctuation.
05:03:00Paul_The_NerdThere is one period in it.
05:03:28Falco98how old, by the way, does a forum post have to be before it expires?
05:03:44Paul_The_NerdThey don't, really
05:03:47Falco98i'm trying to find a few of my posts from a few months ago and they're not there, it seems
05:03:56Paul_The_NerdA lot of old posts are "hidden" right now because we haven't moved them over to the new layout.
05:04:03Falco98i search my nickname and find only 2 threads i'd replied in
05:04:18Paul_The_NerdIf you can give me the "Topic" or something by which I can find 'em, I can move 'em over for you
05:04:21FebsThe period in .rockbox doesn't count as punctuation.
05:04:32Falco98"insert random folders"
05:04:45Falco98"track skip and shuffle" (referring to crossfade)
05:05:10Falco98both talking about patches written by jdgordon and myself (respectively)
05:05:59Falco98BTW... now that the feature freeze is lifted, is there any chance the "shuffle + trackskip" patch could be pushed into CVS? it's rather tiny
05:05:59Paul_The_NerdThey should both be in Audio Playback now
05:08:05Paul_The_NerdFebs: But it verifies that he knows the key exists, and that it works.
05:09:21*Febs tries to stop laughing long enough to type a reply.
05:10:28FebsThanks, Paul_The_Nerd, that comment made my night.
05:10:34Paul_The_NerdIt's tempting to add ONE last member group, called "Clueless" or something, for people who show an utter disregard for the guidelines. ;)
05:10:54Paul_The_NerdUnfortunately, I fear people would want to get in it, just to have a badge
05:13:05rotatorheh, the ipods do seem to bring in the younger and less computer savvy group
05:14:01Falco98ironic, too, that they're the ones requiring more BS to actually get rockbox onto..
05:14:22Paul_The_NerdFalco98: In regards to the Insert Random Folders, JdGordon actually asked about it today, and the core devs seemed uncertain about whether it was worth having in CVS
05:14:52 Join midkay_ [0] (
05:15:16Falco98how about shuffle + track skip?
05:15:28Paul_The_NerdI haven't heard anyone bring that up
05:15:42Falco98do you know which function i'm referring to?
05:16:18Paul_The_NerdOn crossfade, there's a choice "while shuffled" and a choice for "On track skip" but not one for both, correct?
05:16:35Falco98so a few months ago i wrote a patch adding that option
05:16:55Falco98relatively simple.. maybe fewer than 5 new lines of code added, fewer than 10 changed (maybe.. haven't counted)
05:17:33 Join midk [0] (
05:21:31Falco98anyway, i gotta run.. but if you'd be willing to look at that and push it towards inclusion i'd be appreciative (i'm not sure but i suspect it may need to be re-synched first, if so i think it would be minor)
05:22:32 Quit Falco98 ("CGI:IRC")
05:23:59 Quit Killer927 (Read error: 113 (No route to host))
05:32:11 Quit midkay (Read error: 110 (Connection timed out))
05:33:34 Quit midkay_ (Connection timed out)
05:43:27***Saving seen data "./dancer.seen"
05:56:43Davide-NYCquestions about windows: is there a way to hav a single directoy show up in all of my archos folders?
05:57:41Davide-NYCso I can point a shortcut to my music directory and all of my music with be in the simulators without having to actually copy it into the archos folder?
05:58:31Davide-NYCI htink unix can do this, I want to mount a link to a directory that resides elsewhere on my harddrive.
05:58:36Davide-NYCcanbe done?
06:05:25dpassen1Using latest CVS on an H140, scrolling through folders while holding play to quickly scroll does not stop at the end like normal scrolling does.
06:08:04rotatordpassen1: no doubt a result of the recent button handling overhaul
06:08:13rotatoryou might want to report it in the forums:
06:08:20dpassen1Just did :-)
06:08:57Paul_The_NerdDavide-NYC: There is a way to do symbolic links in Windows NT4.0 and later I believe, but I don't know what it is.
06:09:11*Paul_The_Nerd doesn't know if this is exactly what you want or not, though
06:09:18Davide-NYCthank you, I needed the terminology to search for
06:10:29Paul_The_NerdBe careful with it. I remember one of the ways of doing (or simulating) the behaviour could result in you deleting your original files when you tried to delete the links.
06:10:42Paul_The_NerdBut I think that was a separate thing.
06:10:43rotatori seem to recall there was a way to kinda do symlinks under NTFS, but it wasn't really much more than a poorly supported hack
06:12:21Davide-NYCI'm using a single test repository of mp3s. Not my actual gold standard.
06:13:41 Quit jhMikeS ("Client exited")
06:14:07rotatorah, it was ntfs "reparse points" I was think of
06:16:14 Quit pike^laptop (Read error: 110 (Connection timed out))
06:16:14Davide-NYChere's a noce article on the subject if anyone cares.
06:17:01 Join perplexity [0] (
06:23:06 Join aliask [0] (
06:41:26 Quit rotator ("zzzzzzzzzz")
06:41:43 Join webguest72 [0] (
06:42:49webguest72how do i copy mp3s to my ipod in disk mode and still be able to play them with apple firmware?
06:43:00JdGordonu cant
06:43:06 Join Rudy4Pez [0] (
06:44:06 Nick midk is now known as midkay (
06:45:57webguest72was trying to get my sister to use linux and i didnt realize that you couldnt just copy it onto the disk and play it
06:46:12webguest72with apple firmware
06:46:21JdGordonthen get her to use rockbox... duh!
06:46:27webguest72i know
06:47:09webguest72but she doesnt find linux easy
06:47:21webguest72and she might kill me if i switched her to rockbox
06:47:55webguest72cause on my ipod 4th gen grayscale rockbox is still pertty buggy
06:48:23Paul_The_NerdWell, this isn't a general iPod support channel.
06:48:27webguest72reboots itself every so often playing oggs
06:48:36Paul_The_NerdSo it's kinda the wrong place to be asking questions about the Apple firmware
06:49:09webguest72well since someone had to figure out how to dual boot the two, thought someone might know a little about how it worked
06:49:47JdGordonhey Paul_The_Nerd, do u know why there is so much opposition to "option bloat"? is it because the menu's are a mess anyway? or code size/complexity?
06:50:17aliaskI think code size is a big issue.
06:50:24aliaskWell... binary size.
06:50:30midkayon the archoses..
06:51:07JdGordon... because i have an absolutly nuts idea to probably halve code size and make the menus much easier to add...
06:51:17webguest72do midis still pay at double speed rockbox ipod ?
06:51:21aliaskPlease share
06:51:58Paul_The_NerdJdGordon: It's almost wholly code size I think.
06:52:05JdGordonaliask: im still trying to figure the idea out in my head
06:52:29Paul_The_Nerdwebguest72: Yes, midis still aren't officially supported.
06:52:34JdGordonPaul_The_Nerd: awesome... ill bviously double check that with the swedes,, but if it is only code size then i tihnk this could work :D
06:53:19webguest72what about nsf files?
06:54:56Paul_The_Nerdwebguest72: What about them?
06:55:22webguest72is there any development for playback in rockbox?
06:55:36webguest72or are they alraedy supported?
06:55:53Paul_The_Nerdwebguest72: I'd suggest skimming the manual, as it covers the various currently supported features.
06:56:16webguest72i did
06:56:46webguest72i didnt see it so i was wondering if it was in development
06:56:53Paul_The_NerdJdGordon: There's two types of negative option bloat. One is the filesize, which I think is the primary objection to that shuffle. The other is the "support nightmare" variety such as customizable windows that, while possibly fun, open up all sorts of potential for user mayhem as well as making the code harder to maintain and localization / blind user support more difficult.
06:57:24JdGordonPaul_The_Nerd: ok
06:57:41midkayJdGordon: you better not be talking about offloading menus into a plugin. ;)
06:59:21webguest72i think the development of doom for ipod 4g grayscale should stop,,,it cant possibly amount to anything of interest on that platform...unless there is reprogramming like there was for bubbles that made it work well in grayscale
06:59:42JdGordonthe generel gist of my idea involves the menu layout being converted from a text file to a c array by a perl script before compiling, so users couldnt actually change the layout in official buids, but unoficial ones could. the script would obviously have to manage the translation support (which should be easy), and the code size would shrink (assuming the function code takes more room in the binary the const'ed arrays)
06:59:59JdGordonmidkay: doing that would shrinn about 15K i tinhk (i checked it a while ago)
07:00:17Paul_The_Nerdwebguest72: One of the goals of Rockbox is to have as close to identical features on all platforms. Doom can be moderately playable on a grayscale target, so why not?
07:00:56JdGordondoes anyone think that would almost be workable?
07:01:13*JdGordon should be more patient and start discussing this in 5 hours when ppl are awake :p
07:01:23webguest72moderately playable? youve actually palyed it on an ipod 4g greyscale?
07:01:47webguest72it might be able to play demos
07:02:06Paul_The_Nerdwebguest72: I've played it on an H120, which has a screen with similar properties.
07:02:07webguest72and you might be able to see it if you can hold the right angle long enough
07:02:39Paul_The_Nerdwebguest72: I assume you have adjusted the gamma?
07:02:49Paul_The_NerdIt's not like there's "work" going into Doom these days anyway
07:02:57webguest72there was no gamma adjustment when i adjusted it
07:03:02Davide-NYCPaul_The_Nerd: I want to open a discussion about remote support in the simulator and swapping out bitmaps of the different possible remotes.
07:03:04webguest72it stayed the same
07:03:31Paul_The_NerdDavide-NYC: General actually, since it's relating to the simulator, not the actual hardware.
07:03:52Paul_The_NerdDavide-NYC: There's not really a good place for "Simulator" discussion, but there's also *very* few topics that only affect the sim.
07:03:56JdGordonDavide-NYC: thats proabbly best on the dev mailing list
07:04:15JdGordonmidkay: aliask: what you guys think? ^
07:04:20webguest72i think the battery life has been improving with almost every new release though
07:04:22Paul_The_Nerdwebguest72: Hrm, well that would probably make it better.
07:04:27Davide-NYCJdGordon: but I just made these sweet mockup jpgs and I want the rockbox world to see them!
07:04:42JdGordonlink them in the email?
07:04:43midkayJdGordon: have you tried it at all? no idea how much that'd benefit code size...
07:05:14webguest72the ipod gets really hot at times...i wonder if it will blow up
07:05:17aliaskJdGordon: It sounds like a good idea, but yeah, I don't see how it reduces binary size.
07:05:18JdGordonmidkay: no, i just thought of the idea on my way home from uni... it would remove heaps of "duplicated" code... all of *_menu.c would be removed
07:05:42Paul_The_NerdJdGordon: But the thing is, how would it affect compiled size?
07:05:46aliaskJdGordon: So it just decreases the size of the C code in CVS, not the rockbox.____ file?
07:05:49midkayJdGordon: i see.. hm...
07:05:59midkaywell, that should in turn affect the rockbox binary..
07:06:04midkayhow much, i have no idea.
07:06:10JdGordonPaul_The_Nerd: thats the thing.. i dont know how it would compare.. i would assume it would be smaller.. but i have no diea
07:08:38Davide-NYCJdGordon: pardon my naivete but how do I email the developer mailing list?
07:08:58JdGordonumm... there is a link on the left of the rockbox page to the lists
07:09:38webguest72do you have to reboot rockbox to exit a doom level?
07:09:59Paul_The_NerdJdGordon: Wouldn't an easier feature than "Shuffle Albums" be to simply move to a random folder when playing in non-playlist mode, as an alternative to next-folder?
07:10:14JdGordonive done that patch also
07:10:27JdGordonbut thats the one which has a 20+s lag time on the first go
07:11:30Davide-NYCI was subscribed to the cvs list, which I've discovered is different from the dev list.
07:12:30*JdGordon rekons this could work
07:14:54 Quit midgey34 (Read error: 110 (Connection timed out))
07:15:22Paul_The_NerdJdGordon: I think that when re-pitching the "Shuffle Folders" idea, you should mention there's no other way to duplicate this behaviour in CVS Rockbox.
07:16:02JdGordonna, i'll just wait untill there is heaps of commit happening and just sneak it in without them realising :D
07:16:13 Quit Davide-NYC ("Chatzilla 0.9.75 [Firefox]")
07:16:33Paul_The_NerdI'm sure somebody would notice the addition of the feature ;)
07:17:20Paul_The_NerdI think it is a good idea though. It's essentially treating your device as a multi-disk changer, where the disk is chosen randomly, rather than the track.
07:18:39JdGordonexcept the delay is really annoying, but i dont think there is any way around it
07:21:01Paul_The_NerdWell, if playlists could contain playlists, that would be another option, but that unleashes all kinds of havok. :-P
07:26:00 Quit bb-generation (K-lined)
07:26:19BigMacthere any work on that model yet?
07:27:42Paul_The_NerdNone at all.
07:28:16BigMacwell then i guess i will have to reencode that stuff
07:29:54LandusI'm using an iRiver h3x0 and haven't recompiled since the 13th of last month.
07:30:02LandusHas there been any big changes?
07:30:12Paul_The_NerdThere's a changelog.
07:31:11LandusNot seeing it.
07:31:53Paul_The_NerdFront page of
07:34:35LandusThe CVS activity?
07:35:20LandusThat's recent. It's not going to show changes from 4+ weeks ago.
07:35:30Paul_The_NerdNotice the link that says "Changes since 2.5" at the top?
07:35:35Paul_The_NerdThere's over a year of changes available.
07:35:39Paul_The_NerdWhich is well over "4+ weeks"
07:35:48Paul_The_NerdSorry "all commits since 2.5"
07:41:02 Quit Febs (Read error: 110 (Connection timed out))
07:43:28***Saving seen data "./dancer.seen"
07:43:59 Join midkay_ [0] (
07:46:17 Join dwihno_ [0] (n=dw@
07:46:17 Quit dwihno (Read error: 54 (Connection reset by peer))
07:48:27Bg3ri.e. morning
07:48:38JdGordonhehe mornign Bg3r
07:48:59Bg3ri'm still sleepy
07:51:16 Join dwihno [0] (n=dw@
07:51:16 Quit dwihno_ (Read error: 104 (Connection reset by peer))
07:56:17 Join dwihno_ [0] (n=dw@
07:56:18 Quit dwihno (Read error: 104 (Connection reset by peer))
07:57:21scorchei am starting to think we need a basic english test before we let people post on the forums...
07:57:35 Join theli_ua [0] (
07:57:50 Quit dwihno_ (Read error: 104 (Connection reset by peer))
07:58:20Paul_The_NerdIt would be unfair to those for whom English really is a second language
07:58:33JdGordonliek mee
07:58:38scorchebut, you get my point
07:59:46Paul_The_NerdThough even people who don't speak English natively should be able to at least achieve proper spelling.
08:00:10Paul_The_NerdSeeing as that's something that you can actually *check* whereas grammar requires some context and usually human help if you're uncertain.
08:00:39scorchethose are excusable though....what worries me are the people that are from the US and
08:01:31scorchemakes me worry about the state of the language in 50/100 years
08:02:00Paul_The_NerdThe people who feel that their key movements are so valuable, to waste even one extra letter on a punctuation mark or vowel in a word that can be shortened to three letters if they're removed, would be a sin?
08:02:16 Quit midkay (Read error: 110 (Connection timed out))
08:02:19 Join dwihno [0] (n=dw@
08:02:23perplexityIt's ok, the rest of the english speaking world are holding the fort.. We just won't be able to converse with people in the US ;)
08:02:43perplexitys/are/is/ :\
08:02:46scorcheamong others..
08:03:19JdGordondoes anyone know how to remove a file from the play queue in JuK?
08:06:30 Join Bagder [0] (
08:06:58 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
08:09:09 Quit dongs (Read error: 104 (Connection reset by peer))
08:12:54 Join dongs [0] (
08:20:34 Join mantono [0] (
08:22:09 Quit dongs (Remote closed the connection)
08:22:12 Join dongs [0] (
08:26:09 Quit perplexity (Read error: 113 (No route to host))
08:26:25 Join perplexity [0] (
08:34:25 Nick midkay_ is now known as midkay (
08:36:43 Join ender` [0] (i=null@
08:41:49 Quit dongs (Remote closed the connection)
08:41:56 Join dongs [0] (
08:52:07 Quit dongs (Remote closed the connection)
08:53:37 Quit mantono ("Don't worry, be happy")
08:57:19 Join _FireFly_ [0] (
08:57:33 Join dongs [0] (
09:05:37*ender` yawns
09:07:46 Join dongs_ [0] (
09:09:02 Join dan_a [0] (n=dan-mirc@
09:13:06 Join einhirn [0] (
09:13:06 Quit einhirn (Client Quit)
09:14:17 Quit dongs (Read error: 104 (Connection reset by peer))
09:30:24 Quit twisted` (Read error: 110 (Connection timed out))
09:40:45 Join Stas_ [0] (
09:43:31***Saving seen data "./dancer.seen"
09:44:15 Join daurnimator [0] (n=quae@unaffiliated/daurnimator)
09:51:49 Join t0mas [0] (n=Tomas@rockbox/developer/t0mas)
09:57:46midkayyour mother.
10:01:06 Join pondlife [0] (
10:02:19 Quit dan_a (Read error: 110 (Connection timed out))
10:04:21 Join barrywardell [0] (
10:05:41aliaskEhm, in (note: no WebHome on the end), the 10 most recent changes are a bit dated... what's the deal with that?
10:06:57linuxstbClick on the "Refresh Cache" link at the bottom.
10:12:19theli_uado rb->pcm_play_data waits for data to be played or just sumbmits buffer for playing?
10:13:14pondlifeMorning all! Anyone fancy committing - as I now have confirmation that it fixes the crashing problems for 2 users on the ML.
10:16:49LinusNpondlife: has anyone tested it on an archos?
10:18:03pondlifeNo, but it shouldn't affect Archos; it mainly reverts the earlier changes in that area. If you could test it that would be nice though.
10:18:12pondlifeIt does compile this time
10:18:39pondlifePlayback.c is not used on MASCODEC, right?
10:18:52pondlifeOK, so there are no Archos mods in there
10:19:09pondlifeThe talk.c changes are all in SWCODEC #if blocks
10:19:23pondlifeAside from some comments
10:20:14pondlifeIt's definitely better than current CVS too.
10:21:55LinusNthank *you*
10:22:06pondlifeAlso.... how about that buffer_alloc mod for tagtree...
10:22:46 Quit BHSPitLappy ("Leaving")
10:23:11pondlifeI'd like to look into the year problems in tagcache, but would rather get one mod committed first so I don't end up with any "cross-purpose" patches
10:24:37 Join Lear [0] (
10:30:27 Quit dongs_ (Read error: 110 (Connection timed out))
10:33:58 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
10:38:29Learjdgordon: you around?
10:38:35JdGordoni r
10:39:21LearJust wondering about bookmark selection screen... ACTION_BMS_SELECT and _EXIT aren't in any keymap definition. Should they be?
10:39:49LinusNthe bookmark selection screen sucks imnsho
10:40:11JdGordon... but untill its redone it shuold be fixed.. ill go fix the mappings now
10:41:20Learbtw, linusn, you were interested in the "dynamic enum" part of patch 4802?
10:41:41LearI've made an (simple) implementation of that...
10:41:53LinusNme wants it....preciousssssssss
10:42:07pondlifeMaybe one for Bagder? Why does the CVS log display "barrywardell" rather than "Barry Wardell" when everyone else is preoper cased...? Not very important, but these things annoy me...
10:42:48LearWhat I have now is almost backwards compatible; only thing is I suspect most WPS:s don't include a "unknown battery level" bitmap. They'd need one with it.
10:42:48linuxstbpondlife: Just check out another copy of the code from CVS if you want to work on something unrelated to your other work - I normally have a few local copies at any one time, each with different sets of changes.
10:42:52Bagderpondlife: fixed now!
10:43:04pondlifeThanks and thanks
10:43:20LinusNBagder: you beat me to it
10:43:23pondlifelinuxstb: You mean parallel to rockbox-devel I take it.
10:43:27BagderI won!
10:43:34*Bagder cheers and runs around his room
10:43:52pondlife...scaring his co-workers somewhat
10:44:02BagderI have no co-workers!
10:44:03barrywardelli hadn't even noticed!
10:44:08pondlifeLucky you
10:44:13Bagderonly machines
10:44:14 Join Kohlrabi [0] (
10:44:14linuxstbpondlife: Yes. I normally do something like "mkdir tmp ; cd tmp ; cvs co -d..... rockbox-devel ; mv rockbox-devel ../rockbox-myproject ; cd .. ; rmdir tmp"
10:44:29Lear(Not a big issue; the shown battery level would be a bit lower than it should be.)
10:44:29pondlifeAh, ok
10:44:32linuxstb(i.e. you can rename rockbox-devel to anything)
10:44:59Bagdercvs can even check out to a different name
10:45:03linuxstbYou can even just do "cp -pr rockbox-devel rockbox-myproject" if you want.
10:45:10pondlifeEven easier
10:45:19linuxstbBagder: How?
10:45:30Bagdercvs co blabla -d anotherblabla
10:46:07linuxstbAh, I love the way cvs gives the same option different meanings depending on where in the command-line it is...
10:46:15Bagderoh yes, highly confusing
10:46:32Bagderand depending on what subcommand you use
10:47:02*tucoz waits for preglow to mention svn
10:47:08linuxstbAnd it makes searching the man page nice and simple as well...
10:47:19*Bagder resolves
10:47:26pondlifeBagder: I assume Barry Wardell will appear on Barry's next commit... I still see lower case on the frontpage...
10:47:44Bagderat next table update it'll fix itself
10:48:12tucozBagder, really? so it's somewhere in the not so distant future?
10:48:20pondlifeRight... onto something more important....
10:48:22 Join MarcoPolo [0] (
10:48:44Bagdertucoz: hopefully, yes
10:49:21Bagdertucoz: you can in fact checkout from svn://
10:49:31Bagderbut its not updated
10:50:02tucozoh. I've heard so much nice about svn, that it's almost tempting to try it out.
10:50:21Learlinusn: maybe I should make the first battery enum the "unknown level" value, rather than the last?
10:51:34LinusNyes i think so
10:55:47tucozanyone here that uses the gnome terminal? In that case, do you know what the standard font is called?
10:56:24JdGordonLear: ive just commited the fix for the bookmar screen buttons.. 15min and it will be on cvs
10:56:59LearGreat, thanks.
10:58:19tucoznevermind. It's probably monospace.
10:59:00JdGordonguys, RE the option rework email (which seems to have gone off track).. is it worth the effort? do i try it and see how ti goes?
11:00:52JdGordonactually ignore that q.. ill reply to the email
11:01:26*amiconn seems to no longer understand his own pixel value range mapping considerations properly :/
11:01:35JdGordonhehe welll done
11:01:44JdGordondont you wish you commented? :D
11:01:48 Join midkay_ [0] (
11:01:52JdGordon... the code that is
11:03:44amiconnThe code is simple, no comments needed on that one
11:03:58Paul_The_NerdJdGordon: I think the consensus on that was "Reworking the way it works is good. Offloading it to an externally parsed file in XML, not so great."
11:04:01 Join wehn [0] (
11:04:38Bagderthat's my feel too
11:06:38 Join Arathis [0] (
11:07:43 Join KN|stiff [0] (
11:09:16pondlifeJdGordon: It would be good if the rework could allow users to customise menus in some way... is that under consideration at all?
11:09:32Bagderit usually pops up every now and then
11:10:30amiconnImho the main menu should not be configurable
11:11:10Paul_The_NerdI think the customization should be a compiling step, if anything.
11:11:14pondlifeI was thinking this could solve the call for "simpler" menus perhaps.
11:11:21amiconnInstead, I would deem it a good idea to replace the quickscreen(s) with a user configurable menu
11:11:28 Join bb-generation [0] (
11:11:30pondlifePersonally I am very strongly against any need to hide options or use a CFG editor
11:11:32 Join dan_a [0] (
11:12:18tucozI agree. Instead a reorganization is on the other a better solution.
11:12:21linuxstbMy view is that there should be a standard, fixed menu where everything is accessible - so if someone asks how to do something, we can tell them. But (as amiconn just said...) there could be an additional user-configurable menu, or maybe users could add their favourite options to the root of the main menu.
11:12:30amiconnThe user configurable menu could double as a root menu, with a sensible default configuration
11:12:35Paul_The_NerdAny thoughts on my context-menu post to the thread on the list?
11:12:50*linuxstb thinks the same as amiconn
11:12:50Paul_The_NerdReconsidering the use of the context menu somewhat?
11:13:54pondlifeamiconn: Why not allow all menus to be customisable ,as long as there's a way to restore defaults at all time
11:13:57amiconnpondlife: The user configurable menus wouldn't require to edit the .cfg by hand
11:14:20amiconnpondlife: How do you restore the default when you've hidden the menu item to do so?
11:14:20pondlifeNope, I'm referring to the idea on the ML that some options become accessibly only by editing CFG files
11:14:47 Quit midkay (Read error: 110 (Connection timed out))
11:15:18amiconnAlso, having the main menu configurable will lead to worse accessibility by the blind
11:15:27pondlifeamiconn: Reset settings at startup? Or similar..
11:15:45amiconnImagine a misconfigured menu with the voice menus switched off.
11:15:59amiconnHow would a blind user find the menu item to reenable voice?
11:16:14Paul_The_NerdI think the main menu should be static, contain everything user-configurable somewhere within, and should be avoidable in almost all cases.
11:16:18pondlifeWhy would a blind user go to the effort of customising their menus though?
11:16:25LinusNimagine all the "my menu if b0rked" forum posts we will receive
11:16:27pondlifeI mean removing the voice options
11:16:47amiconnpondlife: Maybe others use his unit as well? Or there is a glitch in rockbox?
11:17:04pondlifeI'm assuming that editing a menu will involve some text file editing at least, not something that can be done easily or mistakenly.
11:17:06Paul_The_NerdIt seems to me that since we have a context menu, the main menu should primarily be used for set-up, while the context menu should contain what a user needs at any point while they're using the player (assuming they don't want to change settings for another context, that is)
11:17:23LinusNfor example if we add a few menu items, and the user doesn't upgrade his menu config
11:17:31pondlifePersonally I just want to see the menu include ALL options
11:17:37amiconnpondlife: I'm not talking about _removing_ the options, it's enough that the options are no longer at the same place in the tree because other options were hidden...
11:17:42LinusN"i don't see the new XXXX feature in the menu"
11:18:12pondlifeThe only reason I mentioned it was to see if it could be used to stop people suggesting that some options be removed from settings...
11:18:17LinusNconfigurable menus is a support nightmare
11:18:19pondlifeSo no argument from me
11:18:42pondlifeAs long as all options remain accessibly in the menus that is.
11:18:57LinusNto quote bagder: "you'll drag me kicking and screaming into that mess"
11:19:00Paul_The_NerdI don't think anyone in here thinks options should be removed from the menus, but stay in code.
11:19:33pondlifeNot here, but on the ML
11:20:00*JdGordon back
11:20:10pondlifeSo we all agree, all options remain, only "F3" menus to be customisable, if any.
11:20:34JdGordonre customizing the menu... we agree thats bad, but I tihnnk adding a user menu which the user can populate how he likes would be acceptable?
11:21:24Paul_The_NerdI still think a more intelligent context menu would help a lot.
11:21:47Paul_The_NerdWhy do people need a customizable menu?
11:21:52Paul_The_NerdIn the root, create a folder called "Menu"
11:21:58Paul_The_NerdInside, create folders named after each option you want.
11:22:08Paul_The_NerdIn those place .cfg files with single entries for the values you want to be able to change them to
11:22:15amiconnPaul_The_Nerd: That works... as long as you don't use tagcache.
11:22:19JdGordonhehe.. thats a solution
11:22:54Paul_The_Nerdamiconn: And a "Filetree" list entry in TagCache that takes you to File View, or the option to point at Folders with TagNavi.config bits would also allow this to be worked around without configurable menus. ;)
11:23:03amiconnAnd it means spinning the disk for every option change
11:23:10Paul_The_NerdAh, this is true.
11:23:29daurnimatorh iall
11:23:46Paul_The_NerdBut, either way, I'm surprised nobody has done this yet if they want configurable menus so badly.
11:23:51 Quit dan_a (Read error: 104 (Connection reset by peer))
11:24:18pondlifeNoone really strongly wants configurable menus AFAIK. Some people on the ML seem to think that there are too many options and it is confusing. Me, I like it how it is...!
11:24:18Paul_The_NerdMy thinking is that it's likely to result in a lot of people tinkering a little bit, one person coming up with a well liked layout for the configurable screen, and 90% of users using that one.
11:24:24 Join dan_a [0] (
11:25:39Lynx_how about having a scrolling line at the bottom of the screen that shows a short explanation for the currently selected option in the settings menu?
11:26:02Paul_The_NerdLynx_: I think the confusion is in finding things.
11:26:19Lynx_Paul_The_Nerd: hmm, right, it doesn't help there
11:26:44 Join PaulJam [0] (
11:26:55linuxstbIMO the current menu structure is confusing because it tries to meet two incompatible requirements - make commonly used features accessible easily, and have a logical, easy to navigate hierarchy.
11:27:19Lynx_maybe it would actually help to have options that do not clearly belong into any one category present in two or even more categories? that would increase the menu tree size, but still make stuff easier to find.
11:27:25Paul_The_Nerdlinuxstb: Again: Context menu for commonly used features in a given context, full menu for easy to navigate hierarchy of everything. :-P
11:27:25daurnimatori think there should be a universal OS for all mp3 players
11:27:32linuxstbPaul_The_Nerd: Exactly.
11:27:39scorcheBagder: something going on with builds?
11:27:45daurnimatorprogrammable in a script language - with ability to execute binaries
11:28:04Paul_The_Nerdlinuxstb: I think that the context menu is underappreciated by most people. I get the sense it's primarily used for playlist-insert, and file deletion.
11:28:05 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
11:28:25Bagderscorche: yeah
11:28:41Lynx_daurnimator: port a scripting language to rockbox, and you're almost there ;)
11:28:41daurnimatorie - you're given all functions (load files, play music etc) , and you make your own ui.
11:28:57 Join Kohlrabi [0] (
11:28:57daurnimatorand, a place where you can just select your hardwaare, and its all in there
11:29:02linuxstbPaul_The_Nerd: I also don't think it should be much more than that - keep the context menu for that purpose, and allow the user to create duplicates of their favourite menu options in the root of the main menu, or in a new root menu.
11:29:11daurnimator - like a dropdown list :P
11:29:53scorcheBagder: just making sure... =P
11:30:35Paul_The_Nerdlinuxstb: I think there's a lot missing from the context menu. For example, you can't change replaygain while music is playing, nor are some of the playback options (shuffle, crossfade) available from it, etc.
11:32:30Lynx_is there still a quick menu on the h300, reachable from the wps?
11:32:43Paul_The_Nerdlinuxstb: I think with that addition, the Quick-menu could probably be dropped on SWCodec players (since it still requires many button presses to cycle values anyway) and be replaced with the user configurable one.
11:32:45Lynx_like the archos f-key menu
11:33:24JdGordondoing this rework also means its likely we will be able to easily add the menu title :)
11:34:39Paul_The_NerdFor example, why isn't "File View" available when you hit the context menu in the filetree?
11:35:42 Join _hotwire_ [0] (
11:36:53 Quit wehn (Read error: 110 (Connection timed out))
11:37:07JdGordonwhats the rule about structs with unions/structs nested in them? is it ok?
11:37:09Lynx_how do i report typos in the manual? bugreport?
11:39:26_hotwire_amiconn: you around?
11:40:36tucozLynx_, a bugreport is a good way of reporting stuff like that
11:41:15tucozLynx_, have you found many typos?
11:41:16Lynx_tucoz: ok. it seemd a bit much to open a bug report for one typo ;)
11:41:26tucozLynx_, oh. tell me and I fix
11:41:31Lynx_tucoz: no, just one. maybe i should collect them.
11:41:45scorchePaul_The_Nerd: been watching mr. ritesh?
11:42:04 Join A_M [0] (
11:42:20tucozLynx_, I can fix that one right away :) (as long as you tell me where it is)
11:42:23Paul_The_Nerdscorche: Hm?
11:42:24Lynx_tucoz: section 3.4.3, 'shortchut' under sound settings should obviously be shortcut
11:42:39scorchePaul_The_Nerd: he posted a "cheat" for brickmainia....
11:42:56Paul_The_NerdI saw that, didn't pay attention to who posted it though
11:43:06Bg3rJdGordon: i don't see why it wouldn't be ok
11:43:20JdGordonthats what i like to hear :D
11:43:25scorcheand hw is already up to 51 posts....
11:43:32***Saving seen data "./dancer.seen"
11:43:50Bg3rJdGordon: what r u working on ?
11:44:08JdGordoncomplete rework of the menu system...
11:44:09Lynx_tucoz: section 3.4.2, 'rightend' after The Peak Indicator is missing a space ;)
11:44:15JdGordonread the ml
11:44:28 Join Genre9mp3 [0] (n=yngwiejo@
11:44:31scorchePaul_The_Nerd: 52 posts in about 3 days...
11:44:35 Join petur [0] (n=petur@rockbox/developer/petur)
11:44:51scorcheabout 2 days actually
11:45:23 Quit billytwowilly (Read error: 104 (Connection reset by peer))
11:45:59 Join billytwowilly [0] (
11:46:18tucozLynx_, fixed.
11:48:39barrywardellBagder: turns out we haven't got the h10 scrollpad solved yet :(
11:49:07barrywardellchannel 4 of the ADC is conneceted to the pad
11:49:23barrywardellbut it seems to give pressure sensitivity rather than position sensitivity
11:49:30 Join Mmmm [0] (
11:50:09barrywardellthe adc is a 'dual 4-channel'. does that mean that there could be another 4 channels yet to find?
11:51:04Bagdersounds like it could be
11:53:26barrywardelli guess i'll look at the OF to see if I can get any hints
11:59:23Genre9mp3Paul_The_Nerd: Nice work on the forums... I like the new categorization. I was away for some days and didn't check the forums for quite some time, so I had a total impact of the changes...
11:59:36*JdGordon thinks menu rework will be able to remove (or at least massivly increase) the amount of allowable submenus.... weather that is good or bad.. i dunno D
12:00:45 Quit PaulJam (".")
12:02:15A_MJdGordon: a q about your menu reworking for settings idea: there are plenty of settings that require additional code to be executed rather than just setting the global_settings member variable, how would your system with external .xml (or whatever) files deal with that?
12:03:12JdGordonA_M: check my reply about 5min ago.... the menu allows for functions to be called which can manually run the options gui, or do anything really
12:03:22amiconnImo xml is overkill in a system like rockbox
12:03:23Paul_The_NerdGenre9mp3: Well, most of the change happened overnight. Forums were down 3 hours, came back up looking like that. ;)
12:03:40JdGordonand there is no more external text file.. it was stupid and wrong :p
12:03:44Paul_The_NerdGenre9mp3: Surprisingly few people have complained about the lack of player specific forums, nicely.
12:03:49 Join AM [0] (
12:03:52AMJdGordon: a q about your menu reworking for settings idea: there are plenty of settings that require additional code to be executed rather than just setting the global_settings member variable, how would your system with external .xml (or whatever) files deal with that?
12:04:09 Quit A_M ("CGI:IRC (EOF)")
12:04:19 Join San||Away [0] (
12:04:51midkay_haha. good response.
12:04:53AMdid you just answer that or something? my IRC client sort of died so I may have missed something...
12:05:05JdGordongrr.. stupid pm wont go through
12:05:10JdGordon<JdGordon> A_M: check my reply about 5min ago.... the menu allows for functions to be called which can manually run the options gui, or do anything really
12:05:22tucozAM, check the logs
12:05:48JdGordonreply to the mailing list i mean...
12:07:00 Quit nudelyn ("At Argon, we're working to keep your money.")
12:07:06AMah I see
12:07:54 Nick midkay_ is now known as midkay (
12:09:28Genre9mp3Paul_The_Nerd: I won't miss the target specific forums. It's better this way now IMO. Besides we have "real" topic specific forums now. :)
12:10:03 Join RoC_MM [0] (
12:10:06Genre9mp3Paul_The_Nerd: So the forum rework is completed now? or there is work to be done?
12:11:07Paul_The_NerdGenre9mp3: There's still a lot of posts to be moved over. We moved at best, maybe the newest 10%. Not that people really miss them.
12:11:11 Join pi [0] (
12:11:23Genre9mp3Paul_The_Nerd: I see...
12:11:27Paul_The_NerdOther than that, I'm trying to get some bugs we're experiencing resolved.
12:12:00Paul_The_NerdAnd then maybe I'll look into upgrading the functionality a little bit.
12:12:50Paul_The_NerdOh, right, and in a few days I'm getting rid of post-based member groups. Posts will still be counted, but everyone will simply be "Member" if they aren't in one of the assigned groups
12:13:34 Quit barrywardell ()
12:14:51Genre9mp3Paul_The_Nerd: Nice! This means that I won't be a newbie anymore! :P
12:15:10AMJdGordon: Have you looked at patch #5833 or otherwise considered what would be required to allow displaying the setting values in the menus? When I wrote that I didn't bother addressing the current menu bloat (actually I tried to minimize need to rework existing code so it sort of increases bloat instead), but I think it ought to work well in conjunction with a menu system such as yours...
12:16:02 Quit daurnimator (Connection timed out)
12:16:17JdGordonI havnt, but depending how i implement the changes, there might be very little work needed (if any) for that to work
12:16:35*amiconn wouldn't want the settings displayed in the menu
12:16:58amiconnAt least not all settings in a 2-column fashion
12:16:59aliaskI quite like the patch, on large lcd's at least.
12:17:00Paul_The_NerdGenre9mp3: Haven't you written patches that have been accepted, or am I confused with who you are now?
12:17:11Paul_The_NerdIt is 5:20 am, and my brain is not working right
12:17:37*JdGordon bbl
12:17:55Genre9mp3Paul_The_Nerd: No not patches...just the iCatcher wps
12:17:55amiconnLooks like crap on low-res LCDs and/or with large fonts
12:18:23 Join dionoea_work [0] (
12:18:33markunGenre9mp3: do you still have the gigabeat iCatcher somewhere?
12:18:37dionoea_workwould it be possible to have a "Last changed" date column in flyspray ?
12:18:43 Join pixelma [0] (i=pixelma@
12:18:52firenx how do i edit my boot loader sequence
12:19:24Genre9mp3markun: yes
12:19:39AMthe way the patch works it's a setting so amiconn would have to have it enabled anyhow. (plus there's always the option to display settings on two lines, or only display settings for the selected item)
12:20:01Paul_The_NerdGenre9mp3: Ah. You already deserve a group in the forums then. You're an "Artist"
12:20:04 Join MadDog011 [0] (
12:20:06Paul_The_NerdGenre9mp3: I knew you'd done *something*
12:20:09 Part MadDog011
12:20:17 Join dongs [0] (
12:20:45Genre9mp3Paul_The_Nerd: There's an artist group? Didn't know that....
12:21:40AM*err wouldn't
12:22:33Genre9mp3Paul_The_Nerd: Nevermind. Just saw the badges thread
12:22:34dongsholy fuck this internet is fucking garbage
12:22:43Paul_The_NerdGenre9mp3: I'm rewriting the badges thread right now actually. :)
12:23:53Genre9mp3markun: I 'll have to make a UniCatcher for Gigabeat, too
12:24:41Genre9mp3...and of course a version for X5 remote..
12:24:53 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
12:26:32Genre9mp3markun: So, what's the progress status for the Gigabeat, now? (sorry I was away these days)
12:27:14_hotwire_amiconn: did you get my forum pm?
12:27:25markunGenre9mp3: still working on the bootloader, but we are also working on the drivers
12:27:47markunYesterday we wrote to the DAC for the first time.
12:29:02markunI wrote to the company who made the LCD module about the LCD driver IC. They told me it was a custom job so they can't give it to me.
12:29:07 Join hotwire_ [0] (
12:31:04Genre9mp3That's bad... (for the LCD)
12:32:29Genre9mp3So after the bootloader is done you won't be able to enable the display on target?
12:32:42markunwe already have the display working
12:32:44 Quit MarcoPolo (Read error: 54 (Connection reset by peer))
12:32:58markunbut it's just nice to have for contrast settings, screen flipping etc
12:33:07Genre9mp3I see...
12:33:21*Paul_The_Nerd finishes his "Badges" post update.
12:33:51Genre9mp3markun: :)
12:34:25midkayPaul_The_Nerd: YAY!
12:35:16midkayPaul_The_Nerd: the "do not post in these" forums, are you working them down to 0 posts and then deleting them?
12:35:59Genre9mp3Paul_The_Nerd: How about multiple badges? You deserve the "expert" one as well!
12:36:19Paul_The_Nerdmidkay: Yes, slowly.
12:36:40midkayPaul_The_Nerd: gotcha. nice job on the forum post - reading it now, very well-written so far :)
12:37:07tucozanyone with a ipod photo/4g around?
12:37:16tucozI wonder where the hold switch is located
12:37:18Paul_The_NerdGenre9mp3: I could theoretically do that, but I'm happy the way it is. You're an artist, btw, if you hadn't noticed.
12:38:39Genre9mp3Paul_The_Nerd: Oh...what an honor! :)
12:38:50midkayPaul_The_Nerd: small suggestion - you made each "Staff" and "Bestowed Badges" sections in bold, and the third section is more of a non-bold description ("You may apply for these badges") - maybe you should bold this and cut it down to like "Badges you may apply for".. :)
12:39:04Genre9mp3Though "artist" is very strong word...heh
12:39:12 Quit AM ("CGI:IRC (Ping timeout)")
12:39:17pihmmmw i realy ott to join the forums at some point
12:39:48midkaydon't tell me - please don't - that you said "ott" on purpose..
12:40:38pii amm a yslexic and didnt actualy notice till you mentioned it
12:41:15Paul_The_Nerdmidkay: Well, those are still bestowed. It's like a subcategory. The non-bolding was intention. I guess it doesn't come across like that?
12:41:37midkaypi: er, sorry :)
12:41:37 Quit amiconn (Read error: 110 (Connection timed out))
12:41:37*Paul_The_Nerd needs to learn to type.
12:41:38 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
12:41:53midkayPaul_The_Nerd: i realized that sort of as i was typing that message..
12:41:58San||Awayanyone ever hear of
12:42:09San||Awayit is open source and doesn't look bad
12:42:16 Nick San||Away is now known as [San] (
12:42:18midkayi think these should be split up though. "Staff", "Badges you must earn" and "Badges you may apply for" or something.
12:42:26piit doesnt matter it is just a fact and i apologise for departing from the english language from time to time
12:42:51midkayalso i just discovered the horizontal line forum thing - [hl] - and it'd be well-used here.. to divide categories or badges.
12:43:04Paul_The_Nerdmidkay: Okay, lemme apply some more formatting then
12:43:15midkaypi: just so many people do it because they think it's easier or something.. pet peeve i suppose :)
12:43:34midkayPaul_The_Nerd: *waits in apprehension*
12:43:45crwl[San], there's plenty to read about the GP2X on and
12:44:47[San]i was just showing it to you...
12:44:58 Quit _hotwire_ (Read error: 110 (Connection timed out))
12:45:38Paul_The_Nerdmidkay: [hr] I think you meant?
12:45:48midkaywas that it?
12:46:05midkayi thought "hl" for "Horizontal Line"..
12:46:19midkaysome button in the upper right of the formatting options.
12:46:30Paul_The_NerdHorizontal Rule, I assume
12:46:42midkayyeah, that must be it.
12:47:57Paul_The_Nerdmidkay: I think I'm going to make a very brief mention of who has powers.
12:48:02 Join XavierGr [0] (
12:48:11midkayPaul_The_Nerd: you mean who are admins, etc?
12:48:37Paul_The_NerdNo, just a very brief summary of what each group can do.
12:48:39Paul_The_NerdVERY brief.
12:48:51midkayoh, i see. sure.
12:49:02midkaybut that reminds me of something i forgot about, that occurred to me while reading..
12:49:08Paul_The_NerdForum Admins and Global Moderators have all power necessary to enforce the rules of these forums. As well Developers and Experts have been given some limited additional functions to enable them to better be able to do their jobs. Artist and Coder badges do not grant any additional powers.
12:49:09Paul_The_NerdLike that
12:49:23Mikachu <- play is exit and menu is delete? hello consistency
12:49:24midkayvery good.
12:49:31Paul_The_NerdSomething you forgot about?
12:49:37midkayPaul_The_Nerd: "there are a few people sporting this badge. We are the people with full control over the forums, and are the ones to come to when something in the forums doesn't seem to be working right."
12:49:54Bg3rPaul_The_Nerd: can u add the Bger in the forums as a dev (or whatever is called)
12:50:11midkaywhen i read that, it occurred to me that you should maybe list at least a few of the admins.. for two reasons, one because there are very few, and two because - as you say - you should contact them if there's a problem..
12:50:18midkaymight be nice to know who you can go to in such a case.
12:51:27Paul_The_Nerdmidkay: Good idea.
12:53:18Paul_The_Nerdmidkay: Though I'm not certain they should really contact the other admins unless I'm unavailable. The other two are LinusN and Jeff, the latter not really being around our forums very much at all, and the former probably not wanting to be pestered by random PMs that really don't need to be a PM to an admin in 95% of cases.
12:53:23 Quit scorche (Read error: 104 (Connection reset by peer))
12:53:36Paul_The_NerdBg3r: Give me a few minutes, and I'll do so.
12:53:49midkayPaul_The_Nerd: ah, right..
12:53:50 Join Febs [0] (n=medifebb@
12:53:52Bg3rPaul_The_Nerd: thank u;)
12:55:44*midkay tries working on the developer badge a bit..
12:56:32Paul_The_Nerdmidkay: How's it look now?
12:56:59midkayPaul_The_Nerd: absolutely great, nice job. :)
12:57:18FebsDoes anyone know whether the ipod is capable of playing 48khz/24bit audio?
12:57:33Paul_The_NerdFebs: I've heard it can, but I'm not sure where my brain is pulling this from at this time
12:57:37midkayPaul_The_Nerd: one suggestion, hm - a bit of extra spacing between the last badge and the bit about powers..
12:57:46Paul_The_Nerdmidkay: I was just about to do that. ;)
12:57:57midkayPaul_The_Nerd: haha. great minds think alike ;)
12:58:14midkayof course that always makes me think that crappy ones do too, so.. :)
12:58:51Paul_The_Nerdmidkay: Also, italicized the last bit. I think it sets it off a little better as a "footnote" style entry.
12:59:15perplexitymidkay, great minds may think alike, however fools seldom differ ;)
12:59:22midkayPaul_The_Nerd: perfect... now just await a new Developer badge, in all likelihood..
12:59:36midkayperplexity: that's perplexing indeed :D
12:59:41amiconnhotwire_: I rarely frequent the forums these days...
13:02:31Paul_The_Nerdmidkay: I now have the power to upload badges at will. I actually need a "Member" badge now, for normal members. Because I don't seem to have the option to give them 0 "icons", just 1 at minimum. I was thinking just the R from Rockbox, in a much narrower box, to widely differentiate it from the actual badges.
13:02:48*Paul_The_Nerd was gonna make that sometime soon.
13:03:19midkayPaul_The_Nerd: yeah, ftp access is useful... :) sounds good, i just agree that it should be quite.. clearly different from the others.
13:03:48Bagderless "distinct" I think
13:03:56midkaylet me just see if i can get a better developer one. i'm happy with the others, they're all very legible IMO, except developer is rather hard to read where it overlaps the rockbox logo..
13:04:01Bagderthe badges kind of attract ones eyes
13:04:19Paul_The_NerdYeah, normal members need to not look *anything* like they have a badge.
13:04:30Paul_The_NerdBut I don't want everyone to just have one of those silly little stars after we drop post-based groups.
13:05:30 Quit Kitar|st (Read error: 104 (Connection reset by peer))
13:05:50 Join Kitar|st [0] (
13:06:42Paul_The_NerdWe could always do a black treble cleff on a transparent background (GIF)
13:06:45midkaysigh.. hm...
13:06:49Paul_The_NerdThat way it's very non-detracting
13:07:01midkayPaul_The_Nerd: great minds again, i was thinking about something like that. at least involving the clef. ;)
13:07:26midkayPaul_The_Nerd: hm, any ideas on a potential different color for Developers? red just seems to blend in here.. i wonder if I could get it to stand out a bit more somehow..
13:07:52Paul_The_NerdI have no idea, but since developers names are already Gold-ish in the list, any color that doesn't look too much like the others should be fine.
13:08:11tucozisn't it possible to use .png? This png is transparent
13:08:27Bagdertucoz: in IE too?
13:08:28 Join lorijho [0] (
13:08:31Paul_The_Nerdtucoz: A monochrom gif would be a good deal smaller, though, right?
13:08:46BagderIE at least has a history of sucking at transparant PNGs
13:08:48tucozBagder, ah. the stupid IE.
13:09:16tucozit has a history of sucking at a lot of things
13:09:32hotwire_amiconn: I posted a version of my uisimulator/sdl/button.c online somewhere, but would like a better idea of what you want for cleaning it up.
13:09:46dwihnoAssuming people using IE6, I think 8 bit "transparency" is OK, but alpha channels = mess
13:10:34tucozwhat status does the ipod 3g have? Is it usable? I wonder if we should create a manual for it.
13:10:52Paul_The_NerdIt's "nearly usable"
13:11:16Mikachudwihno: you mean 1bit transparency? :)
13:11:17linuxstbI would say that anything we offer cvs/daily builds for should have a manual.
13:11:24*Paul_The_Nerd agrees with linuxstb
13:11:33*Bagder nods
13:12:01tucozI agree to some extent. I wound at least expect it to be able to play music before we make a manual for it.
13:12:15tucozmaybe that is a criteria to end up as a daily
13:12:16*dan_a agrees with tucoz.
13:12:37Paul_The_NerdWell, 3G can play some music, though.
13:12:43Paul_The_NerdIf I understand, at least
13:13:13dan_aIt can. It only has a little skipping as long as you stay in the menu and not on the WPS
13:13:23dwihnoMikachu: I was thinking transparency in 8bit-mode, but you're true.
13:13:30dwihnoyou're right, and true ;)
13:13:44tucozok. My concern is that unless it is a fairly usable state, we have to make a lot of 'exclude this from 3g'
13:13:54tucoz*in a
13:14:33tucozhmm. But that we could solve quite much with the platform files. Maybe I'll start working on that tonight then :)
13:14:53amiconnhotwire_: My basic idea is to change the way how pc keys are assigned to player buttons. As more and more targets get added, the current method becomes difficult to follow and hence error prone
13:15:12dan_atucoz: I think that apart from the skipping, the only difference between the 3g and the 4g is that the 3g does not detect USB being plugged in.
13:15:15amiconnToday, we check for button defines, and then evaluate pc key values based on that
13:15:38Paul_The_Nerdpondlife: Done.
13:15:45 Join lightyear [0] (
13:15:51amiconnMy idea will add a very lengthy, but regular list of pc keys eventually mapped to player buttons
13:16:00tucozdan_a, ok. Then I'll start making a platform file for it later on.
13:16:06pixelmadan_a: it has a different button layout, too
13:16:18amiconnThe actual mapping will consist of a bunch of #defines per target
13:16:30tucozpixelma, that is handled in the platform files
13:16:40tucoz(to some extent)
13:16:47pixelmaok :)
13:17:09linuxstbtucoz: Yes, playing music is more or less the main criteria for adding a cvs/daily build. By the time that happens, most other things in the port are working - I'm not sure there is much which works on other ipods and not on the 3g.
13:17:12flux__hmm.. can I submit a feature suggestion without creating an account? (I don't have an sf account..)
13:17:13tucozBut this might be a good starting point for the action mappings for the manual as well.
13:17:50tucozlinuxstb, good. know
13:17:58amiconnI considered doing it the other way 'round, but that wouldn't allow multiple keys per button, which is a feature I want to keep
13:18:12amiconnThe numpad mappings are rather cumbersome on a laptop
13:18:47pixelmatucoz: H300 is ready -
13:18:53Mikachuflux__: the tracker hasn't been on sourceforge for a long time,
13:18:55tucozpixelma, I think your drawings look beautiful in the manual
13:18:59Paul_The_Nerdflux__: You don't need a sourceforge account. You need to create a flyspray account at our flyspray page.
13:19:10tucozpixelma, thanks
13:19:15flux__iow, yes :). well, maybe I'll bother at some point :)
13:19:34 Quit RoC_MM (Read error: 110 (Connection timed out))
13:20:10aliasktucoz: What are the svg files used for, they look great.
13:20:26flux__but for the time being, a single simple thing I can /lastlog from irc later: on iRiver iHP-120 the boot loader should check the hold switch again after displaying the 'release hold switch'-message but before shutting down
13:20:43tucozaliask, they are used in the manuals for the key-reference images. I think could come to other uses as well.
13:21:02pondlifePaul_The_Nerd: : cheers
13:22:02tucozaliask, you can check todays built player/recorder/recorderv2fm/h1xx/nano/video/mini manuals in the "The Rockbox interface"-section.
13:22:26midkayPaul_The_Nerd: how about darkish-brown with a glow surrounding it? i can absolutely not find any light-red shade that works here.. the optimal solution would be to shorten it, but i don't think "dev" would be as descriptive.
13:24:14Paul_The_Nerdmidkay: Yeah, dev wouldn't be as descriptive. I can't picture the dark brown in my head, so show me. :-P
13:24:31midkayhaha. it hardly looks brown :) just about to upload what i have..
13:25:06BagderPaul_The_Nerd: 'Lear' should get a dev badge
13:25:20midkayit doesn't look similar to the others yet it's very readable.. i can't really come up with anything better.
13:26:00Mikachucan't you make it taller and write developer under the logo?
13:26:15midkayno, they should all be the same size..
13:26:26Paul_The_NerdWe *could* move to the a new size.
13:26:29midkayred is still kind of hard to see, light red that is..
13:26:30Paul_The_NerdIt'd require remaking all of them.
13:26:53Paul_The_NerdBut that'd give some more freedom with readability, and such.
13:26:53midkayPaul_The_Nerd: i wouldn't mind.. i thought this was kind of a restriction, when you were very specific about the size when i first made these.
13:27:06Paul_The_Nerdmidkay: The restriction was based on the height of the stars we had before.
13:27:07midkayyeah, not overlapping the logo..
13:27:21midkaywell, i could put together a taller Developer badge to see how it looks.
13:27:33Paul_The_NerdI didn't want the Moderator Stars to be any shorter or taller than the badges
13:27:49midkayah, i see.. now that we're all badges, though..
13:27:52Paul_The_NerdNow that we're dropping the badges altogether, we can pick an arbitrary height and width (though I think we're almost at maximum width)
13:28:00Paul_The_NerdErr dropping the stars altogether
13:28:03midkay80x20 currently.. what about like 80x30?
13:28:36Paul_The_NerdI really wouldn't even object to 80x40, as long as it ends up looking nice and readable.
13:28:41hotwire_amiconn: okay... i could use a really small template in that case, just to be sure.
13:28:41tucozQuestion: The buttons on the h300 has a '+' and a '-' print on them. Should these be renamed to Plus and Minus in the manual? (It is now Up and Down)
13:28:43midkaycool, i'll see what works..
13:28:49Paul_The_NerdBagder: Sorry for the slow response. I'll get to that now
13:29:31Paul_The_NerdBagder: Done
13:30:00hotwire_the other problem is that while at the present I have keys assigned to the various iriver remotes, there is no way of identifying which remote is in use in the simulator... and this is something i can't attack, lacking intricate knowledge of the sim, nor time.
13:30:48tucozAnyone with a H300 around?
13:31:21tucozWhat key is the one to the left hand side (next to the display)
13:31:32tucozsimilar on the right hand side
13:31:48pixelmatucoz: it's not a key it's the microphone
13:32:03pixelmaon the right hand it's the hold switch
13:32:03tucozthe one on the right then?
13:32:07tucozah. thanks
13:34:14 Join lodesi [0] (
13:34:49 Quit eGen_ (Read error: 104 (Connection reset by peer))
13:35:27amiconnhotwire_: I was thinking about doing it like that:
13:35:28 Join eGen_ [0] (
13:43:03midkaywhat do you think?
13:43:33***Saving seen data "./dancer.seen"
13:44:07Paul_The_NerdMaybe left align developer and shorten the box as a whole? They don't need to be 80 wide any more either, they can be shorter.
13:45:05aliaskAnd the word developer looks a little blurry, but it may be my eyes
13:45:11Paul_The_NerdThe only constraint we're working with now, really, is that A) they need to not be huge, and B) they all need to share the same dimensions, whatever we choose.
13:45:47midkayaliask: it's slightly shadowed, i'll try to give it more contrast.
13:45:58midkayPaul_The_Nerd: understood, let me try that..
13:46:12midkayI tried centering the logo but it looks awful. left-aligning the text might be the answer.
13:46:21 Join fergie [0] (
13:46:26 Join pike^laptop [0] (
13:47:26 Join Crackerizer [0] (n=root@
13:48:08Crackerizerhello, all.
13:48:19Crackerizerwhat is needed to build a uisim?
13:49:00Bagdera native gcc and build environment
13:49:45LinusNand a clue
13:49:45Crackerizeri have to build sdl too?
13:49:47BagderCrackerizer: you tried and failed, or why are you asking?
13:50:09BagderCrackerizer: build it or download it, yes
13:50:11Crackerizeryep, i tried. I seems like i need the sdl lib.
13:50:25LinusNCrackerizer: windows or linux?
13:50:36firenxugh i get this error with rockdoom.. you cannot - with the shareware version?
13:50:37peturcopy the sdl.lib to the folder where the exe is
13:51:01*petur was talking win32 :/
13:51:06BagderCrackerizer: then get your distros sdl-devel package
13:51:11Paul_The_Nerdfirenx: Yes, you cannot use addons with a shareware base wad.
13:51:21Crackerizerok, thanks you. I'll try.
13:51:31 Quit pi ("Everyone has to believe in something, I believe I will have another drink")
13:51:34firenxso i need .. a different base1.wad?
13:52:03 Join pi [0] (
13:52:07 Quit pi (Client Quit)
13:52:09Paul_The_NerdYou need to either use Freedoom as a base wad, or buy a retail copy of Doom or Doom2 and use that
13:52:09midkayPaul_The_Nerd: alright, now 66x30, new font, less blurry:
13:52:36Paul_The_NerdI like it.
13:52:39 Join pi [0] (
13:52:48fergiejust checking if there is ay development on the USB/Charger detection for iPod?
13:53:09midkayPaul_The_Nerd: so good template to work off of? alrighty :)
13:53:24Paul_The_NerdIn my opinion, yes.
13:53:39encodecan i follow the linux guide for building rockbox on os x?
13:53:48midkayno other conceivable tweaks?!
13:53:58Bagderencode: yeps, it should work
13:53:59CrackerizerI looked at some rockbox's codes today. It's very complex and a lot of global variables.
13:54:18Bagder"very complex" ?
13:54:27Paul_The_Nerdmidkay: Not that I can think of.
13:54:40CrackerizerI mean it's quite complicated.
13:54:56Bagderparts of it might be a bit advanced, sure
13:55:19Bagderbut have a look in linux and compare
13:55:28BagderRockbox is easy after that
13:55:54perplexityrockbox is well compartmentalised and quite easy to read/understand.. I think anyway
13:56:18*amiconn hopes to commit assemblerised lcd_yuv_blit() for h300 tonight
13:56:28encodeits a shame all the ipod works relies on reverse engineering
13:56:42linuxstbamiconn: Is it working yet?
13:56:51amiconnBased on jhMikeS X5 version, which I am planning to tweak a bit as well
13:57:02amiconnlinuxstb: Yes, but I had to change it a bit
13:57:18linuxstbWhat sort of speed improvement is there?
13:57:40amiconnDireclty converting to RGB565 doesn't work well. Results in a greenish image due to roundoff errors
13:57:58firenxPaul_The_Nerd: can i set any wad as the base wad? or do i need the doom registered as base wad?
13:58:07amiconnSo I'll convert to RGB666, then drop one bit from R and B
13:58:44Paul_The_Nerdfirenx: As I said, you need either Freedoom, or commercial versions of Doom or Doom 2
13:59:02Crackerizertoday is my first time looking at the codes. I think i need to spend sometime to understand the structure of the codes.
13:59:03firenxoh freedoom, ok
13:59:11 Join Febs_ [0] (n=medifebb@
14:00:02 Quit Febs (Read error: 104 (Connection reset by peer))
14:01:33*dwihno took a look at the topic. Great idea!
14:02:50Bagderafrica is empty still
14:02:52 Join Rondom [0] (
14:03:40peturthe nordsea has one member ;)
14:03:52Bagderand it seems eastcoasters in the US are more Rockboxish than the westcoasters
14:04:51 Join thegeek [0] (
14:05:15*dan_a is proud to represent "people in the middle of the Irish Sea"
14:06:03*Bagder detects 2 red dots in the Stockholm area
14:07:46*Paul_The_Nerd is proud to represent "One of two people in Texas"
14:08:46XavierGrPaul: who is the second?
14:08:53 Join mikearthur_ [0] (
14:08:54 Join Rob2222 [0] (
14:09:12XavierGrPail: Jeff from MR is from Texas too (IIRC)
14:09:12Paul_The_NerdBHSPitLappy I believe.
14:09:21*perplexity represents one of two in the middle east
14:09:32Paul_The_NerdI'm from the good part of Texas, though.
14:09:51peturno bushes around?
14:10:10pondlifePaul_The_Nerd: would that be Austin?
14:10:21Paul_The_Nerdpondlife: Aye
14:10:43pondlifeHmm, I didn't actually know that. it was a guess... based on the 13th Floor Elevators back catalogue
14:11:12amiconntucoz: What's the change to the associated .png in cvs? Doesn't look like you removed it, as I would have expected...
14:11:22pondlifeI'm surprised how few UK users are on the map. I must pester some of the ones I know.
14:11:38Bagderwell, there are ~200 on the map
14:11:44Bagderwe have >5000 in the forum
14:12:04Bagderso there's room for some more ;-)
14:12:10XavierGrand the forum is quite active I must say
14:12:22Bagderinsanely so
14:12:49tucozamiconn, that is because we need a .png for the html manual
14:12:49pondlifeMaybe put a link to the map from the Rockbox homepage? At least for a while....
14:12:58Bagdergood idea
14:13:21pondlifeIs it actually hosted by Google?
14:13:21tucozamiconn, but the updated .png is based on the new .svg versions
14:13:36pondlifei.e. bandwidth isn't going to hurt Rasher?
14:14:15Bagderyes, the map stuff is google's
14:14:58Bagderyday was "11 months since 2.5 day"
14:18:53Febs_I'm contemplating removing all of the content of the WikiManual and replacing it with cross links to the html manual.
14:19:47amiconntucoz: Can't you just use the .svg for the html manual
14:20:05dionoea_workdo all browsers support svg ? (i guess so)
14:20:13amiconnFirefox displays svg just fine, and afaik there is a plugin for Internet Exploder
14:20:15Febs_The html manual is as up-to-date as we can make it short of another feature freeze, while the WikiManual is falling further and further out of date.
14:20:17dionoea_workmaybe not IE 5.5 ... but who cares ?
14:20:18Paul_The_NerdFebs_: I think this would be a Good Thing. Bu that's just me.
14:20:19tucozI am not sure why. bluebrother had an explanation to why we can not do taht right now
14:20:25amiconn(from Adobe)
14:20:30tucozI think it has to do with htlatex
14:20:32dionoea_workamiconn: i though that IE 6 support was native
14:20:38dionoea_worki can check here if you want
14:20:48dionoea_worki only need a svg to test with :)
14:21:26tucozthat is, the html generator does not support vector formats iirc
14:21:50amiconnAh, hmm.... :/
14:23:04pondlifeLinusN: The problem on the ML probably refers to ID3 view. tree.c, set_current_file() doesn't support the tagtree.
14:23:19dionoea_worklooks like IE 6 doesn't like svgs by default either
14:23:32dionoea_workmaybe if i embed them in a webpage ...
14:24:09*Lear is contemplating going dual core...
14:24:47dionoea_workdoesn't work in a webpage too
14:26:04 Quit Rob2222_ (Read error: 110 (Connection timed out))
14:28:06 Nick Febs_ is now known as Febs (n=medifebb@
14:28:11LinusNpondlife: if so, the poster should mention that he uses the tag database mode
14:29:29 Quit [San] (Read error: 104 (Connection reset by peer))
14:30:08pondlifeWell, that's a bug I noticed a while back, but I'll let him confirm....
14:30:25pondlifeOne of those you notice then forget as you're so excited about tagcache
14:31:15LinusNthe wiki manual should be removed imho
14:31:30pondlifeYes, rather than being left to rot.
14:31:31 Quit Crackerizer ("using sirc version 2.211+KSIRC/1.3.12")
14:31:47pondlifeThere are also quite a few out of date wiki pages...
14:31:53LinusNoh yes
14:32:17LinusNlike the "iriver for noobs" stuff etc
14:32:43pondlifeAbout 50% of the wiki maybe..? ;-)
14:32:59Bagderpondlife: you've got mail ;-)
14:33:04 Join daurnimator [0] (n=quae@
14:33:14pondlifeBagder: you will have soon too
14:34:05 Quit mikearthur_ ("Konversation terminated!")
14:34:14Genre9mp3Does anyone know the resolution of the H10 LCD?
14:34:22 Join rconan [0] (n=richard@
14:34:29BagderGenre9mp3: yes 160x128 and 128x128
14:35:03XavierGr2 resolutions?
14:35:09Bagdertwo different ones, yes
14:35:15XavierGrhow se?
14:35:16Bagderthe 5/6 GB has a smaller one
14:35:27Bagderthe 20GB a bigger
14:35:35*amiconn wonders why both are called H10 when they're in fact different things
14:35:40Bagdeririver also ships different firmwwares
14:35:55Paul_The_Nerd1739 of our forum members have never posted...
14:36:07amiconnIirc there's even a 'H10 junior' which is flash based (!)
14:36:20Bagderamiconn: they must like the name H10 very much! ;-)
14:36:42Genre9mp3128x128 is a weird resolution IMO
14:37:00XavierGrtell that to iriver
14:37:53Paul_The_NerdBagder: It was right before they decided that nearly everything needs to have 10 in its name
14:38:06Paul_The_NerdThere's now, what, H10, E10, U10, N10, T10 and D10?
14:38:18Bagderyeah, those guys like 10
14:38:23Paul_The_NerdThough the T-series has a 20 and 30, the N has an 11 and 12
14:38:30Bagderthose must be mistakes
14:38:35Bagderthey should be cancelled
14:38:35Paul_The_NerdClearly seems that they get very inspired with the namings at iriver
14:40:30piI still want to know what hapens wheen they rum out of letters
14:40:52dan_aThen they release the 1010
14:41:17pondlifeMaybe it's binary?
14:41:23pior go to the greek alphabet
14:41:25 Quit JoeBorn ("Leaving")
14:41:30Paul_The_NerdOkay, post-based membergroups are now a thing of the past.
14:41:30 Quit jaebird ("Ex-Chat")
14:41:37pondlifeSo H1 -> H10 -> H11-> H100
14:42:05Genre9mp3H1 lol
14:42:31pondlifeOr the H00000000000000000000000000000001 as it should be called
14:42:31Bagdersince they already had h100 and h10 they must be going for h1 soon
14:42:35 Join jaebird [0] (n=jae@
14:42:35Bagderrotate right
14:42:44Genre9mp3or the H0,1
14:43:22Bagder"there are 10 kinds of people..."
14:43:55pondlifeplease, no...
14:43:55Genre9mp3"those who understand binary and those who don't..."
14:44:01*pi smirks
14:44:16*dan_a wonders where the H320 fits into all this
14:44:33pondlifeOr maybe H is for hex
14:44:34*LinusN once had to go great lengths to try to explain that joke to a friend
14:44:42peturthe h340 certainly fits nowhere, it's a bit big ;)
14:44:50*pondlife had to too... hence the remarks above
14:45:11perplexitySpeaking of h3x0.. <whispers> bootloader..
14:45:17pondlifepetur: My new H340 is smaller than my old Archos Recorder...
14:45:24pondlifeAh, yes - USB anyone?
14:45:31Mikachuthere are 10 kinds of people, those who understand trinary, those who don't, and those who confuse it with binary
14:46:33*Bagder drowns in virus mails
14:46:50*pondlife didn't send them
14:46:58LinusNthe bootloader usb mode is a problem for me, because it works for me on all my computers
14:47:29pondlifeIs there anything us poor failure cases can do to help diagnose?
14:47:36perplexityOk, what about the remote detection and booting into rockbox when external power is applied ?
14:47:54LinusNperplexity: those are a bit easier
14:47:58 Join San||Away [0] (
14:48:24pii wish i could halp you, because my bootlodar usb dosn't
14:48:29pondlifeBetter to get the current one working properly before adding code paths?
14:48:31LinusNthe ext-power thing is somewhat problematic, since we then need to implement a charging screen in the boot loader
14:48:33perplexityTo be honest, the bootloader failure is 100% reproducible for me on all my machines, but I just don't find it an issue to wait the extra 8 seconds for rockbox to boot before I plug it in
14:48:50pondlifeSame here, but it should work
14:48:56perplexitycan't just have it boot rockbox? it charges while it's running anyway..
14:49:22LinusNno, because there might not be enough battery to spin up the disk to load rockbox
14:49:30pi perplexity:it is just annoying when i forget to turn it on first
14:49:36*pondlife wonders how many others use the sim as their main PC media player...
14:49:47 Join Rondom_ [0] (
14:49:50Genre9mp3bootloader USB mode never works for me, too
14:49:53linuxstbJust flash Rockbox, then you can forget about it...
14:49:57perplexityOh, and you need rockbox loaded to enable the charger I guess..
14:50:12 Part Paul_The_Nerd
14:50:20pipondlife: i did concider it for a we bit
14:50:25 Quit Rondom (Nick collision from services.)
14:50:27 Nick Rondom_ is now known as Rondom (
14:50:40preglowit doesn't work here...
14:50:41pondlifepi: So you can be testing while you work...
14:50:51Genre9mp3pondlife: I'm happy with my foobar2000
14:50:54tucozmarkun, around? what is the program you created the logo called?
14:51:14piunfortunatly i have yet to sucsefuly compile the sim myself
14:51:22pii keep getting errors
14:52:01pondlifeI wonder if there's mileage in making daily sim builds available? Useful for testing maybe...
14:52:07 Quit Rondom (Client Quit)
14:52:10XavierGrUSB Bootloader mode doesn't work for me too!
14:52:23perplexityThe Iriver firmware boots fully here when I apply ext-power, it only drops to charge only mode on idle timeout or manual shutdown..
14:52:40XavierGrAnd I've seen that many users have the same problem
14:52:48XavierGrit is quite strange though
14:53:06pondlifebootloader.c - if (strcmp(username, "LinusN") == 0) do_usb()
14:53:09XavierGrwhy would the H300 be recognized from the main firmware and not in the bootloader?
14:53:23 Quit Febs (Read error: 110 (Connection timed out))
14:53:27Genre9mp3pondlife: LOL
14:53:27LinusNpondlife: :-)
14:53:37pondlifeDoes it work for anyone else then?
14:54:20peturworks fine here
14:54:39perplexityI have 4 different machines and it fails on all of them here..
14:54:40pondlifepetur: Where did you buy your H300?
14:54:51pondlifeWhich country?
14:55:14peturthe Belgian branch of a Dutch internet store
14:55:31 Join Rondom [0] (
14:55:37pondlifeOK. Was it using European firmware then?
14:55:49pondlifeSo was mine :(
14:55:55pondlifeSo not that then
14:56:50pondlifeWeird how it seems to be either "works on all PCs" or "works on no PCs". i.e. it's device-dependent (or maybe cable dependent?)
14:57:29pii have tryed with different cables, no difference
14:57:36peturI have another issue, being that at work RB stays in usb mode when detached. Only when plugging it via the hub of a dell monitor. It's related to the way the original firmware is configured regarding powered hub (ie usb charging)
14:57:55 Join wow [0] (
14:58:01perplexityLinusN: per the ext-power, why can't we do what the iRiver firmware does.. boot rockbox fully, if it idle times out or is manually shut down just close most of rockbox down, spin down the disk and sit in a charging screen until the power is removed, when we then completely power off ? no addition to the bootloader required
14:58:26pondlifeWell that needs the bootloader to handle charging anyway
14:58:31LinusNperplexity: and what to do when the battery is so low that we can't boot?
14:59:00LinusNremember, the iriver firmware boots from flash, rockbox boots from disk
14:59:01perplexityright, so the bootloader has to enable the charger? details I'm obviously hazy on..
15:00:08LinusNit's not a complicated issue, it's just a tiny bit more complicated than just always boot rockbox when the external power is connected
15:00:32perplexityI see..
15:00:44wowis there an rockbox port of elio p722 ?
15:01:06 Quit thegeek (Read error: 104 (Connection reset by peer))
15:01:18markuntucoz: autotrace + sodipodi
15:01:28Bagderwow: no
15:01:37Bagderwow: unless you start one
15:01:38tucozmarkun, ok. thanks
15:02:03wowbagder : i would like to start one
15:02:17 Join BHSPitLappy [0] (
15:02:18Bagderwow: well, its a PP based player so mi4code is your friend
15:02:29Bagderit is likely to be similar to existing targets
15:03:03wowon the mi4code page is an file for the elio
15:03:09wowbut is crypted
15:03:23Bagderso use mi4code to decrypt it
15:04:01wowif I decrypt it wich progamming language will it be ?
15:04:20 Join rotator [0] (n=e@rockbox/developer/rotator)
15:04:33Bagderwow raw binary for the PP
15:06:47wowbut an bootloader do not exists ?
15:06:47wowfor the elio ?
15:07:22amiconnBootloader USB always works for me, but there's the nasty 'clack'
15:07:22Bagderwow: download a firmware package from their site and you'll get one iirc
15:07:46Bagderwow: I believe its called bootloader.rom or similar
15:07:54Bg3rLinusN: regarding hxxx bootloaders... what about displaying the text on the remote too ?
15:08:05 Quit rconan ("Leaving")
15:08:08Bagderwow: but one of those firmwares are decryptable even without BL
15:08:51pondlifeHmm, can anyone explain what does?
15:09:16pondlifeI'm wondering if playback.c should have the same definitions of IRAM size/origin as does.
15:09:16LinusNBg3r: that should be done too
15:10:17amiconnpondlife: Surely not
15:10:32pondlifee.g for X5, specifies IRAM starts at 0x10014000, but playback.c will use a different value
15:10:42amiconnThe core uses one half of the available iram, codecs/plugins the other half
15:10:54pondlifeSo they should be different then...
15:11:04pondlifeBut sometimes they are the same!
15:11:05*petur runs off for a while
15:11:35 Quit perplexity ("*terminated - signal 11*")
15:11:42pondlifeIRiver has IRAMORIG 0x1000c000 for both for example..
15:12:00 Join DeejAy [0] (
15:13:34 Quit DeejAy (Client Quit)
15:13:52 Quit webguest72 ("CGI:IRC (EOF)")
15:14:40wowBagder i don't think im able to port or even modificate the elio firmware
15:15:12daurnimatorBagder: whats the latest on neuros?
15:15:13Bagderwow: using the h3mod you should be able to modify the pics if that's your game
15:15:35Bagderdaurnimator: Zagor got an OSD dev board
15:15:41daurnimatori saw
15:15:46daurnimator(on groups)
15:15:52Bagderthat's the latest
15:16:16daurnimatorwhats the likelyhood of rockbox running on it IYO
15:16:40Zagorpretty big, if anyone wants to do it
15:16:44Bagderit should be fairly straight forward once someone actually tries to do it
15:17:10Bagderand then, the gmini400 port should be a walk in the park :-)
15:17:19daurnimator400 is diff arch
15:17:27wowBagder: the pics in the firmware ?
15:17:29Bagderonly shows how clueless I am
15:17:47Bagderwow: yes
15:17:48wowor the pics in the rockbox ?
15:17:57Bagderthere is no rockbox for elio p722
15:18:12wowi know
15:18:22Bagderwow: go get h3mod, load your mi4 file and start poking about
15:18:41 Join [San] [0] (
15:19:23LinusNpondlife: hmmm, i think you're on to something
15:19:28daurnimatori should probably work on my port more
15:19:38daurnimatorbut - i'm busy this week
15:19:41daurnimatorits a slow process
15:19:51amiconnpondlife: The core should use IRAMORIG 0x10000000 on both irivers and X5, and codecs/plugins should have IRAMORIG 0x1000c000 for irivers and 0x10014000 for X5
15:20:23amiconnThe MCF5249 has 96KB of IRAM, split evenly between core and codecs/plugins
15:20:35Bagderwe should probably change that to have the same ORIG but a "COREUSESTHISMUCHIRAM" define
15:20:47LinusNplayback.c uses 0x1000c000 for both iriver and x5!!!!!
15:20:51amiconnThe SCF5250 has 128KB of IRAM, of which 80KB can be used by the core and 48KB by the codecs/plugins
15:21:04LinusNno wonder the x5 crowd has problems with voice on x5
15:21:16 Join thegeek [0] (
15:21:31amiconnLinusN: Ouch! Iirc that value is for the codec swap - and it means codec swap doesn't work properly...
15:22:03LinusNthose defines should be in the config.h files
15:22:19pondlifeThis explains a crash I'm looking into then
15:23:24pondlifeI'll leave this one to someone who actually understands it.
15:23:49 Quit daurnimator ("daurnimator was using Daurnimator's Legendary Script")
15:23:52wowbagder thx for help, but I think the modification of the firmware is a bit to difficult for me
15:25:21wowcu @ll :)
15:25:46 Quit wow ("Leaving")
15:27:12fergiejust checking if there is any development on the USB/Charger detection for iPod? The faster drainign than charging issue?
15:27:36*pondlife lunches
15:29:29MmmmHow do I use gdb on a sim build? I choose Devel, Debug, Sim and it just freezes "simulator build enabled"
15:31:57 Quit San||Away (Read error: 110 (Connection timed out))
15:32:50 Join JdG [0] (
15:33:30*JdG is bored... have to wait for these silly buggers to finish playing poker and leave before i can go bac t my computer :'(
15:33:31 Quit Rondom ("Ex-Chat")
15:35:09amiconnLinusN: Hmm, I noticed that the X5 uses the old conservative ATA timings, while the irivers use the faster timings. Did you find the time to check whether the fast timings stay within specs?
15:35:25LinusNno i didn't find the time
15:35:29*amiconn wants to complete the speed settings table
15:39:25 Quit TCK (Read error: 104 (Connection reset by peer))
15:42:33 Join merbanan_ [0] (
15:43:09 Quit aliask ("why am i still awake...")
15:43:34***Saving seen data "./dancer.seen"
15:49:33 Quit JdG ("CGI:IRC (EOF)")
15:49:59amiconnLinusN: The strange thing about the ata timings is that when I calculate them given the rules in the datasheet, I get what I've put in system.c
15:50:24amiconnBut they don't match the 'typical' values also quoted in the datasheet...
15:50:25LinusNi think those rules are plain wrong
15:50:44amiconnBut the calculated values work
15:50:48 Join perplexity [0] (
15:51:10amiconnI don't know of any problem caused by these
15:51:28amiconnI should probably compare with the SCF5250 datasheet
15:52:17LinusNamiconn: i meant that the typical values are wrong
16:00:15 Quit actionshrimp ("( :: NoNameScript 4.01 :: )")
16:06:41 Join MadDog011 [0] (
16:06:46 Part MadDog011
16:08:04 Join daurnimator [0] (n=daurnima@unaffiliated/daurnimator)
16:12:12 Join spiorf [0] (
16:14:06 Join barrywardell [0] (
16:15:33 Quit Mmmm ("Byeee")
16:16:36barrywardelli'm going to try getting sound working on the H10
16:16:53dionoea_workdid you get yours back ?
16:17:00barrywardellI'm told it will probably be identical to the ipod
16:17:19barrywardellso can anyone point me in the right direction to look in the rockbox code?
16:17:28barrywardelldionea_work: i got a new one
16:17:35dionoea_workfor free ?
16:18:11barrywardellno. iriver wanted to charge €140 to repair it
16:18:12amiconnbarrywardell: Do you already know what audio codec/dac the H10 uses?
16:18:13 Quit barrywardell (Client Quit)
16:18:18 Join barrywardell [0] (
16:18:24amiconnbarrywardell: Do you already know what audio codec/dac the H10 uses?
16:18:42barrywardellamiconn. it uses the same on as the ipod
16:18:48barrywardellwm8731 i think
16:18:51amiconnThere is no _the_ same
16:19:02amiconnDifferend ipods use different codces/dacs
16:19:13barrywardellsorry, the same as the ipod mini
16:19:20barrywardell1st generation
16:19:52amiconnCodecs used in ipods: WM8711/8721/8731 (identical software-wise, except the first 2 have no recording), WM8758 and WM8975
16:20:03 Join t0mas_ [0] (
16:20:18amiconnOkay, then it's the WM8711/21/31 family
16:20:19barrywardellit's definitely wm8731
16:20:32 Join PaulJam [0] (
16:20:40 Quit JdGordon ("using sirc version 2.211+KSIRC/1.3.12")
16:20:49 Quit t0mas (Read error: 104 (Connection reset by peer))
16:21:27merbanan_barrywardell: wasn't the player covered by warranty ?
16:22:16dan_abarrywardell: Have a look in pcm_playback.c - I've not got a full grasp on how it works, but I think that's where data is fed to the CODEC when the CODEC asks for it
16:22:26barrywardellno. "The HDD within the device is defective and needs to be replaced. The device has been altered by a third party."
16:22:31amiconnThe driver is in firmware/drivers/wm8731l.c
16:22:35barrywardell"Due to the fact of contributory negligence, the repair of the player is not covered by warranty."
16:22:50dionoea_workdevice has been altered ... software ?
16:22:55dionoea_workor did you do hardware mods ?
16:23:16barrywardelldan_a: so hopefully i can find a function there to feed audio data direct to the codec
16:23:40barrywardellno hardware alterations, but I did open it to determine the hardware
16:24:32dionoea_workdid you ask them about what they called alterations ? (just curious :p ... can software modifications void the waranty btw ?)
16:24:34dan_aI think that is what the FIQ handler does
16:24:44Mikachudionoea_work: opening it does anyway
16:25:05dionoea_worknot if you keep the little sticker :) although that might be hard to do
16:25:42dan_a(BTW, I'm bidding on a faulty H10 on ebay, so with luck I might be able to join in the fun)
16:25:49amiconnbarrywardell: I think you just need to include the mentioned driver
16:26:35barrywardelli have defined HAVE_WM8731
16:26:54barrywardellso I think that should include the driver for the codec
16:27:37barrywardelldan_a: great. I think iriver will send my broken h10. if they do, you can use any parts from it you need
16:28:07markunbarrywardell: I'm working on the wolfson codes a bit because the gigabeat also uses one
16:29:19markunbarrywardell: perhaps you can take a look at it:
16:30:15barrywardellmarkun: I'll look at that patch. What changes does it make?
16:30:48dan_abarrywardell: Thank you - it's the charging that's not working, so it should be fine for testing and development. I'll only get it if the price doesn't go too high though!
16:30:50markunit moves the hardware specific code to a seperate file
16:34:31 Quit pi ("Everyone has to believe in something, I believe I will have another drink")
16:36:01 Join pi_ [0] (
16:39:10 Quit PaulJam (".")
16:42:49dan_abarrywardell: Do you know how to use the I2S yet? Is that the same as the iPod? If so, audio might work already
16:43:53barrywardellI think it is the same as the ipod
16:44:20barrywardellI think you are right that audio might work already, but have to figure the whole I2S thing out
16:44:25 Join Criamos [0] (
16:47:19 Join MarcoPolo [0] (
16:49:23barrywardellMrH says the audio codec interface of the H10 is very likely to be the same I2C+I2S interface the ipod uses
16:51:17 Join San||Away [0] (
16:56:54 Join saratoga [0] (
16:58:15 Join PaulJam [0] (
16:58:49 Join jhMikeS [0] (
16:58:58saratogai have questions about profiling support
16:58:59 Quit jaebird (Read error: 104 (Connection reset by peer))
16:59:09saratogacould someone help me out?
16:59:19 Join jaebird [0] (n=jae@
16:59:29 Join actionshrimp [0] (
16:59:58jhMikeSAre all the SWCODEC players with tuners able to record radio? I want to simplify my #ifdefs.
17:01:09 Join Febs [0] (
17:01:17Bagdereven if the current situation is like that, it doesn't sound like an assumption that will hold forever
17:01:48Bagderand I'm not sure if the current situation is so
17:02:07jhMikeSDidn't audio_set_recording_gain(sound_default(SOUND_LEFT_GAIN),
17:02:07jhMikeS sound_default(SOUND_RIGHT_GAIN), AUDIO_GAIN_LINEIN);
17:02:07jhMikeS pm_playback = true;
17:02:07DBUGEnqueued KICK jhMikeS
17:02:07jhMikeS peak_meter_enabled = false;
17:02:07jhMikeSthat's why I ask cause I didn't assume that
17:02:43jhMikeS:D *big time laughing*
17:03:00 Quit thegeek (Read error: 104 (Connection reset by peer))
17:03:06 Join thegeek [0] (
17:03:17CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
17:03:17*jhMikeS somehow inadvertently pasted something
17:03:17 Quit theli_ua ("Leaving")
17:03:31 Quit Zagor ("Client exiting")
17:03:32Mikachui put -mno-long-calls in the Makefile in apps/codecs/Tremor
17:03:34Mikachuand it compiled
17:03:45Mikachu106196 2006-08-23 17:02 apps/codecs/vorbis.codec
17:03:50Mikachu99900 2006-08-23 17:01 apps/codecs/vorbis.codec
17:03:59Mikachubefore and after, contrary to the timestamps
17:04:05Mikachu(on ipod nano)
17:04:10Mikachuand it's playing music happily
17:05:30Mikachui don't know if it will explode if i try to do something else?
17:05:45Bagderonly if there's code in iram
17:05:48 Quit PaulJam (".")
17:06:16 Quit dwihno (Read error: 104 (Connection reset by peer))
17:06:29Mikachuah, iram isn't used for tremor on ipods
17:06:48Mikachulet me try that again
17:06:53Mikachuiram isn't used for code on ipods
17:07:10preglowi'd really love to see that long call problem fixed
17:07:18preglowit's annoying
17:07:22Bagderso why do we still need the long calls again?
17:07:26preglowiram code
17:07:31Mikachuonly one function or two needed to be un-icoded to let you enable icode
17:07:40Mikachubut i didn't notice any performance improvement
17:07:56Bagderah right, I recall linuxstb measuring some little speed diff
17:07:57preglowBagder: also, on 64 megabyte models, the plugin and codec plugin ram areas are past +32 megabytes
17:08:06Mikachudoes iram code outweigh long calls on ipods? i thought it only mattered for data
17:08:17Bagderwell, the plugin ram area could easily be moved
17:08:26preglowit matters _a bit_ for code, but not much at all
17:08:30preglowBagder: how?
17:08:43Mikachucan't you get gcc to only use long calls when jumping between iram and normal ram?
17:08:55preglowMikachu: it's difficult, apparently
17:09:00Bagderpreglow: by editing the .lds file and fitting it somewhere near the start of memory
17:09:02Mikachuyeah i guessed :)
17:09:11 Join webguest47 [0] (
17:09:18dan_aMikachu, preglow: AFAIU IRAM code might have a bigger effect on 3G iPods than others
17:09:21preglowBagder: i suggested that, and amiconn protested, saying it'd break plugin comatability between rockbox compiles
17:09:26Mikachui tried to disable icode globally for fun but it's hardcoded in lots of places in firmware/*
17:09:27preglowBagder: which it does, but i'm not really sure how much i care
17:09:46Bagderpreglow: but both 32 and 64 could have them in the same spot
17:09:50Bagderand besides
17:09:59preglowBagder: yes, but the address would change everytime we added/removed code
17:10:00BagderI'd rather support 64MB than have them binary compatible
17:10:07Mikachuis it really that hard to recompile?
17:10:16Bagderpreglow: not necesarily if we set it at a fixed address like 0x200
17:10:29preglowBagder: true, but that would mean wasting some bytes somewhere :)
17:10:40Bagderyeah, but only very few
17:10:44preglowBagder: but yeah, i'd really, really like that
17:10:53preglowBagder: all code and data in the same area of ram
17:10:55preglowplus the iram issue
17:10:56amiconnThe question is how that would work
17:11:03preglowbut that narrows it down fairly well
17:11:11amiconnThe plugin buffer size is fixed, so the waste is fixed
17:11:31amiconnBut if you put the plugin buffer behind rockbox, the address will vary with binary size
17:11:32Mikachuare long calls only bigger, or slower as well?
17:11:34 Join dwihno [0] (n=dw@
17:11:50preglowamiconn: he suggested putting it _before_ rockbox
17:11:55preglowin which case it shouldn't
17:12:03preglowMikachu: both
17:12:16Bagderyes, it should be first in memory, I mean as early as it can without interfering
17:12:17amiconnpreglow: How would you put it *before*
17:12:17Mikachuso in the meantime, we could put -mno-long-calls in tremor?
17:12:24Mikachu(for ipods)
17:12:25preglowMikachu: a long call is three instructions, totalling 16 bytes
17:12:26amiconnThe vectors are part of the image 0x200 is *within*, not before
17:12:47Bagderso what?
17:12:56amiconnThen the question arises how to load that thing
17:13:01Bagderwhy is 0x200 or 0x100 or 0x60 vad?
17:13:22Bagderhm even
17:13:36Bagderyou'd either get a huge binary or we'd have to update the loader
17:13:36preglowMikachu: and yeah, a short call is one instruction, 4 bytes
17:13:39amiconnEither we need to bloat the binary with 1MB of zeroes, or we need to relocate one part of the binary
17:14:07Bagderif we make it load compressed files, 1MB of zeroes won't be that bad! ;-P
17:14:08preglowbut anyway, this might not be an issue
17:14:17Mikachuokay, you don't want to run rm Tremor/* in the apps dir instead of build/apps dir
17:14:19preglowall calls into and out of plugins/codecs are done through pointers, right?
17:14:42Mikachuhm, i didn't
17:14:53amiconnPointer calls are always long calls
17:14:59preglowi know
17:15:03Bagderamiconn: but a first test would of course bloat the binary, and voila, we can use short-calls
17:15:07preglowin which case it's really irrelevant
17:15:32amiconnBagder: We can use short calls for almost everything right now, without relocating the plugin buffer
17:15:34preglowthe only issue we're facing is still iram/ordinary codes
17:15:43amiconnThe problem is iram <-> dram calls
17:16:16Mikachuwhich platforms are affected by this issue?
17:16:32preglowipod video 64mb
17:16:36preglowbut it's not affected after all
17:16:38Bagderstill, moving the area is still interesting for the auto-detecting 32/64 MB case
17:17:45jhMikeSThe H120 can't record radio?
17:17:50preglowjhMikeS: sure it can
17:17:51amiconnThe gcc solution leaves a lot to be desired
17:18:18amiconnYou can specify which functions must be long-called, but (1) it doesn't always work and (2) it's the wrong approach
17:18:31 Quit [San] (Read error: 110 (Connection timed out))
17:18:40amiconnFunctions need to be long-called if the call goes to a different code section
17:18:55jhMikeSok it's H100 series
17:19:01Bagderyeah, but gcc doesn't know sections at that time, does it?
17:19:12Bagderor even so, most calls can be short even between code sections
17:19:20tucozlinuxstb, is it safe to assume that the 3g and the 4g ipods use the same key presses throughout rockbox?
17:19:29jhMikeSso everything with uda1380 records radio
17:19:52preglowamiconn: why is the flagging functions as long call a bad solution?
17:19:55preglowamiconn: if it worked, it'd be great
17:20:03merbanan_preglow: did you ever try the ffmpeg mp3 decoder ? if so try with the updated code, there are some arm patches pending also
17:20:07preglowmerbanan_: no
17:20:30 Quit merbanan_ (Remote closed the connection)
17:20:31amiconnpreglow: Thing is, a function is never always short-call nor always long-call
17:20:37amiconnIt also depends on the caller
17:20:44preglowamiconn: sure, no, but that's impossible for gcc to deal with
17:20:48preglowgcc doesn't know each function's section
17:21:00Bagderbecause the linker does that magic
17:21:06amiconnIt outputs the asm source, which contains the section names
17:21:08BagderI mean position it
17:21:37preglowamiconn: well, i've tried that anyway. gcc throws a fit at me if i try to tell it that a function is in another section
17:21:45preglowamiconn: you need to do it in the headers for it to see it properly
17:21:49amiconnI think it should be possible to tell gcc: Use long calls when going .text->.icode or .icode->.text, and short calls otherwise
17:22:19tucozlinuxstb, if I grep the code for IPOD_3G_PAD, I see either if (CONFIG_KEYPAD == IPOD_4G_PAD) || (CONFIG_KEYPAD == IPOD_3G_PAD) or#elif (CONFIG_KEYPAD == IPOD_3G_PAD) || \. Which I assume means that the keys are the same for both.
17:22:26Bagderyes, it would be nice to be able to specify it for specific boundaries like that
17:22:32preglowand it should be possible
17:22:38preglowby marking functions ICODE_ATTR in the headers as well
17:22:45preglowbut gcc doesn't like that at all
17:23:31preglowlike i've said before, we need a gcc-maintainer that loves us :>
17:23:33 Join Rondom [0] (
17:23:51Bagderfor all archs we use...
17:23:53dan_atucoz: As far as I know they are, and when I've come across things that don't compile on the 3G due to the keypad not being defined I've been able to use the 4G definitions
17:24:41preglownow, if gcc just wasn't so extremely large and hairy
17:24:51Mikachudo 3G have the separate 4 buttons and all 4G have the click wheel?
17:25:05Bagder"the biggst CVS repo on the planet"
17:25:06tucozdan_a, good. Then I can simply define IPOD_4G_PAD in the platform file. Unless I have to run through the entire manual and add a IPOD_3G_PAD for all tables.
17:25:13Bagdergcc that wsa
17:25:25Mikachuand now it's the biggest svn repo?
17:25:41tucozdan_a, I looked in button.h, and the even there the keys were defined equal.
17:25:46jhMikeSfor short call you mean generating 8 or 16 bit displacements for JSR (on CF anyway)?
17:25:46Mikachui would have thought kde or something is bigger
17:26:37BagderI guess kde is split up in many
17:26:55Bagderbut then, I guess nobody knows for sure who's the biggest
17:26:56dan_atucoz: AFAIK, the buttons are the same absolutely everywhere
17:27:25 Quit barrywardell ("CGI:IRC (EOF)")
17:28:09amiconntucoz: Some places might want to assign buttons differently because of the different mechanical layout
17:28:20amiconnIt's the same as with H1x0 and H300
17:28:33jhMikeSok, I asked a dumb question...:) Some here definitely know that.
17:28:34amiconnSame main buttons, but in different mechanical configuration
17:29:05tucozamiconn, I thought of that. I meant this as a start.
17:29:11amiconnjhMikeS: This short/long call thing here is arm specific
17:29:33tucozhmm. this will take some time I noticed.
17:29:40dan_atucoz: If you do a first draft of the 3G manual assuming the buttons are the same, I'll check that things behave as it says
17:29:43jhMikeSamiconn: So the Coldfire compiler optimizes the JSR size automatically?
17:29:48 Quit webguest47 ("CGI:IRC")
17:30:06preglowjhMikeS: yes
17:30:11preglowamiconn: but yeah, gcc knows for which section it is creating code when it compiles, so all we need to be able to do is mark functions as ICODE_ATTR in the header files
17:30:17tucozdan_a, ok. I just noticed that I probably will not have time to do that today. Maybe tomorrow.
17:30:42preglowamiconn: now, if the gcc developers had a habit of listening to bug reports/feature requests with some merit, this shouldn't be a big issue, i suspect
17:30:46jhMikeS...and he looked upon it thought it was good...
17:31:19amiconngrmph, indeed :(
17:31:41Mikachusend them a couple of ipods
17:31:49preglowit already knows how to create a call based on which section code is in, i think
17:32:14preglowthey don't seem to care too much about their embedded targets at all
17:32:42dan_aDoing Mikachu's test shaves 2k off the size of mpa.codec, but I've not tested whether it will play music yet
17:32:54preglowit probably won't
17:32:58preglowmpa.codec uses iram code
17:33:14amiconnYeah, and if you are going to try and fix things yourself, and ask questions, they seem to assume the gcc code is self explaining, and you know how things work and where a fix has to go :(
17:33:29 Nick t0mas_ is now known as t0mas (
17:33:30dan_aThere was only one ICODE_ATTR definition
17:33:34*amiconn remembers his sh gcc bug hunting :((
17:33:39tucozdan_a, is it possible to set contrast on the 3g?
17:33:46dan_atucoz: It is
17:34:04Mikachuthe man page for gcc says "long calls will always be enabled for functions marked with the section attribute" (assuming you're using -mlong-calls)
17:34:08tucozthen I'll just follow what is possible on the 4g
17:34:36Mikachuyou could sprinkle a _lot_ of #pragma long_calls_off over the code :)
17:35:46Mikachucouldn't you define a function call to insert a pragma before and after to enable/disable long calls?
17:36:19Mikachuhm that would probably not work
17:43:37***Saving seen data "./dancer.seen"
17:47:34 Part pi_
17:56:05jhMikeSuda1380 on H100/120/300 _HAS_ to have recording enabled just to listen to radio?
18:02:29 Quit pondlife ("byebye")
18:03:45 Join dwihno_ [0] (n=dw@
18:03:45 Quit dwihno (Read error: 104 (Connection reset by peer))
18:04:04 Join Paul_The_Nerd [0] (
18:05:21preglowjhMikeS: not at all
18:05:43preglowjhMikeS: the recording subsystem has nothing to do with radio apart being able to use it as a recording source
18:06:49jhMikeSumm...wait a sec...
18:07:08 Join linuxstb_ [0] (n=linuxstb@
18:07:17jhMikeSpreglow: the radio.c code was doing this:
18:07:27jhMikeS uda1380_enable_recording(false);
18:07:27jhMikeS uda1380_set_monitor(true);
18:07:36jhMikeSjust to play it
18:07:40preglowi suggest you check out the first function
18:07:46preglowand you'll find out what the parameter means
18:07:51preglowit's not what you think
18:08:03 Part Paul_The_Nerd
18:08:08jhMikeSdid have a peek at it...
18:08:11linuxstb_dan_a, tucoz: I setup separate IPOD_3G_PAD and IPOD_4G_PAD definitions because even though the 3G has the same buttons, they are physically in different places - so I didn't want to assume that they would always be the same.
18:08:24preglowfalse means use radio as source, i think
18:08:38tucozlinuxstb, ok. I think I go for that option myself. grep is my friend :)
18:08:55preglowand the monitor function just mixes the input with the current output in the uda1380 chip
18:09:55jhMikeSit says "record from line-in" it seems they left recoding monitoring enabled when playing radio.
18:10:12 Join solarflare [0] (
18:11:10jhMikeSfor tlv320 the analog bypass was set
18:11:17preglowhmm, i think making the uda1380 actually send audio samples via the i2s bus is really overkill
18:11:22 Join suge [0] (
18:11:41sugecan I upgrade to another build just by deleting the .rockbox folder?
18:12:17preglowsuge: yes, that and installing a new one complete with a rockbox.* file
18:12:33preglowsuge: but really, you don't need to delete it first
18:12:34Mikachuand not deleting the old one
18:12:39preglowsuge: you can just overwrite the old files with new ones
18:13:00sugeso i got the piesjulies jpod zip... just extract it to root and I should be good?
18:13:08 Quit dwihno_ (Read error: 104 (Connection reset by peer))
18:13:19 Join dwihno [0] (n=dw@
18:13:35jhMikeSpreglow: Do you think it's a hardware nescessity or for another reason (I can't think of one)?
18:14:44preglowi can't, no
18:14:55preglowbut i'm not _that_ familiar with the recording code
18:15:03preglowmy only involvement with it is hacking in spdif support
18:16:36jhMikeSI'm trying to get the input selection correct before posting up that patch...
18:16:43pregloware all the posts in the old forums going to be moved?
18:18:46peturI think so, yes
18:18:56preglowthat's quite a bit of work...
18:18:59jhMikeS...and it left line-in/I2S on _after_ turning off the radio...
18:19:11preglowjhMikeS: that's nasty
18:20:23jhMikeSdatasheet time!
18:24:40 Join mantono [0] (
18:28:21sugei can't seem to find this anywhere... on the pijulius build it supports album art.. is that via an id3 tag or picture in the folder?
18:28:58 Quit KN|stiff (Read error: 104 (Connection reset by peer))
18:31:40 Quit petur ("here today, gone tomorrow")
18:38:07MikachuBagder: you've been slashdotted
18:42:19amiconnpreglow: I changed the recording monitor to pass the data digitally through the i2s bus on iriver, for 2 reasons:
18:42:44amiconn(1) It simplifies spdif monitoring vs. analog in monitoring
18:42:58amiconn(2) It's more 'what you hear is what you get'
18:43:16 Part linuxstb_ ("Leaving")
18:43:59amiconnAnalog bypass monitoring might still sound fine while the actual recording is already clipping
18:44:11amiconnWith i2s monitoring, you will hear it.
18:45:02amiconnThe radio could also go through i2s.
18:45:06preglowyeah, that's nice
18:45:21preglowi was just wondering why just entering the radio screen switches on recording
18:45:26preglowoh right
18:45:36amiconnIt does so on archos as well
18:46:16amiconnIt's a thing that should be changed, as we could even shut down the mas dsp core while listening to radio, unless prerecording is enabled
18:50:36 Join solarflare_ [0] (
18:51:01jhMikeSamiconn: If you're only listening to radio, why not just use analog bypass? I have things working differently now: select the source (Mix,Line,SPDIF,FM) and whether to use record/playback mode.
18:51:31jhMikeSOnly for SWCODEC though
18:51:41 Quit dan_a (Read error: 110 (Connection timed out))
18:51:42amiconnProbably switching between bypass and i2s will cause a pop?
18:52:08preglowamiconn: yeah, exactly
18:52:19jhMikeSCan always mute sound for a moment...this works fine for tlv320
18:53:25*amiconn is watching fullscreen m2v at 12.6fps on his h300 :)
18:54:03preglowany new changes?
18:54:10*Genre9mp3 just drops a wide smile!
18:54:27amiconnasm lcd_yuv_blit() based on jhMikeS' code for X5
18:54:36jhMikeSamiconn: with sound? x5 gets 15.3 w/o sound.
18:54:52jhMikeSamiconn: you think emac would help?
18:54:59amiconnSince when does mpegplayer support sound?
18:55:04 Join midgey34 [0] (
18:55:09 Join Crackerizer [0] (
18:55:22jhMikeSamiconn: I didn't think it did...wonder if you had it going.
18:55:32amiconnemac won't help for yuv I think
18:55:47amiconnI have some extra tweaks on top of your ideas
18:55:52CrackerizerPlease help me. There's something wrong with my ipod 3g.
18:56:08amiconnWill of course port them to X5
18:56:45jhMikeSamiconn: do tell...I still might try the emac just for kicks and to learn it...if it gives any resuls then...
18:56:48amiconnjhMikeS: Btw, fullscreen on h300 != fullscreen on X5. With Mikachu's 176x128 video I get 16.7fps
18:56:52CrackerizerThe directory structure is gone. As i can't use usb nor firewire, I cant use the updator to restore my ipod.
18:57:07amiconn(that jpop thing)
18:57:28preglowjhMikeS: just look for long chains of a += b*c, and you'll emac food
18:57:32Genre9mp3amiconn: X5 screen is 160x128 not 176x128
18:57:35preglowyou'll have
18:57:47jhMikeSemac 0xf00d
18:57:52amiconnGenre9mp3: I know
18:58:09CrackerizerIs there someone here own a 3g ipod?? help me please.
18:58:09jhMikeSIt's more a += b*c >> n
18:58:27preglowjhMikeS: that too works fine
18:58:32 Quit XavierGr ()
18:58:52preglowjhMikeS: depends if you're talking 64 bit or 32 bit wide multiplies
18:59:13 Quit solarflare (Read error: 110 (Connection timed out))
18:59:18*amiconn compares speed vs. the old code now
18:59:18 Join XavierGr [0] (
18:59:20jhMikeSpreglow: I keep the match to 16 bits
18:59:52jhMikeSto use lsr.l #8,Rx
19:00:22amiconnjhMikeS: Btw, I checked the conversion matrix coefficients, and two of them were off by one
19:00:26saratogaCrackerizer: what do you see on the 3G when you mount it in your OS?
19:00:39amiconnOne of them is 128 now, allowing to optimise one muls.w away completely
19:01:33jhMikeSamiconn: I hand checked the results to get pure full RGB colors long as it doesn't cause more saturation it will help.
19:01:46Crackerizersaratoga: I can not mount my ipod to my computer. because the use and firewire do not work.
19:02:22jhMikeSYou can also do for example: x*129 = (x << 7) + x
19:02:36amiconnThat will be slower
19:02:48jhMikeSmuls seems to be pretty fast though
19:02:48Crackerizersaratoga: the only way to update songs to my ipod is to hook it's HDD directly to my computer with an IDE to 1.8 inces HDD adapter.
19:02:58preglowmuls is pretty fast on emac capable chips
19:03:00amiconnA mul[s|u].[w|l] takes 4 cycles on our coldfires
19:03:10saratogaCrackerizer: If you canot mount it, then its probably dead
19:03:22saratogacontact Apple about a replacement
19:03:32amiconncopy, shift and add are 3 cycles - and you need an extra register
19:04:39 Join JoeBorn [0] (
19:05:11Crackerizersaratoga: the HDD is fine. I just need the struture and configs files of an ipod 3g. because i can not use the updater.
19:05:22amiconnIt looks like this yuv conversion can be made extremely efficient on arm
19:05:31 Join dan_a [0] (n=dan-mirc@
19:05:35jhMikeSamiconn: I think there were some extra at that point. 3 vs. 4 sounds good :)
19:05:36amiconnIts shifter will help a lot
19:06:10CrackerizerIs it possible to boot rockbox without ipod directory structure and config files?
19:06:13saratogaCrackerizer: Adding files isn't going to help if you can't mount the disk
19:06:23jhMikeSamiconn: what did you change the matrix to?
19:06:37amiconn * |R| |74 0 101| |Y' - 16| / 256
19:06:37amiconn * |G| = |74 -24 -51| |Cb - 128| / 256
19:06:37amiconn * |B| |74 128 0| |Cr - 128| / 256
19:06:51Crackerizersaratoga: I can copy any files to my ipod's HDD.
19:07:02saratogabut you cannot mount it, correct?
19:07:13saratogaor rather, mount the ipod itself via USB
19:07:55amiconnWow, the speedup is almost constant 3 fp irrespecive of the video size
19:08:02preglowwith the first bars signifying a matris and the second bars abs() ?
19:08:08preglowright, vector
19:08:23amiconnThat means +23% at 176x128 and +31% at 224x176
19:08:30jhMikeSyou changed all the chroma ones
19:08:31Crackerizergaratoga: why is mounting important?
19:08:46Crackerizerhello dan_a.
19:08:54amiconnAh, yes
19:09:08amiconnThat's what I use on h300 now, it's looking good
19:09:14saratogaCrackerizer: because if you can't mount, then the Ipod is probably dead
19:09:21amiconnIt's what proper rounding produces
19:09:31dan_aHi Crackerizer:
19:09:32jhMikeSamiconn: why not change the R/B for 5 bits on iriver? That's for 6 bits.
19:09:51amiconn[13:57:40] <amiconn> Direclty converting to RGB565 doesn't work well. Results in a greenish image due to roundoff errors
19:09:57amiconn[13:58:07] <amiconn> So I'll convert to RGB666, then drop one bit from R and B
19:09:58Crackerizerdan_a: can you boot rockbox on 3g without any original ipod file and directory structures?
19:10:24preglowemac will work for yuv->rgb, it seems
19:10:29 Join Massa [0] (n=Massa@
19:10:32Crackerizersaratoga: no, it was working fine even the usb and firewire dont work.
19:10:45dan_aI think so, but I've never tried it
19:10:58saratogahow is it working fine if the USB doesn't work :)
19:11:13jhMikeSamiconn: I didn't want to do proper rounding...just truncation. The appearance was excellent w/o it on x5 anyway.
19:11:23saratogawait, did you ever have USB working on your PC? could be a problem with your PC not the ipod
19:11:52Crackerizersaratoga: I mean it could play any song i copy to my ipod using 3rd party tools. :)
19:12:20Crackerizersaratoga: No, my pc is fine too. I have both usb and firewire port.
19:12:29saratogathat doesn't make any sense
19:12:29tucozdan_a, there. Now you can build the manual for the ipod 3g :)
19:12:39saratogahow could you add files if you couldn't mount the disk
19:12:44dan_atucoz: Brilliant!
19:13:09tucoznow I need some food, after all that search and replace
19:13:11amiconnjhMikeS: The problem with direct-to RGB565 conversion is probably twofold: (1) The coefficients for R and B become smaller, so the roundoff error becomes bigger. (2) The resolution for G is different from R ab B.
19:13:37Crackerizersaratoga: there are tools such as retune, ephpod etc. which can be used to copy songs to my ipod HDD and sync the ituneDB.
19:14:02Crackerizerdan_a: could you do me a favour, please?
19:14:06dan_aI've tried the same thing as Mikachu (taking out ICODE_ATTR statements and setting mno-long-calls) and the codecs I've tried still work (MP3, WAV, SID) and the codecs shrink by a bit
19:14:12dan_aCrackerizer: What's that?
19:14:14 Join Rondom_ [0] (
19:15:24saratogaCrackerizer: ephpod requires you to mount the ipod . . . you said you couldn't mount it
19:15:49dan_aI get a lot more skipping with no ICODE_ATTR statements, though
19:16:36amiconnOn PP5002 you will
19:16:40 Quit suge (Remote closed the connection)
19:16:42amiconn...because of the broken cache
19:17:17preglowdan_a: the mp3 codec still works?
19:17:22dan_aamiconn: I'd been told that, but I wanted empirical evidence!
19:17:31amiconnOn PP5002 icode makes sense, on PP5020+ it's debatable
19:17:36dan_apreglow: Yes, if you take out the ICODE_ATTR statements
19:17:45preglowdan_a: ahh, yes, sure
19:17:49preglowthen small wonder, heh
19:17:56preglowno reason they shouldn't work then
19:17:57Crackerizerdan_a: could you send me your ipod 3g file structure. Especially, the iPod_Control\Device folder.
19:18:41dan_aamiconn: When I tried semi-randomly adding in ICODE_ATTR statements (to small functions which I guessed would be called often), it made it skip more because there was less codec buffer
19:18:48dan_aSo I just wanted to be sure
19:18:57amiconnjhMikeS: Doing the extra RGB666->RGB565 conversion doesn't make it slower. It's only 2 extra shifts, but they're accounted for by the simpler clamping
19:19:01CrackerizerWell, I mount my ipod hdd with an USB to 1.8 inces HDD adapter.
19:19:36Crackerizeri have to open my ipod and take its hdd and then hook it with the adapter.
19:19:52dan_aCrackerizer: I will do, but I have to go out right now. I'm supposed to be 20 miles from here in 10 minutes and the main road is closed
19:19:56jhMikeSIs that claping code in CVS now?
19:20:18CrackerizerAnytime dan_a. Thank you for your help.
19:20:18amiconnNot yet. I'm just diffing to see what I did, then commit
19:20:30 Nick dan_a is now known as dan_a_away (n=dan-mirc@
19:20:45amiconnProbably have to add a couple of dirs to cvs first - lcd-as-h300.S is starting the h300 target tree...
19:21:46jhMikeSwhat, cvs up -dP won't do it? :?
19:22:22amiconn_I_ have to add them
19:22:37jhMikeSyeah, *duh*
19:23:59 Quit Rondom (Nick collision from services.)
19:24:03 Nick Rondom_ is now known as Rondom (
19:29:58 Quit dionoea_work ("CGI:IRC")
19:31:45 Quit Lear ("Chatzilla 0.9.74 [Firefox]")
19:33:23*jhMikeS is trying to guess the clamping code and can see a possible approach w/o bra
19:33:51 Join TeaSea [0] (
19:34:38CrackerizerIt seems like the original firmware on the firmware partition is corrupted.
19:34:52amiconnIt uses bra, but only one per colour component
19:35:03amiconn(and the overall check in front, which is pretty efficient)
19:35:14CrackerizerThis is unlucky if you can not restore your ipod with updator tool.
19:35:14*jhMikeS thinks you don't need bra for low clamp
19:35:35jhMikeSlsr.l / add.l #1 / and.l
19:35:39amiconnJust take a look at my solution...
19:35:45jhMikeSgetting it now...
19:36:21jhMikeScvs up is taking its sweet time
19:36:40amiconnDon't forget the -dP ;)
19:37:22*jhMikeS didn't forget the -dP !
19:38:34preglowamiconn: did you try to not use iram for the mem* functions? i'm curious if it's really something that needs to be there
19:39:10amiconnIirc using IRAm had a tiny speed advantage on coldfire
19:39:27preglowyeah, i'm talking coldfire
19:39:32preglowsh probably benefits much more
19:39:40amiconnIf we have better use for the IRAM and want to take them out, we'd need to do it in a target specific fashion
19:39:52amiconnYes, sh benefits significantly
19:40:05tucozBagder, the ipod 3g manual is now ready to be added to the daily builds.
19:40:55*amiconn broke the h300 sim build :/
19:41:26amiconnrofl, x86 doesn't understand coldire asm
19:41:42preglowhaha, for some reason
19:41:44Mikachuwho would've thought
19:43:04preglowamiconn: yeah, just memcpy is almost 1kb, i'd say those are prime candidates to remove if we ever need more iram
19:43:38***Saving seen data "./dancer.seen"
19:43:52amiconnI know. memmove adds about the same again
19:44:22amiconnThere are 2 options, depending on how much iram we need, and how the speed impact is
19:45:01preglowthe arm memset probably doesn't need iram either, and we're a lot tighter on iram there
19:45:12preglowfor data, that is
19:45:37amiconn(1) Move the whole thing to dram. (2) Undefine FULLSPEED in the asm source. This will slow down word aligned copies (long aligned ones stay as fast as they are, as well as byte aligned ones), but also reduce size by > 1/3
19:45:48 Join Doomed [0] (
19:46:00preglowi say just letting FULLSPEED stay and moving to dram instead is the best thing to do
19:47:33 Join mikearthur [0] (
19:47:51 Part tucoz ("Leaving")
19:48:10 Quit Rondom ("Ex-Chat")
19:48:28Doomeddoes rockbox recognize album art in the tag?
19:52:12MassaDoomed: no, it does not (to be honest, it does not even detect album art as separate file - without a patch)
19:52:13preglowrockbox doesn't recognize album art at all
19:52:59Doomedwell i use this optimized build for H120 so it has an album art patch, but thanks for the heads up
19:54:08MassaDoomed: and I'm sure it'll take some time until rockbox will be able to read the corresponding id3 tags - if ever
19:55:07amiconnCalling images in a tag 'album art' is really misleading imo
19:55:24amiconn'track art' makes more sense
19:55:24jhMikeSok, I see how that saturates low (and with 0x00 if low) (spl does what those lsr/add would've done) but I don't see how anding the component with 0xff saturates high.
19:55:55amiconnThe trick is the dual use of the cmp result
19:56:21Massaamiconn: Yes, I always found it confusing - actually it's often called "file cover art"
19:56:38amiconnIf the component is in range, C or Z will be set, so bls.s skips the whole rest
19:56:40Doomedwell i guess i should start saving it as the album art in folder then
19:57:01preglowisn't the album art duplicated in all files if you don't?
19:57:04preglowsounds really stupid
19:57:42jhMikeSoh's already lt 64...
19:57:49MassaDoomed: don't forget to save it as BMP with corrseponding size ;)
19:58:00MikachuThere may be several pictures attached to one file, each in their individual "APIC" frame, but only one with the same content descriptor. There may only be one picture with the picture type declared as picture type $01 and $02 respectively.
19:58:09amiconnIf it's out of range, both C and Z are unset. Then spl.b evaluates the N flag, and sets the lower byte to 0 when negative, and to 0xff when positive
19:58:10Massapreglow: yes it is - but that's how it is with that ID3 tags ;)
19:58:30Doomednah man, jpeg, it shows up in rockbox
19:58:50Mikachu is the relevant part of the id3 standard i think
19:58:57amiconnThe and.l with 63 is just to bring the 0xff to the correct upper bound (and map our eventually higher bits, since there's only spl.b)
19:59:16MassaMichachu: and they may be in JPG or PNG format...
19:59:24Mikachui still love $11 A bright coloured fish
19:59:43*Doomed wants a 60gb video but cant let go of H120
19:59:45MikachuMassa: "should" be png or jpg, could be anything
20:00:33 Join Rondom [0] (
20:00:38MassaMikachu - you're right - as far as I remember there was something about "interoperability"...
20:00:40 Join eli_sherer [0] (n=eli@
20:01:02 Quit Crackerizer ("CGI:IRC (Ping timeout)")
20:01:21Mikachuooh, here's something interesting
20:01:23Mikachu4.20. Linked information
20:01:31MikachuTo keep information duplication as low as possible this frame may be used to link information from another ID3v2 tag that might reside in another audio file or alone in a binary file.
20:01:50eli_shereramiconn: will this function help optimizing the jpegviewer (it uses YCbCr as well...)?
20:01:52MikachuThe "AENC", "APIC", "GEOB" and "TXXX" frames may be linked with the content descriptor as additional ID data.
20:01:57MassaMikachu: and the picture types have defined meanings...
20:02:13amiconneli_sherer: jpeg uses a slightly different YCbCr format
20:02:49amiconnAnd I don't see why the conversion needs major optimisation in the jpeg viewer
20:03:00eli_shereramiconn: You've really done your homework :)
20:03:14Mikachuwhat i want in the jpeg viewer is the ability to start a zoom while decoding
20:03:20Mikachuie to interrupt the current decoding
20:03:40eli_shereri don't understand why the firmware's jpegviewer is faster then ours...
20:03:44MassaMikachu: about link information: is there any program known which is able to generate and _use_ such tags?
20:03:53 Join webguest03 [0] (
20:03:55MikachuMassa: no idea, want to make the first? :)
20:04:02Doomedhopefully this source updates to the latest cvs with no problem
20:04:08amiconnMy experience on ipod is that decoding is blazingly fast
20:04:38 Part lorijho ("Kopete 0.11.3 :")
20:04:38 Quit webguest03 (Client Quit)
20:04:40amiconnMaybe I should try larger images
20:04:48 Join GabrielArellano [0] (
20:04:50Mikachui have a map of the city, it takes a while to open
20:04:56 Join Crackerizer [0] (
20:04:59MassaMikachu: no - not until at least JPG decoding is part of rockbox itself ;)
20:05:00Mikachuactually i had to split it in 4 because it was too large decoded
20:05:01pregloweli_sherer: which device?
20:05:02eli_shereripod can decode really fast can scroll through your pictures with the wheel and the pictures will fly
20:05:09jhMikeSthat spl.l does similarly to a sequence I'd use on the x86 for the same effect.
20:05:17eli_shererpreglow: iriver...(sorry!)
20:05:20pregloweli_sherer: i thought itunes resizes the jpegs you transfered
20:05:26Mikachuit does
20:05:28Mikachuyou can't zoom
20:05:29Doomedwho uses itunes
20:06:05jhMikeSSo...analog bypass in OK for iRiver radio or not?
20:07:04GabrielArellanoHello, I need write access to the Wiki
20:07:16 Quit Crackerizer (Client Quit)
20:08:21Doomedsomehow i need to get $400
20:08:30eli_shererWe all do!
20:09:00jhMikeSamiconn: now you've got a .size directive in there that wasn't there before.
20:10:21amiconnMonochrome target JPEG decoding speed comparison, test image 658x900 (full zoom): iPod Mini g2: 1.00s, Iriver H140: 0.88s, Archos Recorder: 20.64s
20:10:22Doomedhmmm how long does this updating cvs taeke
20:11:07 Quit GabrielArellano ("CGI:IRC (EOF)")
20:11:27amiconnSH1 suffers from not having a shift-by-n instruction here. The huffman decoding stage would profit from it
20:11:48preglowit not having that _still_ makes me chuckle
20:12:04pregloweverything should have a barrel shifter
20:13:00 Join Crackerizer [0] (
20:13:23 Join pike^lapt [0] (i=amiga@
20:14:53jhMikeSLooking for a final word (please): it's ok for analog radio playback on Hxxx? It's just a "pop" issue iirc?
20:15:20amiconnIt's what's used currently afaik, so at least it won't make it worse
20:15:30amiconnI would prefer i2s though
20:16:03jhMikeSamiconn: I can just leave it that way for uda1380 then...
20:17:53jhMikeSfor normal audio playback this is correct:
20:17:55jhMikeS uda1380_disable_recording();
20:17:55jhMikeS uda1380_set_monitor(false);
20:18:28 Quit Febs (Read error: 110 (Connection timed out))
20:22:51jhMikeSWell...ok...I'll put the updated recording patch in the tracker soon and you can tell me if everything goes haywire. :)
20:24:28Doomedwhen u go to update ur source with the cvs does it just stay cvs up -dP or do u get lines under that?
20:27:52 Join sharpe [0] (
20:28:23MikachuDoomed: it's supposed to print a whole lot of stuff and then drop you back at the command line
20:30:21 Quit pike^laptop (Read error: 110 (Connection timed out))
20:30:35sharpeDoomed: How long have you tried waiting for it to start to update?
20:30:49Doomedjust before was like 3-4 min
20:31:09sharpeCan you do a 'cvs status' ?
20:31:30Mikachuwell, the rockbox server was slashdotted, could be related?
20:31:40Mikachunope, i can connect fine
20:31:47Doomedsee when i go to login, it downloads the cvs
20:32:01Doomedk ill let it download to rockbox-devel
20:32:05Doomedthen i cd to my source
20:32:10Doomedthen cvs up -dP
20:32:28sharpeUpdates fine for me...
20:33:38 Join fatherfork [0] (
20:33:52Doomeddamn i couldve used that video card :(
20:34:40Doomedo shit do i have to compile the source first then update with cvs?
20:36:20Doomedk now the cvs downloaded to rockbox-devel
20:36:30Doomednow i cd to my rockbox (source) folder right
20:36:40amiconnjhMikeS: The .size isn't strictly necessary but a lot of other asm sources have it, so I added it just in case
20:36:43sharpeYes. 'rockbox-devel'
20:36:59Doomedi have a source i downloaded
20:37:07Doomedbut the rockbox-devel is the cvs
20:37:29Doomedthis is what my C:\Rockbox folder looks like :
20:37:36Doomedrockbox rockbox-devel
20:39:30markuncd rockbox-devel
20:39:37markuncvs update -dP
20:40:05markunI thought that is what you wanted
20:40:15Doomedthats the cvs already
20:40:30markunwhat's the problem then?
20:40:51Doomedi have a downloaded source not from cvs
20:41:06markunyou can't update the source that is not from cvs
20:41:30Doomederr no considering ive done it before
20:41:35Doomedi just dont remeber what i did
20:42:07markunYesteday I tried to talk to someone with a similar problem and he walked away in anger
20:42:41markunSo perhaps I'm not the best person to help you
20:43:26markunwhy do you want to update yout non-cvs source tree? Did you edit some of the files?
20:44:14Doomedim only starting to compile sources so i have no clue what im doing
20:44:24markunok :)
20:44:41markunAnyone here with a T series iriver?
20:45:02sharpeI thought that said 'T shirt iriver'...
20:45:47fergieI just tried to do my first build with cygwin, but after loading it on my ipod, it hangs on the boot screen after rockbox loaded appears
20:45:47markunJust having some firmware update problems with my new T30
20:47:35sharpefergie: Are you sure you built Rockbox for the right iPod?
20:47:54 Join bbroke [0] (
20:48:09Doomedwhats the point of taring the source twice?
20:48:16sharpeAnd you extracted it and it froze after the bootloader?
20:48:31fergiei rebooted it now and now it get a bit further but it hangs again at the menu
20:48:45markunDoomed: you had a rockbox.tar.tar file?
20:48:53sharpefergie: Have you applied any patches?
20:48:58 Quit eli_sherer ()
20:49:01Doomedno i mean from the daily build page
20:49:23fergieyep the scroll margins patch
20:50:08markunDoomed: you mean the tar.gz extension?
20:50:18Doomedyea nm tho
20:50:24sharpeOkay, I doubt that's causing it, but could you try a build without it?
20:51:05markunDoomed: tar pack many files into 1 file (uncompressed) and gzip can compress 1 file
20:51:16 Quit XavierGr (Read error: 113 (No route to host))
20:52:27fergiei had to change some calls to lcd_setmargins in debug_menu.c
20:53:09 Join Febs [0] (n=medifebb@
20:53:56fergiewould doing cvs up -dP make any difference?
20:54:05 Quit Crackerizer ("CGI:IRC (EOF)")
20:55:06sharpeTry just unpatching it.
20:56:23fergiewill try a clean build from latest cvs
20:56:29sharpeThat works.
20:57:52Doomedthis is so confusing
20:59:35fergiebtw I asked it earlier this day but is there any change on the charging on iPod ?
20:59:57sharpefergie: Any change since when?
21:02:26fergiewell i mean does it still drain energy faster than it can charge when playing?
21:03:03sharpeIt doesn't seem to for me...
21:03:23fergiestrange, which ipod do you have?
21:04:01Mikachusharpe, fergie: do you charge via usb or wall charger?
21:04:29sharpeI chargeth via usb. :)
21:04:42Mikachuthere goes that theory
21:06:26fergiemy mp3 files are mostly lame vbr aps
21:14:03fergiehmm the nonpatched build also hangs :(
21:15:14sharpeStrange. Has this just recently started to happen?
21:15:28sharpeI mean, any times before your patched build?
21:15:33 Join barrywardell [0] (
21:17:14fergiethe one from the website works. This is my first own build ;)
21:17:35fergiewill have a look at the build warnings
21:18:04sharpeAh... well... yeah, those may have something to do with it if you get them...
21:19:43saratogaanyone seen this? :
21:19:45saratogaIn file included from thread.c:95:export/profile.h:57: error: conflicting types for ‘current_thread’
21:19:55saratogasorry that should have been two lines
21:20:32saratogaIn file included from thread.c:95:i'm trying to make a build with profiling support for the 3G from latest CVS
21:21:11amiconnI guess dan_a's multi-core changes broke profiling...
21:21:31saratogadidn't realize that had been commited
21:21:57saratogai suppose downloading CVS from a couple day ago would work around this problem for now?
21:25:53 Quit Arathis ("[rl_bot quit]")
21:27:16saratogaalso, should i file a bug report over this or contact dan_a?
21:31:52fergieplugin.c:440: warning: initialization from incompatible pointer type
21:33:06fergiecould that account for the hangs?
21:34:16fergieit's the only warning
21:34:30 Quit apo (Read error: 110 (Connection timed out))
21:34:43sharpeDoubt it, but it's worth a try fixing it.
21:34:58 Join apo [0] (
21:36:41 Quit barrywardell ("CGI:IRC")
21:43:16 Join doobeh [0] (n=me@
21:43:41***Saving seen data "./dancer.seen"
21:44:12doobehHi, I've installed Rockbox on my ipod, it seems to have locked up, how can I force a reset?
21:44:17 Join Davide-NYC [0] (
21:44:36doobeh(It's just locked up while creating the tag database, I think part of my hdd on the ipod is dead/dying)
21:45:35Davide-NYChotwire_: Hey Hotwire, how's the UISim remote code going?
21:45:49Davide-NYCI just put a feature request up on flyspray:
21:46:02Davide-NYCcheck my my corny animated GIF! :-P
21:47:59Davide-NYCThis feature request has to do with Conditional Background Image in UISim for the various iRiver Remotes.
21:50:06 Join murdock [0] (
21:52:41 Quit fatherfork ()
21:53:39 Quit MarcoPolo ("Bye !")
21:59:43 Join _hotwire_ [0] (
22:00:28 Quit Febs (Read error: 104 (Connection reset by peer))
22:05:09preglowbtw, is that left/right scrolling thing still in?
22:09:07LandusAfter I apply patches, I just have to cd back into the directory I made when I ran ../tools/configure, and make zip, right?
22:09:23LandusThen extract the stuff in the zip file into my DAP?
22:09:26 Join barrywardell [0] (n=barrywar@
22:10:59 Quit mikearthur ("Konversation terminated!")
22:12:59 Join mikearthur [0] (
22:13:32 Quit meshuga ("Changing server")
22:15:25 Quit hotwire_ (Read error: 110 (Connection timed out))
22:15:59 Join meshuga [0] (
22:21:13Davide-NYCquestion: I signed up for the rockbox-dev mailing list. I have yet to receive any emails submitted by devs (only the automagic mailman emails)
22:21:14Davide-NYCI noticed while brosing the web interface that I was not subscribed to any topics, yet I couldn't find an option to subscribe in the web form.
22:21:16Davide-NYCIf I email rockbox-dev-request with the word 'list' in the subject am I on the right track?
22:23:57 Join mirak [0] (
22:24:36jhMikeSshould the cpu be boosted only during recording or during simple line monitoring too? Right now it's boosted just because you're looking at the recording screen.
22:25:04 Join lowlight [0] (
22:25:35dionoeait should be boosted only when you need the extra power. does monitoring require the same amount of CPU as recording ?
22:26:08 Quit mirak (Remote closed the connection)
22:27:35jhMikeSI don't believe it does, there's no file writing or compression happening but there are DMA1 interrupts being serviced.
22:28:34 Join Drkepilogue [0] (
22:28:43*lowlight breathes easy at the sight of a green table :)
22:30:05 Join damaki_ [0] (
22:32:16 Quit Drkepilogue (Client Quit)
22:33:10preglowjhMikeS: it depends
22:33:21preglowjhMikeS: when doing spdif recording, it should always be boosted
22:33:31preglowat any other times, it should be boosted when it needs some extra juice
22:33:40preglowi don't know how it's currently implemented, though
22:38:16lowlightshould I commit my checkmarks for the current settings?
22:38:35 Quit murdock (Remote closed the connection)
22:39:49Davide-NYCguys, should I be receiveing emails fairly regularly from the rockbox-dev mailing list?
22:40:08jhMikeSpreglow: I suppose I'll leave the SPDIF boost alone then :)
22:40:56barrywardellDavide-NYC: the last one I got was on the 19th of july
22:41:25Davide-NYCthank you. I was totally confused
22:41:45BagderDavide-NYC: yes you should
22:41:53 Join RoC_MM [0] (
22:41:55lowlightrockbox-dev was busy today.
22:42:29Davide-NYCbadger: I'm not and it seems that I'm subscribed tot he list, but not to any topics
22:42:42Bagderthen you have a problem
22:42:49Bagdersince you get all or none
22:43:20Davide-NYCin other words there are no topics, just one big list
22:43:33Bagderyes it is a mailing list
22:44:10 Quit damaki (Read error: 110 (Connection timed out))
22:44:41Davide-NYCI re-subscribed
22:44:49*Davide-NYC fingers crossed
22:45:26 Join adiamas [0] (
22:45:50Bagderrockbox-dev has 284 subscribers
22:47:10Davide-NYCIt's not working for me, should I send an email to an admin?
22:47:25Bagder /msg me your email address
22:47:47BagderI am admin
22:50:11 Quit _FireFly_ ("Leaving")
22:56:06 Join BHSPitMonkey [0] (
23:00:46 Join mirak [0] (
23:01:38 Join _jhMikeS_ [0] (
23:02:18markunhi mirak
23:03:35 Quit meshuga (
23:03:35 Quit Rob2222 (
23:03:35 Quit dongs (
23:03:35 Quit pixelma (
23:03:35 Quit ShadowdogMU (
23:03:35 Quit chendo_ (
23:03:35 Quit Mikachu (
23:03:35 Quit cismo_ (
23:03:35 Quit gtkspert (
23:03:35 Quit netmasta10bt (
23:03:35 Quit Nibbier (
23:03:35 Quit ToyKeeper (
23:03:35 Quit Galois (
23:03:35 Quit Spida (
23:03:44NJoinmeshuga [0] (
23:03:44NJoinRob2222 [0] (
23:03:44NJoindongs [0] (
23:03:44NJoinpixelma [0] (i=pixelma@
23:03:44NJoinShadowdogMU [0] (
23:03:44NJoinchendo_ [0] (
23:03:44NJoinMikachu [0] (
23:03:44NJoincismo_ [0] (
23:03:44NJoingtkspert [0] (
23:03:44NJoinnetmasta10bt [0] (
23:03:44NJoinNibbier [0] (
23:03:44NJoinToyKeeper [0] (
23:03:44NJoinGalois [0] (
23:03:44NJoinSpida [0] (
23:05:31Bagderthat was in SanDisk's general direction
23:06:24 Quit mantono ("Don't worry, be happy")
23:07:49 Quit bbroke ("Chatzilla 0.9.75 [Firefox]")
23:08:52 Quit TeaSea (Remote closed the connection)
23:09:20 Quit doobeh (Remote closed the connection)
23:11:37merbananBagder: how's the webserver taking the slashdotting ? :)
23:11:55Bagderits the same as the main site
23:11:59Bagderseems to cope well
23:12:59merbananBagder: anyway I think MrH is working on making it possible to dualboot a possible rockbox firmware and a unmodified sandisk firmware
23:13:13Mikachuis this mrh guy anonymous?
23:13:15 Quit _Veseliq_ (Remote closed the connection)
23:13:19BagderMikachu: entirely
23:13:34Mikachuso this no anonymous contributor stuff only applies to code that goes into rockbox then
23:13:55Bagderthe real name policy is for Rockbox, yes
23:14:11 Quit jhMikeS (Read error: 110 (Connection timed out))
23:14:17Bagderbut yes, we might come to an interesting dilemma...
23:15:07Bagderfor the mi4code case I intend to do a re-implementation for the simple case we need, no crypto and dummy signing
23:15:45 Part lowlight
23:16:26 Join _Veseliq_ [0] (n=veseliq@
23:17:19merbananBagder: that sounds reasonable, you don't need all the mi4code features
23:17:37Bagdernope, we need very little
23:18:06merbananand the best thing is that you don't need to add the encryption keys to rockbox either
23:19:17merbananwhat videos do the sansa play ?
23:19:27BagderI don't know
23:19:32merbananwhat format ?
23:20:05BagderI don't know
23:20:14Bagderbut I believe it (can) transcode them
23:20:32Bagderit has some weird windows program for transferring video to the unit
23:21:01BagderI was at work with a Windows box nearby when I got it
23:21:33merbananI read somewhere that it was mjpeg, but I can't seem to confirm that
23:21:51Bagderhm wait...
23:22:24Bagderjust need to restore the firmware first ;-)
23:23:39BagderI think I have a demo video still around on it
23:25:09Bagdermjpb vlc says
23:25:26Bagderwhatever that is
23:25:59Bagdermplayer says "QuickTime/MOV file format detected."
23:26:11BagderFourcc: mjpb Codec: 'Motion JPEG B'
23:26:39merbanancan you play it ?
23:26:49merbananand what is the audio sample rate
23:26:57Bagderneither mplayer nor vlc play it
23:27:08merbanancan you send it to me ?
23:27:13BagderMOV track #1: 7 chunks, 0 samples
23:27:14BagderAudio bits: 16 chans: 2 rate: 11025
23:27:14BagderFourcc: twos
23:27:19BagderOpening audio decoder: [pcm] Uncompressed PCM audio decoder
23:27:19BagderAUDIO: 11025 Hz, 2 ch, 16 bit (0x10), ratio: 44100->44100 (352.8 kbit)
23:27:19BagderSelected audio codec: [pcm] afm:pcm (Uncompressed PCM)
23:27:29Bagderyeah, I'll upload it...
23:28:45Bagder160x208 at 15fps
23:29:10merbananno mpeg4-sp there
23:33:20Bagderthere's a second video in the same dir now, twice the size
23:33:26Bagderthat one works in mplayer for me
23:34:20BagderFourcc: mjpa Codec: 'Motion JPEG A'
23:36:45 Join MarcoPolo [0] (
23:37:18 Quit apo (Remote closed the connection)
23:38:15 Quit San||Away ()
23:41:11 Quit fergie ()
23:41:53merbananBagder: ffplay can play both, I'll inform the mplayer guys so they can add the fourcc if they havn't done that already
23:42:10BagderI'm not using the latest so it might work already
23:43:05 Join Shadowrazor [0] (
23:43:06merbananI should have tested it before I spoke, it worked
23:43:45***Saving seen data "./dancer.seen"
23:44:06Shadowrazorhi, could someone help me out with vlc? I want to convert a avi to m2v. It looks like vlc does the job but I dont know where vlc stores the file :(
23:44:34 Quit Doomed ("( :: NoNameScript 4.01 :: )")
23:45:33*merbanan looks at the name of the channel
23:46:32Bagderwell m2v is a bit rockboxish
23:46:55merbananhow about using ffmpeg instead then
23:46:59Shadowrazorwhat do u mean, isnt it just a mpeg file without sound?
23:47:00merbananworks all the time
23:47:10Shadowrazorffmpeg, I'll try it out
23:47:35merbananI'm not biased at all :)
23:48:08 Quit mikearthur (Remote closed the connection)
23:50:50 Quit midkay (Read error: 104 (Connection reset by peer))
23:52:49 Join San||Away [0] (
23:54:13 Part billytwowilly ("Konversation terminated!")
23:54:16Shadowrazordamn mplayer doesnt have standard a gui :S
23:54:47 Join lodesi_ [0] (
23:55:26Bagder"Run gmplayer if you want the GUI."
23:55:33 Join bluebrother [0] (
23:55:37 Join apo [0] (
23:56:27ShadowrazorI downloaded the mpayer, then I downloaded a skin, but I cant find a gmplayer.exer
23:56:55 Quit San||Away (Client Quit)
23:57:04 Join San||Away [0] (
23:57:17Bagderwell I have no idea
23:57:31BagderI use mplayer perfectly fine without any gui
23:57:59Shadowrazorbut i'm to newb to can do that ;)
23:58:24 Quit Davide-NYC ("Chatzilla 0.9.75 [Firefox]")
23:58:41 Quit barrywardell ()
23:58:46 Quit apo (Remote closed the connection)

Previous day | Next day