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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2009-05-20

00:01:23amiconnNot having the dotted frame essentially means tabbed navigation doesn't work, imo
00:03:19CIA-38New commit by 03alle (r20991): Forgot emdash and endash for 19-Nimbus
00:03:50pixelmathat's an easy fix though. I also did not understand why he chose it this why
00:03:54pixelmaerrr... way
00:04:40_fmlkugel (for the logs): what was the cause of the problem with loading/dumping the state in rockblox and how did you fix it? The code snippet you posted here seemed to be correct.
00:04:45 Quit _fml ("CGI:IRC 0.5.9 (2006/06/06)")
00:06:02*domonoky thinks that on the new website design, the wiki changes and mailing list thing doesnt really fit, maybe it would be better to put them also into the topbar ? (with scroll out, like svn)
00:07:29Zagorsvn is going to the dev page
00:07:51gevaertsYay, a firm decision! :)
00:08:51LloreanZagor: Even though it's users who want it on the main page, and only devs who think it should go to the dev page?
00:10:10 Join bubsy [0] (i=Bubsy@
00:10:17Zagorlet's call it the "details" page then
00:11:08Zagorthe main thing is that the front page, at least in my first variant, is simply introducing Rockbox. what it is, what it does and how you get it. and how you find more details.
00:11:26LloreanI just meant, users like to see that things are changing even on casual visits. I think the full SVN details is too much, but even something as simple as "last update to the current build happened at (time+date)" somewhere
00:11:53domonokythen i would suggest to move the wiki changes and mailing list list also to the details page :-)
00:11:59LloreanA way that even new people can see that the project was active quite recently. I prefer a list of a few things, but that adds clutter.
00:12:13Zagoryes, I agree some sort of activity indicator has its' place
00:12:30domonokymaybe we could include some "activity" in the yellow bar to the right ?
00:13:16domonokysomething like a nice realtime graph of commits/activities or something :-)
00:13:41pixelmaI always wanted to try if the recent changes things look more friendly in a list form (first line with date and author, below the topic), maybe that would already help
00:15:09 Quit bluebrother ("leaving")
00:18:43dfktwould it be acceptable to ask in flyspray for balance implementation on the x5's line-out?
00:19:56 Quit domonoky (Read error: 54 (Connection reset by peer))
00:20:26pixelmawhy do you want to adjust balance for line-out on the player by the way (not the attached stereo)?
00:20:46dfktfor a portable headphone amp, and because i'm hearing impaired
00:21:40 Quit jgarvey ("Leaving")
00:22:42 Quit arohtar (Client Quit)
00:22:54Lloreandfkt: Flyspray is not for feature requests.
00:23:42dfktthat's why i ask, is it a "bug" is it a "feature"? since the OF does have balance
00:24:01AlexPThe OF doesn't matter
00:24:02LloreanThe OF doesn't determine anything
00:24:21dfkti shouldn't have mentioned that, sorry
00:24:38LloreanIt's a bug if something doesn't work the way it's "supposed" to work as per how the developers intend it to work.
00:24:48LloreanIf you want it to do something new that it wasn't intended to do before, it's a feature request.
00:25:01 Quit Zagor ("Clint excited")
00:25:11 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
00:26:29dfktso i guess asking about it here in the channel is the reasonable thing to do then, or in the forums?
00:26:57LloreanGenerally the response you're going to get hear is that you should probably work on it yourself if you really want it.
00:27:50LloreanYou can post it in the "feature ideas" section of the forums, but Rockbox is entirely a volunteer effort.
00:27:54AlexPAnd in the forums you'll get a couple of people saying "Yeah, that'd be great" but nobody does anything
00:28:58froggymanhe could post it in FS after he gets something "working", right?
00:29:13AlexPSure, if he writes a patch
00:29:47Lloreanfroggyman: No "s around it. It needs to actually do something.
00:31:36froggymanLlorean: ohh, sorry
00:32:48pixelmanot sure if it's wanted. The X5/M5 (M3 too?) seems to be a true line-out independent from the headphones - it always has the same volume, except you can mute it. I thought it's intended then that it always works "at the same level", but am not sure
00:34:44Lloreanpixelma: Generally, yeah, the policy has been that the line out should be as independent as the hardware will allow from sound altering affects
00:34:54pixelmadfkt: maybe try starting a discussion at another time, maybe in the ml or so
00:35:14LloreanMailing list.
00:35:34high-rezSo something weird. The iPod port on my Subaru seems to work sort of. However I get audio drops... It plays for 5 seconds and drops for 10 (times are actually estimated).
00:35:51high-rez(5th gen ipod video 30g)
00:37:04dfktpixelma, the eq precut works on the line-out for volume change, but of course it's not what it's intended to do
00:38:12Lloreandfkt: The eq precut works because it's an effect applied in software.
00:39:10pixelmaI guess that has something to do with the fact that some sound altering effects have to be done in software (same as bass and treble on the Iaudios). But I am really on slippery ground now...
00:40:01Lloreanpixelma: That's basically it. Software effects like EQ go to both outputs no matter what since, iiuc, they happen before audio even gets into the PCM buffer.
00:40:25*Llorean is pretty sure he told this to dfkt once already a day or two ago
00:44:28 Quit ender` (" Is there like a way to put a compiler in "Just trust me on that one" mode?")
00:47:01CIA-38New commit by 03bluebrother (r20992): RFC2616 requires requests made to proxies to use the absoluteURI form while HTTP/1.1 clients shall only create it when sending a request to a proxy. ...
00:52:22 Join faemir [0] (
00:54:46 Quit bertrik (Read error: 113 (No route to host))
00:55:26 Join Zarathu [0] (i=zarathu@
00:55:52ZarathuHey guys, does Rockbox support the iTrip now or do I still need to recompile with that patch?
00:56:40evilnick_homeZarathu: Upgrade to the latest build and check it out
00:56:52Zarathuevilnick_home: I'll take that as a yes. Thanks.
00:56:54 Part Zarathu
00:57:40 Quit Mathiasdm2 ("Yuuw!")
01:02:17 Join wincent_balin [0] (
01:07:48 Quit Thundercloud (Remote closed the connection)
01:13:04 Part mcuelenaere ("Leaving.")
01:14:09 Quit advcomp2019 (Connection timed out)
01:19:20 Nick robin0800 is now known as robin0800_ (
01:19:23 Nick robin0800_ is now known as robin0800 (
01:20:40 Quit mt (Read error: 113 (No route to host))
01:22:14 Quit froggyman (" ajax IRC Client")
01:22:21 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
01:22:37 Quit amiconn (Remote closed the connection)
01:22:37 Quit pixelma (Read error: 104 (Connection reset by peer))
01:24:19 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
01:24:20 Join pixelma [50] (n=pixelma@rockbox/staff/pixelma)
01:30:46 Join Strife89 [0] (n=michael@
01:31:47*Strife89 would like to improve FS #10202 but can't find figure out how to write keymap tables.
01:32:30***Saving seen data "./dancer.seen"
01:32:41 Part toffe82
01:35:55 Quit l403 (Read error: 113 (No route to host))
01:36:01 Join JdGordon| [0] (i=46079d89@gateway/web/ajax/
01:36:30pixelmahave you looked at other plugin tex files?
01:36:41 Quit SirFunk__ (Read error: 110 (Connection timed out))
01:39:21 Quit robin0800 (Read error: 104 (Connection reset by peer))
01:40:07 Join Strife1989 [0] (n=michael@
01:40:17 Quit Strife89 (Nick collision from services.)
01:40:23Strife1989pixelma: A little.
01:40:27 Nick Strife1989 is now known as Strife89 (n=michael@
01:41:06Strife89pixelma: What's a color-only plugin aside from Rockpaint?
01:44:12pixelmaaccording the manual's plugins/main.tex the ppmviewer is
01:50:19JdGordon|Llorean: you get a chance to test out the usb charge atch yet?
01:50:33JdGordon|.. and maybe fix the manual language while your atr it?
01:59:05 Quit BHSPitMonkey (Nick collision from services.)
02:06:27 Quit JdGordon| (" ajax IRC Client")
02:11:39 Join thomas_ [0] (n=thomas@
02:12:26 Quit amiconn (Nick collision from services.)
02:12:28 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
02:12:28 Quit pixelma (Nick collision from services.)
02:12:28 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
02:12:48 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
02:12:48 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
02:13:36 Join kkurbjun [0] (n=kkurbjun@rockbox/developer/kkurbjun)
02:16:09 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
02:19:18 Join Strife1989 [0] (n=michael@
02:25:21 Join michaelcarr [0] (n=michael@
02:26:21 Join SirFunk [0] (
02:28:15 Quit Strife89 (Read error: 110 (Connection timed out))
02:31:31 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
02:31:48michaelcarrFS #10202 updated.
02:31:48 Quit thomas_ ("Leaving")
02:32:25 Nick michaelcarr is now known as Strife89 (n=michael@
02:32:41 Quit GodEater (Read error: 110 (Connection timed out))
02:38:39 Quit JdGordon ("Leaving.")
02:39:12Strife89I know the keymaps aren't complete, but that didn't stop other manual entries. Does anyone object to commiting this?
02:41:57 Join CaptainKwel [0] (
02:42:07 Quit Strife1989 (Read error: 110 (Connection timed out))
02:42:43 Join JdGordon [0] (
02:52:36 Quit efyx_ (Remote closed the connection)
02:57:42LloreanStrife89: The manual's gotten along fine without Rockpaint. Why not put the rest of the work in first?
02:58:04LloreanJdGordon: Sorry, haven't been able to. Housesitting and don't have anything with me to test on.
02:58:59 Nick fxb is now known as fxb__ (
02:59:01Strife89Llorean: Because I'm not sure how much of the keymaps I can even get done....
02:59:33LloreanStrife89: The keymaps are defined in the source, right?
03:00:07Strife89... Oh, yeah.... ^^:
03:00:22LloreanJdGordon: The forum post seems to be about the feature that was there briefly to leave USB mode by pressing Select after your device was already connected.
03:04:03 Join JdGordon| [0] (i=43a02c5a@gateway/web/ajax/
03:04:16 Quit JdGordon| (Client Quit)
03:04:49 Quit barrywardell (Remote closed the connection)
03:06:07Strife89Llorean: Thanks for reminding me of that, that'll make things much easier. :)
03:06:33 Quit Strife89 ("Huzzah!")
03:07:27 Join lymeca [0] (
03:18:46 Part wincent_balin ("Kopete 0.12.7 :")
03:20:21 Quit itcheg (" ajax IRC Client")
03:32:34***Saving seen data "./dancer.seen"
03:39:03 Quit _lifeless (Read error: 110 (Connection timed out))
03:52:45 Quit lymeca (Read error: 113 (No route to host))
03:56:09 Join Strife89 [0] (n=nds@
03:58:30 Quit JdGordon (Read error: 104 (Connection reset by peer))
04:00:49 Join _lifeless [0] (n=lifeless@
04:03:24 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
04:04:58 Join sharp1 [0] (
04:06:38 Part sharp1
04:19:37 Quit Strife89 ("ClIRC - IRC client for Nintendo DS")
04:28:31 Join bluefoxx [0] (
04:29:04bluefoxxSo is there a bootloader for the e200 that entirely disables the OF?
04:29:59bluefoxxSeeing as rockbox does everything that DAP firmware should do and more(emulators, apps, games, washes your pants and makes great julian fries, etc)
04:32:14GaloisI find the e200 rockbox slightly unreliable for USB transfers, so I still use the OF for that
04:35:57Unhelpfulif i'm understanding this correctly, i can't really make used of the coldfire multiply accumulate with load instruction to load in packed 8-bit values - it *looks* like it reads in the same size as used for the MAC operation, and i would assume that needs to be aligned, as well?
04:37:21 Join lymeca [0] (
04:37:52 Quit daurnimator (Read error: 60 (Operation timed out))
04:38:22tmztbluefoxx: there was some discussion of that a while a go, I think it's possible but don't know how or if you would really want to do it
04:38:46bluefoxxbecause i have an 8 gig microsd card in the thing that i can't access otherwise
04:38:53bluefoxxand i loath the OF with a firey passion
04:39:18bluefoxxyou have no idea how many times i returned one of these because the OF barfed on itself before i found rockbox
04:40:43bluefoxxit's done stupid things like refusing to show up on a system at all, to barfing a green screen on bootup, or a blue one, to magically making my files vanish, or how about having the thing loaded with music, and it refusing to show anything
04:41:07bluefoxxor it barfes a green/blue screen when it goes to update the database...
04:41:15bluefoxxfor that matter, it sucks just for not doing sdhc...
04:41:36bluefoxxi can think of dozens of ways why i want to obliterate it and use rockbox only
04:41:41Unhelpfulbluefoxx: there's a bootloader on flyspray that doesn't automatically boot OF on USB insert startup :P
04:42:10bluefoxxfor that matter, i do, but its a bit of a pain to have to manually start the thing up into rockbox before i go and plug the usb in...
04:42:21bluefoxxdo share a link if you don't mind...
04:42:48bluefoxx(i have a habit of remembering i meant to ask something in here when i'm grabbing the thing and going out the door but want to update the music...)(
04:45:12LloreanGalois: Unreliable how? Have you filed a bug report?
04:47:33bluefoxxThe only unreliable thing i've found with it so far is when it boots back into rockbox after having been in the OF, the microsd card refuses to show up until i reboot rockbox again and than reinsert the card...
04:47:50bluefoxxotherwise its actually faster ime
04:47:57Lloreanbluefoxx: I more meant, what problem is he having with USB
04:48:36LloreanBug reports mean at least someone knows about it. If one's not filed, and nobody else has it, nobody may even know it needs fixing
04:49:34 Quit miepchen^schla (Read error: 101 (Network is unreachable))
04:50:12bluefoxxRight...I'm assuming that with those bootloaders i run sansapatcher and direct it to that file somehow??...
04:51:40tmztI assume since e200v1 has a recovery mode it's possible to recover the firmware without OF
04:52:26bluefoxxboot the device with the center button held down and the hold switch on
04:52:51bluefoxxiirc (its been a good year and a bit since i managed to brick mine seriously enough to warrent such)
04:52:53tmztyeah, that's what I mean
04:53:23tmztthere was a rockbox mi4 file made though, right?
04:54:15tmztto allow flashing rockbox in place of OF've lost me now...
04:55:04bluefoxx(than again, its not hard to right now..)
04:55:40tmztmi4, the portal player firmware
04:56:15bluefoxxyes, i got that much..
05:00:37Unhelpful"rockbox.mi4" is the rockbox firmware that the rockbox bootloader loads, isn't it?
05:01:33 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
05:01:49 Join lee321987 [0] (
05:01:53bluefoxxas far as i know, yes...
05:02:47bluefoxxmy knowlage is a tad limited, not being a programmer and such, anyways, i have to go, someone has a surely horrific laptop for me to 'fix' or something
05:12:07lee321987anyone −− you know how your computer alerts you when someone types your name in IRC? Is this case-sensitive?
05:13:10lee321987anyone seeing this?
05:13:11rvvs89lee321987: That's done client-side, and depends on the client
05:13:59rvvs89lee321987: Although I expect that most clients ignore case when looking for hilightable words
05:14:50lee321987if anyone talks to gevaerts, please give him my next to messages (I think they concern recent commits of his).
05:14:56lee321987Gevaerts: USB transfer rate has slowed down. DAP: c200v1 ~~ RB:r20981 ~~ Computer to internal rate average = 3.59MB/s ~~ Computer to SD card write average = 3.78MB/s
05:15:02lee321987RB:r20992 ~~ Computer to internal rate average = 2.98MB/s ~~ Computer to SD card write average = 3.23MB/s
05:15:37lee321987*next TWO messages
05:16:26tmztUnhelpful: I thought it was rockbox.sansa but I'm not sure, maybe that's just ams
05:16:37tmzthaven't had a v1 for a year now
05:20:35 Join |ahmina| [0] (
05:20:50cool_walking_The archives on contain .rockbox/rockbox.mi4 ...
05:21:08lee321987I think I need a new DAP. Is RB working on any that are currently in production?
05:21:25cool_walking_lee321987: I don't think so.
05:21:40bluefoxxcheck your local craigslist for an e200
05:22:06lee321987do you know if that takes the same battery as the c200?
05:22:06 Quit lymeca (Read error: 113 (No route to host))
05:22:14 Join hobbs [0] (n=nnnandre@p3m/member/hobbs)
05:22:18lee321987sorry - off topic.
05:23:17bluefoxxoh yay, crashed rockbox while doing USB transfers. Note to self: don't do more than two moves from player to sd at once...
05:23:49hobbsSilly opinion question: I have a Gigabeat S30 that I've been running Rockbox on, but the hardware is flaking out on me and I'm probably not up to fixing it. Which would probably make me happier? A new Gigabeat S off eBay or a supported (classic) iPod off eBay? And which iPod is the best?
05:24:10bluefoxxavoid ipods at all costs is my opinion
05:24:38bluefoxxbut i have to fix the occasional apple product, and detest them more and more, so i'm somewhat biased against them
05:24:43hobbsI lean that way a bit myself ;)
05:24:53 Join JdGordon| [0] (i=836b0069@gateway/web/ajax/
05:25:03hobbsiPod seems like mostly decent hardware though. I just want something that will be reliable
05:25:29bluefoxxoh this is just rockbox completely freaks out on USB connects...
05:25:31hobbs(and have a goodly amount of storage and run rockbox well, which limits me to those two devices as far as I can tell)
05:25:33bluefoxxtime to downgrade again.
05:25:53Unhelpfulthey're much *slower* hardware than S, though.
05:26:49lee321987bluefoxx: Sansa c200v1?
05:27:15lee321987e200erase saved me once
05:27:22hobbsUnhelpful: didn't really consider that
05:27:34bluefoxxtheres no chance in hell i'm wiping files off of my mp3 player...
05:27:37lee321987(maybe it's "e200erasetool")
05:27:52bluefoxxresorting 6 gigs of music by hand out of 80 gigs takes far too long..
05:27:56hobbsUnhelpful: (on the flipside the gigabeat runs that processor at 100% all the time as far as I can tell... I wish it throttled for better battery life :)
05:27:57Unhelpfulit's not going to be a big issue for common audio formats. mpegplayer is quite pitiful on ipod video, though.
05:28:47bluefoxxugh ugh ugh
05:28:51Unhelpfulhobbs: it's trickier than that - merely changing the clock on the S CPU doesn't have much impact on power consumption, we need to use voltage scaling to get savings.
05:28:55bluefoxxno no no no thats NOT how that should work dammit...
05:29:33hobbsUnhelpful: understood. I've screwed with it enough to know it can be non-trivial. Just an "I wish" :)
05:29:50hobbsUnhelpful: (no, not on the gigabeat, on some PXA devices)
05:30:42tmztwhat pxa targets?
05:31:34hobbstmzt: Zaurus (clamshell type)
05:31:50tmztoh, not a rockbox target
05:32:36***Saving seen data "./dancer.seen"
05:32:36 Quit Horscht ("Verlassend")
05:32:55tmztbluefoxx: well my point was you should be able to flash the .mi4 but you want to ask someone else here about that and I can't remember who it was
05:33:19bluefoxxugh ugh ugh
05:33:28hobbsanyway I guess I'll go for the S30. I see a few putatively new ones on ebay for affordable prices. Less than iPod 5Gs. :)
05:33:53bluefoxxnext time i come in here asking something someone do me a favour and kick me with the reminder that every time i do do this, i completely fuck my sansa up...
05:33:54lee321987bluefoxx: you do know I was talking about this −−- right? It was originally for the e200 (AND THE "e200" VERSION IS _DIFFERENT_ , but I don't know exactly where it is).
05:34:39bluefoxx...yea, i don't erase data, ever.
05:34:57bluefoxxsomeplaces have a data destruction policy, i have a keeping data policy...
05:35:15lee321987my c200 would only boot to manufacturing mode, and that tool worked.
05:36:05lee321987tmzt: maybe it was gevaerts
05:36:54lee321987his tool saved a $90 brick for me :)
05:39:52 Join froggyman [0] (n=47ba0b80@gateway/web/cgi-irc/
05:40:21bluefoxxok, that solved that. rockbox 3.2 is broken for me, and 3.1 works fine. and that ancient version of kugel build also seems to work just as well, with all the features i liked but without the usb connect...
05:40:25bluefoxxblah, i need to run now..
05:40:27 Quit bluefoxx ()
05:42:02 Quit froggyman (Client Quit)
05:43:00lee321987last time −− please tell him if you see him:
05:43:02lee321987gevaerts: DAP: c200v1 −−- r20981, computer to internal = 3.59MB/s −−−−−−−− r20992, computer to internal = 2.98MB/s
05:43:10 Quit lee321987 ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
05:44:03 Quit |ahmina| (Read error: 110 (Connection timed out))
05:46:10GaloisLlorean: general instability, crashes. Much like what bluefoxx just described. Not consistently reproducible. I also have way too many things going on in my life with family to shepherd bug reports at this time, but that's it fwiw
05:47:39LloreanWell, don't expect it to improve if you can't be bothered to provide useful reports
05:47:46LloreanFor what it's worth, I haven't seen Rockbox crash or freeze in months
05:48:01GaloisI am in fact not expecting it to improve. It works fine.
05:48:28GaloisI dont' even need the USB, as I usually transfer using the microsd card
05:51:08 Quit JdGordon| (" ajax IRC Client")
05:51:37LloreanHave you been using a remotely recent version?
05:51:54 Part hobbs ("Leaving")
05:52:02Galois20090422 build or something
05:52:38LloreanThat's a date, not a revision
05:52:50LloreanIt just tells what day it was compiled, could be any revision prior to that day
05:53:58tmztis there some reason the svn revision is not in the about?
05:54:31Lloreantmzt: What "about"?
05:54:34LloreanIt's in "rockbox info"
05:54:40tmztoh, ok
05:54:58LloreanWhat "about" are you referring to?
05:55:30tmztabout is what windows programs call it
05:55:38Galoislooks like r20774?
05:56:05Lloreantmzt: So why were you asking why it's not there if it is?
05:56:46tmztI assumed if it was there Galois would have reported it, I assume wrong
05:57:03LloreanPlease, try not to make any assumptions.
05:57:07LloreanIt's a bad habit and leads to further contusion
05:57:44GaloisI don't normally navigate to the about screens, it's faster just to reboot the thing and look at the bottom of the screen
05:57:50Galoisbut anyway, dinner
06:11:42 Quit CaptainKwel (Remote closed the connection)
06:19:09 Quit saratoga (" ajax IRC Client")
06:19:30 Join Makuseru [0] (
06:19:53 Part Makuseru ("Konversation terminated!")
06:20:23 Join saratoga [0] (i=9803c6dd@rockbox/developer/saratoga)
06:21:08 Join BXCracer [0] (
06:51:24 Join nibbler__ [0] (
06:53:48 Quit perrikwp (" ajax IRC Client")
06:54:04 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
07:07:16 Quit ze (Read error: 104 (Connection reset by peer))
07:07:18 Join ze0 [0] (i=ze@
07:08:08 Nick ze0 is now known as ze (i=ze@
07:09:00 Join itcheg [0] (i=62db4c46@gateway/web/ajax/
07:09:03 Quit kkurbjun (Read error: 110 (Connection timed out))
07:13:52 Nick fxb__ is now known as fxb (
07:19:40 Join ze0 [0] (i=ze@
07:24:27 Quit itcheg (" ajax IRC Client")
07:27:53 Quit BXCracer ("Quit")
07:29:59 Join matsl [0] (
07:30:29 Quit ze (Read error: 60 (Operation timed out))
07:30:29 Nick ze0 is now known as ze (i=ze@
07:32:40***Saving seen data "./dancer.seen"
07:50:56 Nick fxb is now known as fxb__ (
07:51:42 Join lymeca [0] (
07:53:52 Join |unjustice| [0] (
07:54:15 Join bertrik [0] (
07:56:41 Join daurnimator [0] (n=daurnima@unaffiliated/daurnimator)
08:03:20 Join ender` [0] (
08:03:31amiconnUnhelpful: Coldfire MAC with load always loads 32 bit data. And it doesn't need to be aligned, as usual on coldfire, but it should be aligned for performance reasons.
08:04:16Unhelpfulamiconn: hrm... i'm not sure it would be useful in the h-scaler, which will need to load (packed) bytes and zero-extend them
08:04:44Unhelpfulthe v-scaler is using the h-scaler's output, though, which is 32-bit and aligned, so that can use mac+load
08:05:48amiconnMAC with load only helps speeding up things under certain conditions
08:06:19amiconnMAC without load takes 1 cycle (apart from result latency of course), MAC with load takes 2 cycles
08:06:46Unhelpfulah. so it wouldn't be much help for, say, loading a value for the next MAC operation
08:07:40amiconnWell, it saves an instruction (but at the cost of an extension word, so it may not save binary size)
08:07:52amiconnIt also helps saving registers (versus preloading)
08:08:42Unhelpfulwell, actually, i guess if i pre-zero some registers, i could use the MASK to load zero-extended bytes?
08:09:11amiconnIn IRAM it's usually a good idea to use MAC with preload if possible. In DRAM it is faster to not use it if you can preload at least 16 bytes aligned to 16 bytes using movem.l
08:10:11Unhelpfulmovem is essentialy a streaming load operation?
08:10:33amiconnmovem.l is the equivalent of arm's ldm/stm
08:11:56amiconnIt's also the only instruction that issues line bursts (128 bit, 4 registers) for loading and storing with suitable alignment
08:12:03Unhelpfulthe vscalers could use movem if the loop were unrolled a bit. they basically perform the same operation on one after another of several rows of 32-bit ints
08:12:36amiconnWell, as mentioned it depends
08:13:34Unhelpfulright... pulling 4 values from each of two scaler work lines would fill all of the data registers...
08:14:28Unhelpfulhere's the scaler patch i've been working on. the C math works nicely on my ARM devices, and times faster on them and on coldfire (i don't have benchmarks for SH1 yet).
08:16:18Unhelpfulsh1 could benefit from an inline 16x32->32 multiply in the horizontal scalers, if we're willing to assume downscaling is never to more than 1/257 of original size. that's the same assumption already used to do hardware multiplies with the current scaler code.
08:16:21 Quit nibbler__ (Read error: 101 (Network is unreachable))
08:16:29amiconnIn IRAM the various burst modes don't matter. 4x MAC with preload are 8 cycles, versus 4x MAC + 1x movem.l +1 addition (for advancing the pointer) are 10 cycles
08:16:39amiconnBut in DRAM, the latter is *much* faster
08:17:12 Quit antitrons ()
08:19:17Unhelpfulthe scaler and jpeg decoder are pretty much always operating from DRAM, from what i can tell - the buffer where image temp data goes is DRAM, correct? and the jpeg DCT data block goes on stack, which also appears to be DRAM in the case of the audio thread, which does the album art loads.
08:20:09amiconnThe question is whether it would be worth to put some things into IRAM
08:21:45Unhelpfuli understand that to be a very limited resource on most of our targets... :/
08:22:42 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
08:30:21 Join n1s [0] (n=n1s@rockbox/developer/n1s)
08:31:20 Quit |unjustice| ("KVIrc 3.2.4 Anomalies")
08:36:09 Join einhirn [0] (
08:39:55 Join GodEater [0] (n=bibble@rockbox/staff/GodEater)
08:40:46 Quit einhirn (Read error: 104 (Connection reset by peer))
08:43:39 Join Rob2223 [0] (
08:43:51 Quit bertrik (Read error: 113 (No route to host))
08:44:30 Join flydutch [0] (
08:48:28 Join petur [50] (n=petur@rockbox/developer/petur)
09:00:23 Quit perrikwp (" ajax IRC Client")
09:00:35 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
09:00:56 Quit Rob2222 (Read error: 110 (Connection timed out))
09:10:21 Join Grahack [0] (
09:11:13 Quit goffa (Remote closed the connection)
09:11:19 Join goffa [0] (n=goffa@
09:15:23 Join antil33t [0] (i=antil33t@
09:21:12 Join petur2 [50] (n=petur@rockbox/developer/petur)
09:21:16n1shmm, there seems to be a bug with the dropdown menus on the new website, the menu disappears when i try to move the mouse down from the heading tio the menu about 7 out of 10 times
09:21:29 Quit petur (Nick collision from services.)
09:21:31 Nick petur2 is now known as petur (n=petur@rockbox/developer/petur)
09:22:16n1smost visible when moving the pointer kind of slowly
09:22:47n1soh and i think the "Home" button looks very much out-of-place
09:23:05n1sbut overall i think it looks nice
09:23:14 Nick J-23 is now known as Moarc (
09:23:36 Nick Moarc is now known as J-23 (
09:23:56 Nick J-23 is now known as Moarc (
09:24:15GodEatern1s: which browser / OS combo ?
09:24:18 Nick Moarc is now known as J-23 (
09:24:22GodEaterworks fine for me
09:24:27n1sdebian ff3.0
09:24:33n1seh iceweasel i mean
09:24:54GodEaterany weird extensions ?
09:25:16n1sI'll try disabling them
09:25:49BagderJ-23: stop that please
09:26:39n1sGodEater: disabled the only 2 i have and it makes no difference
09:27:24GodEatercurious - I've checked it on FF3.0 on windows and Ubuntu
09:27:30GodEaterworks every time =/
09:27:33GodEaterI wonder what the issue is
09:27:36Zagorn1s: I see it too. there's a thin line where neither the button or the menu is active, and hence the menu disappears
09:28:00*GodEater hovers his mouse over this line, and the menu fails to disappear
09:28:17 Quit Xerion (Read error: 104 (Connection reset by peer))
09:28:28Zagoryeah it stays open with opera too. but the cursor changes from hand to arrow.
09:28:36GodEatersame here
09:29:00ZagorGodEater: what os/browser?
09:29:02 Join Xerion [0] (
09:29:14Zagorah, saw it
09:29:17GodEatercurrently FF3.0 and windows
09:29:40GodEaterah ok - it breaks in IE8 64 bit
09:30:22 Join Thundercloud [0] (
09:30:28Zagorit breaks in arora (webkit browser) too
09:30:29GodEaterand the 32bit version
09:30:48*GodEater goes to grab safari too
09:31:15Zagorit doesn't really matter though, it breaks in enough mainstream browsers that it has to be fixed.
09:32:43***Saving seen data "./dancer.seen"
09:43:17 Join ender1 [0] (i=krneki@
09:44:31 Quit ender` (Read error: 60 (Operation timed out))
09:47:05 Quit matsl (Read error: 104 (Connection reset by peer))
09:47:27 Join matsl [0] (
09:54:19 Quit Thundercloud (Remote closed the connection)
10:14:42 Quit perrikwp (" ajax IRC Client")
10:15:04 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
10:22:19 Quit advcomp2019 ("I didn't set up UberScript properly!")
10:27:16 Quit perrikwp (Remote closed the connection)
10:27:16 Quit saratoga (Remote closed the connection)
10:27:16 Quit cool_walking_ (Remote closed the connection)
10:29:00 Join efyx_ [0] (
10:32:27 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
10:35:39 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
10:36:56 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
10:40:02 Quit cool_walking_ (" ajax IRC Client")
10:44:19 Join barrywardell [0] (
10:44:41 Join PaulJam [0] (
10:45:06 Quit barrywardell (Remote closed the connection)
10:47:07 Join barrywardell [0] (
10:51:32 Join webguest100 [0] (n=5498329b@gateway/web/cgi-irc/
10:52:05 Quit _Auron_ (Remote closed the connection)
10:52:54webguest100hi, how do i change the display font on my H140, I have browsed around in the settings and I cannot find a menu which says, change font or new font etc...
10:54:37webguest100there's no menu under fonts that says change font, or new font, OOOPS, i just found it under theme's my apologies
10:54:45 Quit webguest100 (Client Quit)
10:56:10 Quit BHSPitMonkey ("Ex-Chat")
10:58:00 Join keetah [0] (
11:02:54 Part keetah
11:04:39 Join robin0800 [0] (
11:06:03PaulJamHi, i have a little problem with the album art mode of the sliding puzzle plugin: For most albums i have the albumart in cover.75x75.bmp (75x75), cover.bmp (176x176) and a <albumname>.jpeg with a high resolution. And it seems, that the sliding puzzle plugin tries to use the jpeg file which doesn't seem to fit in the plugin memory and so the album art mode isn't available. if i rename the jpeg, then everything works and it uses the cover.bmp file. do
11:06:03PaulJames anyone know a solution that does not imvolve renaming/deleting the jpeg?
11:07:54tmztthat sounds like a bug, somebody has been working on the jpeg decoder recently
11:08:30 Quit perrikwp (" ajax IRC Client")
11:08:36 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
11:08:37tmztnot sure if they intend there to be a fallback to another filename, but there has a been a discussion of what filenames to support
11:08:43Unhelpfulthe jpeg decoder is new. there's a function in pluginlib that will try to use the right loader for an image file, but sliding puzzle is not using it yet
11:09:59PaulJamso it tries to decode the jpeg with the bmp loader?
11:10:43Unhelpfulprecisely. this can probably be fixed quickly. falling back to search for another file if the first one can't load is a bit less likely to happen.
11:16:53 Join einhirn [0] (
11:18:55Unhelpfulugh, slightly less trivial, as sliding_puzzle currently uses a static buffer for image loads. that's probably just not going to work for jpeg :)
11:24:56linuxstbUnhelpful: At least for the targets with 512KB plugin buffers, that shouldn't be a problem - I can't see it needing that much RAM itself...
11:25:41linuxstbi.e. if it doesn't already use it for other things, you can get a pointer to the unused plugin buffer RAM, and use that.
11:25:52Unhelpfullinuxstb: it's more a matter of it being less trivial than it was for pictureflow, which was already using the plugin buffer when it was adapted to bitmap scaling :)
11:26:57linuxstbIt looks like sliding_puzzle doesn't call rb->plugin_get_buffer(), so you can do that - almost trivial.
11:28:05Unhelpfulindeed, it's just a matter of changing the image buffer to a pointer, and adding a call... and also using read_image if the target has albumart. since sliding_puzzle doesn't use greylib, there's little point loading album art in it on mono targets.
11:30:36 Quit perrikwp (" ajax IRC Client")
11:32:47***Saving seen data "./dancer.seen"
11:32:53*Unhelpful notices that sliding_puzzle.c is another of those evil plugins (which put an #ifdef around the entire plugin code)
11:34:43pixelmaI guess that's because it's an old one (in the origin) before SOURCES and no-one bothered removing it as it shouldn't hurt
11:36:31Unhelpfulpixelma: it doesn't, as long as it's managed correctly in SOURCES. when the loader-initialized API pointer went in i remember there being issues with some that could be built under certain circumstances, but have the plugin code #ifdef'd away
11:37:09Unhelpfuli believe they stopped the build with link errors... this one didn't get changed then because it didn't cause trouble :)
11:47:16CIA-38New commit by 03unhelpful (r20993): Support JPEG album art in sliding_puzzle.
11:47:37UnhelpfulPaulJam: that should fix it, worked for me on ipodvideo sim
11:48:16PaulJamThank you :)
12:22:53 Nick fxb__ is now known as fxb (
12:41:03 Join RandomDestructn [0] (
12:45:20 Quit PaulJam (Read error: 104 (Connection reset by peer))
12:45:45 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
12:54:17 Quit Zagor ("Don't panic")
12:54:42 Quit R[a]ndom (Read error: 110 (Connection timed out))
12:55:44 Join Horschti [0] (n=Horscht@xbmc/user/horscht)
13:01:11 Quit Horscht (Read error: 60 (Operation timed out))
13:01:45 Join funman [0] (n=fun@rockbox/developer/funman)
13:11:53 Quit martian67 (
13:11:53 Quit suom1 (
13:11:53 Quit robin0800 (
13:11:53 Quit Rob2223 (
13:11:53 Quit amiconn (
13:11:53 Quit jfc (
13:11:53 Quit kachna|lappy (
13:11:53 Quit shodanX (
13:11:53 Quit gevaerts (
13:11:53 Quit courtc (
13:11:53 Quit dionoea (
13:12:31NJoinrobin0800 [0] (
13:12:31NJoinRob2223 [0] (
13:12:31NJoinamiconn [50] (n=jens@rockbox/developer/amiconn)
13:12:31NJoinjfc [0] (
13:12:31NJoinkachna|lappy [0] (
13:12:31NJoinshodanX [0] (
13:12:31NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
13:12:31NJoinmartian67 [0] (n=martian6@about/linux/regular/martian67)
13:12:31NJoincourtc [0] (n=court@unaffiliated/courtc)
13:12:31NJoindionoea [0] (n=dionoea@videolan/developer/dionoea)
13:12:31NJoinsuom1 [0] (
13:14:08 Join SirFunk_ [0] (
13:14:11 Join PaulJam [0] (
13:14:18 Quit SirFunk (Operation timed out)
13:15:12CIA-38New commit by 03funman (r20994): Sansa Fuze : fix r20680 for pong plugin (BUTTON_REPEAT can't be used)
13:16:24Bagderfinally pong on the fuze!
13:16:38 Quit suom1 (
13:16:38 Quit martian67 (
13:16:38 Quit shodanX (
13:16:38 Quit robin0800 (
13:16:38 Quit Rob2223 (
13:16:38 Quit courtc (
13:16:38 Quit jfc (
13:16:38 Quit gevaerts (
13:16:38 Quit kachna|lappy (
13:16:38 Quit amiconn (
13:16:38 Quit dionoea (
13:16:47funmani think it'll only be ready for release once rockblox1d is working efficiently
13:17:04Bagderyeah, that's the most crucial part of rockbox
13:17:25funmanperhaps that's why AMS people ignored us : pong not working :(
13:17:50funmanmore seriously, I plan to ping AMS again for the AS3531 datasheet when we'll have official builds for Sansa AMS
13:18:46NJoinrobin0800 [0] (
13:18:46NJoinRob2223 [0] (
13:18:46NJoinamiconn [50] (n=jens@rockbox/developer/amiconn)
13:18:46NJoinjfc [0] (
13:18:46NJoinkachna|lappy [0] (
13:18:46NJoinshodanX [0] (
13:18:46NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
13:18:46NJoinmartian67 [0] (n=martian6@about/linux/regular/martian67)
13:18:46NJoincourtc [0] (n=court@unaffiliated/courtc)
13:18:46NJoindionoea [0] (n=dionoea@videolan/developer/dionoea)
13:18:46NJoinsuom1 [0] (
13:22:33CIA-38New commit by 03funman (r20995): Fix my mistake in r20987/FS #10219 : live update of Sansa AMS clock freqs (fix by Jack Halpin)
13:22:47 Join renke [0] (
13:26:59 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
13:32:40funmanbertrik (for the logs): when using caches on the Sansa AMS, i2c fails at 400kHz because the "check if still busy" at the beginning of ascodec_read() & ascodec_write() fails : i2c is still busy. If I add 100 loops of 5 consecutive nops, i2c works fine. But I wonder if we couldn't just put a loop waiting for readiness, like at the exit of these 2 functions?
13:32:50***Saving seen data "./dancer.seen"
13:33:19funmanIf I add 100 loops of 5 consecutive nops >>>just before the check for i2c business<<<
13:33:27 Quit Grahack ("Leaving.")
13:46:11 Quit funman ("leaving")
13:47:09 Join PaulJam_ [0] (
13:51:40 Quit Horschti (Read error: 110 (Connection timed out))
13:55:26 Quit PaulJam (Read error: 104 (Connection reset by peer))
13:56:04 Quit timc ("Leaving")
13:59:07 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
14:04:30 Join froggyman [0] (n=187b533e@gateway/web/cgi-irc/
14:20:16 Quit _lifeless (Read error: 104 (Connection reset by peer))
14:20:55 Join _lifeless [0] (n=lifeless@
14:21:29 Quit PaulJam_ (Read error: 110 (Connection timed out))
14:24:45 Quit froggyman ("CGI:IRC")
14:27:50 Join BXCracer [0] (
14:31:34 Quit Ridayah (Read error: 60 (Operation timed out))
14:32:12 Join LambdaCalculus37 [0] (i=44a0430d@gateway/web/ajax/
14:32:23 Join Grahack [0] (
14:37:59CIA-38New commit by 03mcuelenaere (r20996): Fix Ingenic Jz4740 USB driver (USB IN <-> OUT) and some whitespace cleaning
14:43:27CIA-38New commit by 03mcuelenaere (r20997): Speed up LCD driver on Ingenic Jz4740 by doing some sort of partial LCD updates.
14:48:35 Join daurn [0] (n=daurnima@unaffiliated/daurnimator)
14:54:12 Quit robin0800 (Remote closed the connection)
14:56:17 Join robin0800 [0] (
15:02:46 Quit daurnimator (Read error: 110 (Connection timed out))
15:10:22 Join froggyman [0] (n=187b533e@gateway/web/cgi-irc/
15:12:32 Quit SirFunk_ (Read error: 110 (Connection timed out))
15:15:09 Join evilnick_7 [0] (i=0c140464@gateway/web/ajax/
15:32:53***Saving seen data "./dancer.seen"
15:39:05 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
15:54:15 Quit gevaerts (Nick collision from services.)
15:54:24 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
15:55:06 Quit froggyman ("CGI:IRC")
16:03:33 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
16:08:26 Quit evilnick_7 (" ajax IRC Client")
16:18:31 Join __lifeless [0] (n=lifeless@
16:21:38 Quit robin0800 (Read error: 104 (Connection reset by peer))
16:25:00 Join robin0800 [0] (
16:25:06 Quit _lifeless (Read error: 110 (Connection timed out))
16:27:32 Join evilnick_230` [0] (i=0c140464@gateway/web/ajax/
16:29:43 Quit kachna|lappy (Read error: 113 (No route to host))
16:33:00 Quit bmbl ("Woah!")
16:33:14 Nick evilnick_230` is now known as evilnick_230 (i=0c140464@gateway/web/ajax/
16:40:15 Join toffe82 [0] (n=chatzill@
16:41:10 Quit GodEater ("Leaving")
16:47:38 Join saratoga [0] (i=9803c6dd@rockbox/developer/saratoga)
17:07:11 Quit evilnick_230 (" ajax IRC Client")
17:07:17 Join Strife89 [0] (n=michael@
17:07:54Strife89Could someone look and tell me if I have this right so far?
17:08:27Strife89I don't want to make the entire table and then it be completely wrong.
17:08:43 Join wincent [0] (
17:11:09LambdaCalculus37Strife89: Looks good so far, but here's an example of a table you can use to follow along with:
17:12:18Strife89LambdaCalculus37: Alrighty, thanks. :)
17:13:14Strife89LambdaCalculus37: By the way, we're going camping this weekend, so if you haven't sent it yet, just wait 'till this Saturday.
17:14:35 Join evilnick_230 [0] (i=0c140464@gateway/web/ajax/
17:15:17pixelmayou could at least close (end) the table, I think
17:16:08LambdaCalculus37Strife89: Ahh, didn't notice what pixelma mentioned. Close the table and it's good.
17:17:08pixelmaand the button part should be capitalised I think (in the beginning there is\ buttonLeft etc., should be \ButtonLeft
17:19:01 Join wpyh [0] (n=william@
17:19:20pixelmaStrife89: have a look at the keypad-something.tex files under manual/platform to know exactly (it's also called ButtonRec I believe, not ButtonRecord)
17:19:38LambdaCalculus37pixelma: Yes, that's correct.
17:19:56LambdaCalculus37Strife89: So far so good, but make those revisions that pixelma pointed out.
17:20:04pixelmaStrife89: keymap, not keypad
17:20:11wpyhsaratoga: I'm not sure how I should test the Fuze. there seems to be 2 versions, your version and funman's version
17:21:46saratogawpyh: i didn't really understand funman's changes so I'm not sure
17:21:58saratogayou could just try both and see if either gives the expect values when reading from the disk
17:23:46pixelmaStrife89: but a good start and the explanantion you have in the patch is also important. If you want to help the possible committer a bit, make sure the lines aren't (much) longer than 80 characters - and add the new rockpaint.tex to manual/plugins/main.tex (which works a bit like SOURCES in the rest of the code)
17:24:44CIA-38New commit by 03gevaerts (r20998): It's no longer possible to switch to charge-only mode by pressing a button, so remove that bit from the manual
17:25:31 Join ender` [0] (
17:25:50wpyhwell, your version yielded the correct results, and funman's version too (I've tried with 7964671 and 7964672)
17:26:57wpyhhowever, if I go ahead and make some more changes funman suggested (use wordbuf instead of buf2 to display the result), I get garbage
17:27:20saratogamine gives the correct result on an 8GB fuze?
17:27:39wpyhthe one with 8026112 gives the correct result
17:29:50saratogacan you post your patch?
17:31:07 Quit robin0800 (Read error: 104 (Connection reset by peer))
17:31:16wpyhwell, I'm using this:
17:31:30wpyh(that's the one you posted on the forum)
17:31:41saratogayes but you said you made changes that broke it
17:32:40wpyhI always get DIR:0 and DIR4:1
17:32:55***Saving seen data "./dancer.seen"
17:33:53wpyhah, that part
17:33:56wpyhI changed
17:33:58wpyh+ _DEBUG_PRINTF("DIR0: %x DIR4: %x", *(buf2+512), *(buf2+512+4));
17:34:02wpyh+ int *wordbuf = (int*) buf2;
17:34:02saratogayes the part i don't already have
17:34:04wpyh+ _DEBUG_PRINTF("DIR0: %x DIR4: %x", wordbuf[512/4 - 1], wordbuf[512/4]);
17:34:06wpyhper funman's suggestion, and I get garbage for DIR0 and DIR4
17:38:38wpyhwell, the code is actually on the forum...
17:39:26*wpyh doesn't understand what saratoga needs
17:39:39saratogaa patch so i can see if you're doing this correctly
17:40:22 Join robin0800 [0] (
17:40:30wpyhah, ok
17:41:31 Quit ender1 (Read error: 110 (Connection timed out))
17:45:34 Quit einhirn (Read error: 104 (Connection reset by peer))
17:45:56 Join einhirn [0] (
17:49:53wpyhgood patch:
17:49:53wpyhbad patch:
17:52:54Strife89Okay, I've got several keymaps in, and I've converted it to a patch.
17:53:06 Join Ricky26 [0] (
17:53:34Strife89I'm still working on it, I just figured I'd post it so that if I made a big mistake, it can be noticed before I finish.
17:53:42Ricky26Anyone in here know which settings the Zen V needs in IDA Pro? :P
17:56:31 Quit wpyh (Read error: 104 (Connection reset by peer))
17:57:20LambdaCalculus37Strife89: Only one thing: some of the lines are rather long, longer than 80 characters, even.
17:57:45 Quit petur ("work->home")
17:57:47Strife89LambdaCalculus37: So, shorter descriptions?
17:57:56Strife89Or split lines?
17:58:02LambdaCalculus37Strife89: Split some lines.
17:58:12LambdaCalculus37Try not to make them longer than 80 characters if you can.
17:58:23Strife89Will do. :)
17:58:31LambdaCalculus37Cool. :)
17:58:41LambdaCalculus37Post a new patch once you're done.
17:59:33Strife89Alright. :)
18:00:23 Join ender1 [0] (
18:01:55 Join JdGordon| [0] (i=4601d1d7@gateway/web/ajax/
18:02:02Strife89Is it okay if several lines are a few more than 80 chars (85 maximum)?
18:07:25linuxstbStrife89: Just don't draw attention to them. Oops...
18:07:59*Strife89 just sticks with a strict 80-char-per-line limit.
18:08:05Strife89Just to be safe.
18:08:55 Join mt [0] (n=MTee@
18:12:00Ricky26Yay, I now have a Zen V with a .png instead of firmware. :D
18:13:40 Quit robin0800 (Read error: 110 (Connection timed out))
18:16:30 Quit ender` (Read error: 110 (Connection timed out))
18:17:42 Join PaulJam [0] (
18:21:46Strife89FS #10202 has been updated. All keymaps are in.
18:23:19LambdaCalculus37Strife89: Did you test to see if the manual builds correctly with it?
18:23:25Strife89LambdaCalculus37: Not yet...
18:23:46LambdaCalculus37Strife89: If it does, I'll commit it later on when I have net access on my MacBook.
18:23:51 Join _Auron_ [0] (
18:24:11 Part Grahack
18:24:27Strife89I forget how to build a manual.
18:24:53LambdaCalculus37Strife89: Hang on, let me find the page for that.
18:25:09Strife89I know I need to get some LaTex packages, but I don't know their names. >.<
18:25:32 Join Hillshum [0] (n=chatzill@unaffiliated/hillshum)
18:25:58 Quit Ricky26 ()
18:26:14Strife89LambdaCalculus37: Alright, I'll read up and report back.
18:26:26 Nick hd is now known as HellDragon (i=jd@Wikipedia/HellDragon)
18:27:29LambdaCalculus37Strife89: If you're using Linux currently, just install the needed packages on that page (the tetex-* sets), and you'll be set.
18:27:42Strife89I have a lot in packages to get, so this'll be a while.
18:27:54Strife89My download speed is 100 KB/s.
18:28:02LambdaCalculus37No worries, we'll be here. :)
18:28:36*Strife89 gets some lunch while he waits.
18:29:16 Join archivator [0] (n=archivat@
18:29:20 Quit _Auron_ ("Infinity repeatedly denies rumours of plotting with zero to bring down the Universe.")
18:36:37 Quit renke ("leaving")
18:38:14 Quit HellDragon (Client Quit)
18:38:35 Quit JdGordon| (" ajax IRC Client")
18:43:20 Join JdGordon| [0] (i=836b0070@gateway/web/ajax/
18:43:22 Join HellDragon [0] (
18:44:48 Quit matsl (Read error: 110 (Connection timed out))
18:46:07 Quit PaulJam (Read error: 113 (No route to host))
18:54:22 Quit archivator ()
18:55:41 Join archivator [0] (n=archivat@
18:57:20CIA-38New commit by 03gevaerts (r20999): Don't force double-buffering for sd devices. They apparently are not faster with it, ans possibly actually slower
18:57:54mtlinuxstb : Apparently it's better to modify the test program first. :)
18:59:57 Join petur [0] (n=peter@rockbox/developer/petur)
19:00:29mtBut I'm still thinking about how I'd approach this. Currently, I think I'll write get_uint* which are similiar to read_uint* in all but 2 aspects : 1. They'll take in a memory buffer. 2. They will not entail copying of data.
19:02:41 Quit saratoga (" ajax IRC Client")
19:02:53mtAnd they'll have to take a "pointer to the pointer to the memory buffer. " to be able to advance the pointer with the amount of the data read.
19:03:45mtWhat do you think ?
19:06:25 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
19:07:03gevaertsdo those things read 1 int at a time? If so, is copying a pointer less expensive than copying the value?
19:09:21 Quit archivator (Remote closed the connection)
19:14:36 Part wincent ("Kopete 0.12.7 :")
19:15:44 Join bzed_ [0] (
19:17:59mtgevaerts : no .. there would be get_uint8, get_uint16be and get_uint32be
19:18:49*gevaerts should probably shut up if he doesn't know what's being talked about :)
19:19:10 Join steiny [0] (
19:19:39steinyIs there a known issue with iPod 3Gs and rockbox 3.2 and up? I have weird bahviour :/
19:21:53gevaertswhat sort of issue?
19:22:52steinythe keys seem to behave odd
19:23:15steinyfor instance. I have to touch the forward key up to 6 times till it enters a menu and the central select button does absolutly nothing anywhere
19:23:31steinymind you, it works in apples original firmware ... all buttons act on first touch, including select
19:23:58steinywith one oddity: since rockbox is on it, apples firmware starts a new song every time I hit select to navigate its menus .... very odd
19:24:09gevaertsI haven't heard of anything like this
19:24:22steinyyes, I feel quite lonely with that issue :)
19:24:36steinyI assume select should enter me into a rockbox menu, not the foreard button
19:24:37*gevaerts isn't an ipod specialist, so he probably doesn't know all known issues with them
19:25:09steinyi soon also get a H120 .... I thought of RB there, but now I'm reluctant .. the H120 is much more valuable to me than this 3g
19:25:09gevaertsyes, it's select
19:25:33AlexPsteiny: To see what the buttons should do, you can have a peak at the manual. Having said that, select seems sensible here :)
19:25:51amiconnBoth buttons will enter a menu item /submenu
19:25:53AlexPsteiny: I have had Rockbox on a H140 for years with no issues
19:25:58steinyalexp: yeas I .. er ... "read" it (overflown is maybe more true ... I admit) :)
19:26:32steinysomething is clearly b0rken, though
19:26:50 Quit bzed (Read error: 110 (Connection timed out))
19:26:50 Nick bzed_ is now known as bzed (
19:27:08AlexPsteiny: Part of the problem may be that next to nobody has a 3G
19:27:34steinywell ... it is old, yep :(
19:27:59gevaertsold isn't the problem :)
19:28:02LambdaCalculus37steiny: It's not really a matter of being old.
19:28:06*LambdaCalculus37 is too slow ;)
19:28:19AlexPI just mean that with very few users it doesn't get much testing, and with very few devs with one it is harder to fix any problems that may come up
19:28:20LambdaCalculus37steiny: It's just that very few developers or staff have one.
19:28:44steinyah ... bummer
19:28:49steinyhow about the H120?
19:28:56AlexPsteiny: For instance nobody has reported this fault before, and if no devs have a 3G (or a 3G with this problem) then it is very hard to fix
19:29:01gevaertse.g. the Archos Player is a lot older, but nearly everyone has it, so it works well :)
19:29:03AlexPsteiny: Lots more have those
19:29:13AlexPAs they are much better :)
19:29:28LambdaCalculus37AlexP: They are *the* player to have. :)
19:29:33steinyhm ... I'd be heartbroken if the H120 gets killed with some bootloader issue ... I went to trouble to get hold of one recently
19:29:55AlexPsteiny: If you use a released bootloader, then it should be absolutely fine
19:30:00gevaertsif you stick to the official released bootloader, you shouldn't have any problem
19:30:06 Join miepchen^schla [0] (
19:30:19steinyokies ... I'm not planning on the CFlash bootloader and all .... yet .... I'm scared of a nonbooting iriver
19:30:29AlexPThere hasn't been a single broken H100 when using the released version to my knowledge
19:30:58steinyneat to know
19:31:20steinyit's so sad nobody makes players with the H100 series soundwuality AND optical out anymore
19:31:25AlexPOn most targets you can recover easily from bootloader issues - because the H100 and H300 can be bricked with a bad one, any released ones undergo *a lot* of testing by people with specialist recovery equipment before release
19:31:55steinygood :)
19:32:08steinyI kow people hoarding dozesns of H100's ... it's so hard to get one
19:32:38steinyI won mine on ebay, because I never do ebay and I was not realizing a bid has a confirm ...
19:32:46 Join bertrik [0] (
19:32:47steinyI frantically hit confirm and made it in 2 seconds before the end ....
19:32:58***Saving seen data "./dancer.seen"
19:33:28steinyif I have to do that again, I'll die from heart attack
19:33:49AlexPhehe - I think a H100 is worth it :)
19:34:06steinygot it for 110 US dolalrs .... I hear that's not a bad price
19:34:16gevaertsthat's an excellent price
19:34:34steinyit supposedly mint, only 2 days used and all accessoires included
19:34:48steinyI'll see if that is true or not in a few days
19:34:50AlexPThey usually go for loads more than that
19:35:01steinyI'm worried about its battery tho, if nobody used it
19:35:09AlexPThey die anyway
19:35:15AlexPBut replacements are fairly easy
19:35:35AlexPYou can put an ipod 1G/2G in it, but you need to switch the leads
19:35:36steinywell I was prepared to shell out up to 220 USD ... i was surprised nobody bid along ... maybe cause of my bid confirm sillyness
19:36:01steinyyep I saw 2200 one son ebay ... but I'll opt for a non bad one .. those all get balloney
19:36:09 Join evilnick_7 [0] (i=0c140464@gateway/web/ajax/
19:36:22AlexPBuy a named brand like Cameron Sino
19:36:29steinyintedn to, yes
19:36:43steinyalready got me a usb/charge cable from boxwave as well
19:36:58steinyeven tried hunting for the 40 GB toshiba, but that seems pointless
19:36:58AlexPI have a 2200 mAh battery in my H140 and last time I tested the runtime (a long time ago) it was still going after 29 hours when I turned it off (under Rockbox of course)
19:37:20steinywith a harddrive or CF?
19:37:27AlexPhard drive
19:38:37steinyin theory I don't need RB i guess ... i do VBR mp3 ... I tried flac, mp3, wav comparisons at high bitrates .. I cant hear the diff in rap and trance etc :)
19:38:54steinynot even by optical feed into the iBasso D10 and beyerdynamic DT880s
19:38:58steinymaybe I'm deaf
19:39:14 Join einhirn [0] (
19:39:21evilnick_7It's also very nice to make dynamic on-the-go playlists
19:39:25 Join Jaykay [0] (
19:39:40gevaertswell, the iriver firmware doesn't have doom
19:39:50steinymmh with my bugged ipod 3g, I could not create them ... bad keyboard
19:39:58AlexPsteiny: There is much more to it than extra sound formats
19:40:06steinyafter crysis warhead on a gtx 285
19:40:21AlexPI would never use the H100 OF now - it would drive me mental
19:40:28steinyalexp: I know ... I'm happy to explore that :)
19:40:45steinyoriginal H100 is not very intuitive I hear
19:41:20steinya nice EQ would be nice ... as far as I know, RB has it
19:41:26steinynot sure about original H100
19:41:31*evilnick_7 bought an H140 purely for Rockbox and hasn't used the OF much if ever. Maybe to flash to Rockbox right at the start!
19:42:09AlexPI had mine before Rockbox was ready for it - the start up time alone was a killer
19:42:27AlexPsteiny: Yes, Rockbox has a five band parametric equaliser
19:43:08steinyand I noticed it doesn't automatically jump to a next file it seems ... all purely playlist based?
19:43:35LambdaCalculus37steiny: If you would like to donate your iPod 3rd gen to a developer, you're more than welcome to. Just don't expect results right away.
19:43:46LambdaCalculus37Things like Rockbox take time to make perfect. ;)
19:43:56evilnick_7It will create a playlist out of the files in the current directory, and then play through that.
19:44:27steinylambda: once I have my H120 I would not even mind ... as long as I get it back functional in some way :) I have sentimental value with it
19:44:34steinyer it has S V to me
19:44:50steinyev: ah okay
19:45:04evilnick_7I assume that you want this for recording?
19:45:18steinycan you create playlists for it on the PC? I mean ist standard format and USB HD and all ... but how about the relational filepaths?
19:45:32steinyev: The H120?
19:45:40evilnick_7Yes, it uses standard m3u8 files
19:45:41Strife89Making seems to have failed.
19:46:08evilnick_7And also will strip away one "directory level" at a time if it can't find the music file as specified in the m3u8
19:46:12LambdaCalculus37steiny: Then if it's something sentimental to you, then I suggest not donating it. Donated players are usually taken for things like hardware scans and such, and as a result may not come back intact.
19:46:28steinylambda: eeww
19:47:08evilnick_7So that; E:/Music/Beatles/Sgt Peppers/A day in the life.mp3 if it didn't exist on the target...
19:47:11AlexPsteiny: I'd really suggest reading the manual - it tells you all about what Rockbox can do
19:47:40evilnick_7...Rockbox would look for Music/Beatles/Sgt Peppers/A day in the life.mp3 and then Beatles/Sgt Peppers/A day in the life.mp3 etc.
19:47:45steinyAlexP: I know, you're right ... :)
19:48:07steinyev: I see, thanks
19:48:26steinyrecording I wont do ... I wasn't even aware before that the H1x= can
19:48:38steinyI use it for optical out into the iBasso D10
19:48:39evilnick_7Why do you specifically want that model then?
19:48:48evilnick_7Aha, optical out! I see
19:49:32evilnick_7But if you can't tell the difference between mp3 and flac, is it worth having the optical output?
19:49:50steinythat'lll exactly be my setup, even the same cable ... apart from the picture showing a D1, not a D10
19:50:04steinyevil: I think so ... it seems to sound crisper
19:50:14steinyeven for noisy sound like trance, rock etc
19:50:34 Join _Auron_ [0] (
19:50:43Strife89LambdaCalculus37: Looking at the PDF, it seems to have built correctly.
19:51:11steinyare some of you RB devs btw?
19:51:19evilnick_7I'm all for hi-fi sound, but I don't quite understand why you'd be more interested in using optical-out (for the benefit) than lossless.
19:51:57steinyevil: I only tried trance so far ... I'll have to review various loreena mckennit things as ell, I expect more diff there
19:52:01LambdaCalculus37Strife89: Can you email it to me?
19:52:08steinyevil: and: part of it is pure geek value of kmnowing to own the optimum :)
19:52:17Strife89LambdaCalculus37: Sure, just a sec.
19:52:52evilnick_7:) I hear you, but it might be a lot cheaper to go lossless
19:52:53steinyi'm not audiphile, but I do like a solid setup
19:53:21steinyI already own all these things ... too late for cheap :)
19:53:23AlexPsteiny: Yes, many people in the channel are
19:53:33steinyalexp: are you?
19:54:30Strife89LambdaCalculus37: Sent.
19:54:41AlexPsteiny: Only as far as documentation and support goes
19:54:44steinyevil: I liek all the comparing, hunting, checking, owning bits of all this technostuff .... so ... :)
19:55:16steinyah you are not worth my attention then, I only let true devs bathe in the light of my presence :)
19:55:16evilnick_7steiny: Looks like we're drifting off-topic for this room
19:55:38steinybut to be serious: Thanks a lot for that product ... it's great fun (apart from my odd 3g issues)
19:56:10steinyoh excuse me, I wasn't intending to interrupt compiler talk
19:57:02evilnick_7steiny: This channel is logged (specifically for developers)
19:57:03LambdaCalculus37Strife89: Checking it out now.
19:57:33steinyev: IC .. I wasn't aware there were several RB channels, sorry .... got sent here by some forum :/
19:58:12 Quit einhirn (Read error: 104 (Connection reset by peer))
19:58:59LambdaCalculus37Strife89: It built correctly, save for the warning regarding no screenshot.
19:59:09LambdaCalculus37And it looks good, to boot. ;)
19:59:16Strife89LambdaCalculus37: :)
19:59:17gevaertssteiny: support talk is welcome here, but lots of people read the logs, so having lots of random chat makes them lose time. Also, if you want to highlight someone, use the full nick
20:00:02LambdaCalculus37Strife89: Do you want to finish it off with some screenshots?
20:00:13Strife89LambdaCalculus37: I'll try.
20:00:20Strife89It'll take a while...
20:00:30LambdaCalculus37Strife89: No worries. Take your time.
20:00:57steinygevaerts: Okies, I'll try to cut down the babble :)
20:01:01LambdaCalculus37Attach them to the FS entry for your patch and I can commit the entire lot.
20:01:42Strife89LambdaCalculus37: Attached as images or a single patch.
20:02:22AlexPStrife89: They need to be the images (but if you zip them it'll be quicker to download)
20:02:45AlexPStrife89: Then the person committing it has to svn add them and set as binary - you can't do that in a patch
20:02:59Strife89AlexP: Oh, okay then.
20:03:23*Strife89 looks for the list of sims.
20:03:46 Join Toeizik [0] (
20:04:05 Quit Toeizik (Client Quit)
20:05:15 Quit jfc (Remote closed the connection)
20:05:58 Join ricky26 [0] (
20:06:37 Join kugel [0] (i=kugel@rockbox/developer/kugel)
20:07:50 Join jfc [0] (
20:08:42bertrikhi FlynDice
20:08:49FlynDicehi there
20:09:39FlynDiceI remember you were talking about running the ams sansa's below 200 MHz for lower voltage
20:10:19FlynDiceand I've got a 192 max freq running right now. How much of an advantage can we get. do you know?
20:11:04FlynDiceand what would be the best way to test any ideas
20:11:44 Quit Zambezi (Connection timed out)
20:11:58 Quit evilnick_7 (" ajax IRC Client")
20:12:50bertrikFlynDice, I think we can reduce processor current to about square(lower voltage / original voltage) of the original current
20:13:19bertrikI didn't mean we _should_ do that, just pointing out that we could
20:14:32 Join Zambezi [0] (
20:14:49FlynDiceI'm just trying to look at the tradeoff. The performance seems just fine at 192 and I remember you mentioning the square of the voltage was the power savng factor
20:16:48 Quit barrywardell (Remote closed the connection)
20:16:56kugelI think we should look into reducing voltage for unboosted states, not for boosted
20:18:02 Quit n1s (Read error: 104 (Connection reset by peer))
20:18:28FlynDicekugel: So you think we should adjust the voltage on the fly?
20:18:32 Join n1s [0] (n=n1s@rockbox/developer/n1s)
20:19:24FlynDicekugel: or I guess dynamically might be the right term?
20:19:47kugelif that's possible, yes
20:20:12 Join Thundercloud [0] (
20:20:57 Join PaulJam [0] (
20:21:47FlynDicekugel: Do you think we could make just make the change at the same point we reduce the frequency to unboosted?
20:24:11kugelbut probably with a little delay
20:25:46 Quit JdGordon| (" ajax IRC Client")
20:25:59 Quit flydutch ("/* empty */")
20:31:00Jaykaywhat si songdbj ?
20:34:06Horschtiirc, it's a tool for PC that can build a rockbox tagcache/database
20:34:29 Quit Zambezi (SendQ exceeded)
20:34:44 Join tessarakt [0] (
20:35:21ricky26Hrrm, anyone have much experience with hacking the Zen V?
20:36:14 Quit tessarakt (Client Quit)
20:39:21Jaykayand why is songdbj still in the rockbox repository?
20:39:39 Join renke [0] (
20:39:43 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
20:44:24 Join tomers [0] (
20:50:20AlexPJaykay: Why shouldn't it be?
20:51:05Jaykayit's obviously not used since two years
20:51:38AlexPSo why should it be removed?
20:52:32Jaykaywhy should it stay there?
20:52:36AlexPEven if it is currently a little out of date, surely it is good to have there so someone can fix it up if they want?
20:53:15AlexPJaykay: Why do you want to remove stuff for the sake of it when it is useful?
20:54:52Jaykayi can't say whether it's useful or not, but when most people don't know what it is and it wasn't edited in the last two years, i would say it's not useful
20:55:19Jaykayi just asked why it's there. if it helps you leave it there
20:55:45AlexPJaykay: I suspect most developers do know what it is for
20:56:00AlexPAnd just because something hasn't changed in two years means nothing
20:56:27AlexPIt might just mean that it works well and hasn't needed updating
20:57:20Jaykaywhat does it do exactly?
20:57:34AlexP[20:36:01] <Horscht> iirc, it's a tool for PC that can build a rockbox tagcache/database
20:57:52HorschtThere's a FS patch that makes it compile and work...
20:58:25HorschtFS #9371
20:58:29scorcheJaykay: typically things are left (as in rarely removed) in code repositories to satisfy GPL requirements and such as well
20:58:33bertrikHorscht, where can I find this tool? I found but that is a script
20:58:41Unhelpfulamiconn: if i understand how it works correctly, a loop like the one at bottom of scale_v_linear might be a good candidate for using movem on coldfire.
20:58:56Horschtbertrik, i am about to compile it on windows
20:59:36JaykayAlexP: thanks... how does it help?
21:00:01 Join Zambezi [0] (
21:00:08AlexPJaykay: What do you mean?
21:00:30AlexPIt lets you build the RB database using your PC
21:00:36AlexPTherefore it is much faster
21:01:01bertrikmeh, this thing is all over the rockbox source code
21:01:17Jaykayah ok, i didn't understand the first explanation
21:01:25 Join robin0800 [0] (
21:01:51Unhelpfulget rowinc and two copies of rowval into address registers, along with an end address, then movem into 4 data registers from rowinc and rowval, do the four additions, movem the results back to rowval (via the second copy of the rowval address), with movem doing all the address increments for us as well
21:02:55bertrikHorscht, does it replace the songdb script? (does that one work actually)?
21:03:37HorschtI don't know if the script works. The songdb.exe works fine, though (the current one I have)
21:04:04HorschtI am about to build a new one, though
21:09:05bertrikHorscht, so currently the songdbj tool does not work and the patch makes it work, do I understand correctly?
21:10:01Horschthm... i might be confusing something... I can't find songdbj
21:10:55Strife89LambdaCalculus37: Okay, I just sent you an e-mail with the screenshots.
21:12:08 Quit Thundercloud (Remote closed the connection)
21:12:32LambdaCalculus37Strife89: Thanks. I hate to be a pain, though, but can you also attach them to your Flyspray ticket as well?
21:12:47Strife89LambdaCalculus37: Not a problem. :)
21:13:10Strife89LambdaCalculus37: I've been under the impression that _I'm_ the pain, with all the stuff I ask.....
21:14:15Strife89LambdaCalculus37: Done. :)
21:14:23Horschtbertrik, all I can find is tools/ and tools/database/. The database tool does not work, but FS9371 creates a /utils/songdb/ which is doing the same job as /tools/databse/, except it actualy works ;)
21:14:28LambdaCalculus37Strife89: Nah, we give special exceptions to the Padawans. :)
21:16:03UnhelpfulLambdaCalculus37: did you have any luck with those benchmark plugins? they failed to work properly for gevaerts on x5 until i sent him a full build :/
21:16:27LambdaCalculus37Unhelpful: They didn't work for me, either.
21:16:50 Join root [0] (i=root@
21:17:18 Nick root is now known as Guest62396 (i=root@
21:17:33 Nick Guest62396 is now known as jpro (i=root@
21:17:34 Nick jpro is now known as jpros (i=root@
21:17:49LambdaCalculus37Strife89: I think it's safe to say that your patch is ready to commit. Did you build the manual one last time to check that everything's in order?
21:18:14Strife89LambdaCalculus37: Sure, lemme do that real quick....
21:18:30jprosI want to put the rockbox firm in my pmp
21:18:39Strife89LambdaCalculus37: I'll SVN-add my own screenshots first, so hang on.
21:18:49pixelmaStrife89: what was the FS# again
21:18:49jprosit's a sunplus slide pmp
21:18:52jproswith arm9
21:18:58Strife89pixelma: 10202
21:18:58LambdaCalculus37pixelma: FS #10202.
21:19:06jprossomeone know if it is possible?
21:19:21LambdaCalculus37Only players that are listed on the front page will work with Rockbox.
21:19:48jprosit's hard to modify for my pmp??
21:20:14jprosreally I need to make or modify a firmware that have a holly bible
21:20:20pixelmaStrife89: you could combine all the \ButtonUp etc. in one \opt
21:20:26jprosI'm from brazil, so sorry for my english
21:20:30martian67whats major and new since 3.2 besides jpg album art support?
21:20:32pixelmawhen it's the same
21:20:45martian67and usb stacks on portal player
21:22:18LambdaCalculus37jrpos: I suggest you start by reading
21:22:26jprosLambdaCalculus37 tks :D
21:22:59 Quit ricky26 (Remote closed the connection)
21:23:14pixelmaStrife89, LambdaCalculus37: and CREATIVEZVM_PAD and the like don't exist for the manual yet - it won't break the manuals now since the opt is never true, but might prove a pain if someone decides to name it slightly different later on
21:23:55Strife89pixelma: "The like" means the Philips?
21:23:58LambdaCalculus37pixelma: I'll remove CREATIVEZVM_PAD before it gets committed, then.
21:24:09Strife89I was using the names provided in the plugin.
21:24:34Strife89LambdaCalculus37: Don't worry about it, I'll make one last patch.
21:24:53LambdaCalculus37Strife89: No, that's good that you followed along with it, but pixelma's right... the ZVM and GoGears have no manuals yet, so attempting to add those will break things.
21:25:55pixelmaalso the Fuze
21:25:56LambdaCalculus37pixelma: You know the manual layout a little better than I do... is Strife89 pointing to the right location for screenshots?
21:26:18LambdaCalculus37Because IIRC it should point to manual/plugins/images
21:26:28 Join Ridayah [0] (
21:26:28 Quit steiny ()
21:27:06pixelma"Opening A File" and the other subheadlines need to be subsubsections, otherwise they'll show up as own "plugins"
21:27:31Strife89pixelma: So how would I type that?
21:27:52pixelma\subsubsection instead of \subsection
21:28:02pixelmathe rest is the same
21:28:30Strife89Anything else that needs doing? :)
21:29:17pixelmathe tools could be itemized (would have to look myself for a reference, I guess rockboy could have something like it)
21:29:22Horschtworld peace...
21:31:24Strife89pixelma: I'll try that and see how it looks.
21:31:26pixelmaStrife89: as I said already, some opts could be combined but that would just make reading of the code easier - should already work as is
21:32:35pixelmaotherwise it looks good :) I would only test-compile a few manuals
21:32:42LambdaCalculus37Strife89: Once you've got the code cleaned up, post a new patch.
21:32:57Strife89LambdaCalculus37: Will do.
21:32:59***Saving seen data "./dancer.seen"
21:33:10Strife89pixelma: I went ahead and did that. :)
21:33:11LambdaCalculus37pixelma: He test compiled a Sansa c200 manual for me which looked good, apart from the missing screenshots.
21:33:25pixelmaoh, just saw one more - maybe the "Be careful" bit could be a note or a warning
21:33:35LambdaCalculus37pixelma: Thanks for the assistance in going over the code. :)
21:34:12 Join ricky26 [0] (
21:35:02ricky26Is there a hard reset on the Zen V anywhere?
21:35:09Strife89pixelma: I hate to bug, but how do I make it a warning?
21:35:28pixelmathe reference to the screenshot looks alright and the naming of the files too
21:37:05AlexPStrife89: \warn{This is a warning}
21:37:07ricky26bye bye little screw.
21:38:05 Quit blithe ("Lost terminal")
21:38:16 Join blithe [0] (
21:38:44 Join merbanan [0] (
21:38:54Strife89AlexP: What if I need to split it into two lines?
21:39:05Strife89The wiki doesn't address it....
21:39:09AlexP\warn{This is a
21:39:21Strife89AlexP: Okay, just checking. :)
21:39:39AlexPalways best :)
21:40:39pixelmaLambdaCalculus37: I always test compile a few different ones there can be differences (and usually I build all it applies to and one which it doesn't to see if the exclusion works - but that's me ;) )
21:41:19AlexPI always do a selection of those included and excluded by any opts used
21:41:36 Join froggyman [0] (i=47ba0b80@gateway/web/ajax/
21:42:18 Quit PaulJam (Read error: 113 (No route to host))
21:42:26Strife89Okay, making a test c200 manual.
21:42:56LambdaCalculus37Strife89: Screenshots in manual/plugins/images ?
21:43:05Strife89LambdaCalculus37: Yes, locally.
21:43:06 Join {phoenix} [0] (
21:43:12LambdaCalculus37Strife89: Okay.
21:43:46 Join BXCracer_ [0] (
21:44:27 Quit BXCracer_ (Client Quit)
21:44:32Strife89Hmm, that doesn't look right at all.
21:45:56pixelmaif you tell what...
21:46:25Strife89Hang on, I'm trying something.
21:46:46UnhelpfulLambdaCalculus37: if you don't mind, i put up a full build for ajbrv1 at
21:47:43Strife89pixelma: Basically the objects I "\item"ized are showing up scrunched together right before the keymaps.
21:48:06Strife89On one line.
21:49:27pixelmathe line looks like e.g.: \item[Color Picker] The N...
21:49:39pixelmamaybe you missed the ] there
21:49:59 Join LambdaCalculus3 [0] (i=44a0430d@gateway/web/ajax/
21:50:11Strife89pixelma: No, but I did move the descriptions to their own lines.
21:50:56pixelmathat shouldn't be a problem
21:51:11 Join BryanJacobs [0] (
21:51:29 Quit LambdaCalculus37 (" ajax IRC Client")
21:51:41 Nick LambdaCalculus3 is now known as LambdaCalculus37 (i=44a0430d@gateway/web/ajax/
21:52:16Strife89pixelma: In my terminal is says: "! LaTeX Error: Lonely \item−−perhaps a missing list environment."
21:52:53pixelmaif I see correctly in some other plugin tex, \item seems to need being within a \begin{description} - your items - \end{description}
21:54:01pixelmathe LatexGuideline page confirms
21:54:10Strife89Okay, trying again.
21:55:31Strife89There! That worked. :)
21:56:01 Quit ricky26 (Remote closed the connection)
21:56:04*Strife89 makes another manual.
21:57:55Strife89An iPod Video manual appears to have built correctly.
21:58:29 Join stripwax [0] (
21:58:58 Quit LambdaCalculus37 (" ajax IRC Client")
21:59:35Strife89As does a Gigabeat manual.
22:00:08 Join mcuelenaere [0] (n=quassel@rockbox/developer/mcuelenaere)
22:01:44mcuelenaereUnhelpful: does pictureflow use the LCD driver's YUV blit function?
22:01:58Strife89FS #10202 updated:
22:02:59 Part renke
22:03:24Unhelpfulmcuelenaere: no, it does all of the YUV->RGB during cache build, the cache entries are just bitmap data with a header for the dimensions and the axes transposed (to suit pictureflow's column-oriented renderer)
22:03:34Strife89pixelma: By the way, I've been liking the keymap patch. :)
22:04:29mcuelenaereUnhelpful: and these YUV->RGB transformations, do they use lcd_blit_yuv()?
22:04:37mcuelenaerenever mind
22:04:42mcuelenaerestupid question :)
22:04:56 Join kachna|lappy [0] (
22:05:08mcuelenaereI'm having some pictureflow image distortion
22:05:12Unhelpfula jpeg viewer *might* be a good candidate for that.
22:05:39Unhelpfulhrm. i have to go for a bit, but if you can get a description or a screenshot, i can take a look. :)
22:07:07mcuelenaereand the jpeg viewer shows the images just fine
22:07:12pixelmaStrife89: only one thing now - the \ButtonUp and \ButtonDown without an \opt at all look weird
22:07:22mcuelenaere(this is on MIPS target)
22:07:40Strife89pixelma: In what way?
22:07:44pixelmaI know it usually works for Left and Right as all keymap files have it
22:08:13mcuelenaereoriginal image:
22:08:20Strife89pixelma: What should I do, then?
22:08:34Unhelpfulok, i'll try to take a look at it later
22:08:50 Quit BryanJacobs ("null")
22:09:07pixelmaStrife89: in e.g. the Ipod manuals \ButtonUp (if that even exists) and \ButttonMenu would appear
22:09:10 Quit jpros (Read error: 104 (Connection reset by peer))
22:09:52Strife89pixelma: So "\opt{\ButtonUp}"?
22:10:33 Join p3tur [50] (n=petur@rockbox/developer/petur)
22:10:40pixelmayou could put it in a looong opt, listing all the targets it applies to - or in a \nopt with the Ipod pad and the H10 pad which you specify below with different buttons
22:12:37Strife89pixelma: One line or two, to keep with the 80-character thing?
22:13:09pixelmae.g. \nopt{IPOD_4G_PAD,IPOD_3G_PAD,IPOD_1G2G_PAD,IRIVER_H10_PAD}{\ButtonDown}
22:13:57*Strife89 went with the \opt option.
22:14:32pixelmabreaking the list inside the \opt{} is a bit tricky
22:15:19Strife89pixelma: With the \nopt , what does the \ButtonUp need to be?
22:15:24 Quit Jaykay ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
22:15:37 Join Ricky26 [0] (
22:17:03pixelmathe same - \nopt means excluding what's specified there. In your case everything that has its own \opt below
22:17:06 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
22:17:33*Strife89 is unsure.
22:17:47 Join jgarvey [0] (
22:18:03LambdaCalculus37Strife89: How's it coming along?
22:18:08Strife89LambdaCalculus37: Almost there.
22:18:27*LambdaCalculus37 prepares to add
22:18:37Strife89pixelma: Ah, I get it now.
22:20:15Ricky26Anyone have the files from, linked in the wiki?
22:20:23 Join Jaykay [0] (
22:21:04Strife89LambdaCalculus37, pixelma: Updated.
22:23:14pixelmasorry to be nitpicking but there still seem to be some references to keypads, the manual doesn't know about yet (Fuze, D2, Onda)
22:24:31Strife89pixelma: And the keymap table didn't build...
22:25:14 Quit _Auron_ ("Infinity repeatedly denies rumours of plotting with zero to bring down the Universe.")
22:26:50Strife89"Underfull \hbox (badness 10000) in paragraph at lines 106−−110. Misplaced alignment tab character &."
22:28:33AlexPStrife89: `There are always loads of thise warnings
22:28:52AlexPWhich is unfortunate as it makes spotting real warnings very hard
22:29:15Jaykaybtw what do these warnings mean/is there a way to eliminate them?
22:29:28Jaykayunderfull/overfull hbox
22:29:32Strife89I just built another iPod manual and the keymap table is gone.
22:29:42 Join _Auron_ [0] (
22:29:44AlexPWhat is your opt/nopt?
22:30:18mcuelenaereRicky26: what exactly do you want to know regarding the Zen V?
22:31:42Ricky26mcuelenaere: I want to write firmware for it. Just digging information at the moment.
22:32:00 Join Spoonman [0] (
22:32:43Ricky26I need to know how to get from code into a format that the rockbox tools can deal with. I have an arm toolchain set up, I have sendfirm, which works with the stock fw.
22:33:12 Join Ubuntuxer [0] (
22:33:28AlexPStrife89: Does it work for a different target?
22:33:38 Quit {phoenix} (Remote closed the connection)
22:33:41mcuelenaereRicky26: do you have any docs/information regarding the chipset?
22:34:04Ricky26They are the specs floating around on the Wiki.
22:34:11Strife89AlexP: Lemme build a Gigabeat manual...
22:34:24Spoonmananyone here can help me with button mapping in recording screen for ipods?
22:34:41Unhelpfulmcuelenaere: which target is this?
22:35:00AlexPSpoonman: For button mapping you want apps/keymaps/keymap-xxx.c
22:35:03mcuelenaereUnhelpful: Onda VX747
22:35:09Strife89AlexP: Nope, key table's gone for that target too.
22:35:26AlexPStrife89: Could you pastebin the section?
22:36:06pixelmaStrife89: the \begin{btnmap}{}{} seems to got lost
22:36:40AlexPStrife89: yes - what pixelma said
22:36:40mcuelenaereRicky26: I don't think there's much information except for the linux tarball regarding the STMP36XX chipset available..
22:36:50 Join Thundercloud [0] (
22:37:03pixelmaStrife89: also... there are evil tabs in your code ;)
22:37:06Unhelpfulanything weird i should know about the target? i see it's RGB565, which is the only color format PF knows about. the scaler uses some 32x32->64 unsigned multiplies, but i wouldn't expect gcc to compile that to anything that doesn't do what the math says...
22:37:19mcuelenaereand IIRC linuxstb has a mirror of
22:37:29mcuelenaereUnhelpful: it's MIPS
22:37:31Strife89pixelma: So, eliminate all tabs?
22:37:35AlexPSpoonman: If you look in apps/keymaps/keymap-ipod.c you will see a section starting "const struct button_mapping button_context_recscreen[] = {" - under that are the keymaps
22:37:43AlexPStrife89: Yes, it should be spaces only
22:37:56pixelmaStrife89: replace by spaces (usually 4)
22:38:13AlexPSpoonman: Keymap for the recording screen that is
22:38:35Unhelpfulyuv_to_rgb uses ints for some 0-255 values that get shifted left by 7, but that shouldn't be a problem even if int is 16-bit
22:39:02mcuelenaereint is 32-bit on this MIPS CPU
22:39:13Spoonmansorry, but i am not a programmer, so i want to ask if someone can do a change for me...
22:39:33mcuelenaereUnhelpful: are these functions shared among other things? (core jpeg decoder?)
22:39:48Unhelpfuli'm just trying to think of any way some value might overflow and wrap, or flip its sign bit, etc.
22:39:55Strife89AlexP, pixelma, LambdaCalculus37: Okay, changes made, making a test manual.
22:40:04AlexPSpoonman: What change? Also, the keymaps are not so bad to do
22:40:19mcuelenaereUnhelpful: the GCC version used for MIPS is 4.x.x, perhaps that matters?
22:40:23linuxstbSpoonman: Can you compile Rockbox yourself?
22:40:29 Quit BXCracer (Remote closed the connection)
22:40:38Spoonmanyes, but to set upa build environment and such things....
22:41:31AlexPSpoonman: What are you wanting to change and why?
22:41:38Spoonmanthe change should be to map record split from long play/pause wich is not really useable to long select...
22:41:41Unhelpfulmcuelenaere: the core jpeg decoder is used by pictureflow, but pictureflow uses a custom output format. if the cover appears correct in WPS, the first place i'd look is output_row_32_transposed_fromyuv in apps/plugins/pictureflow/pictureflow.c
22:41:57AlexPSpoonman: Long select looks to be used to go to the menu
22:41:58Strife89AlexP, pixelma, LambdaCalculus37: It works. I'll upload the patch one last time....
22:42:17Spoonmanbecause long play/pause effectively stops the recording
22:42:24*mcuelenaere looks up the WPS syntax
22:42:29AlexPShort select seems unused
22:42:39Spoonmaneaven better...
22:42:40linuxstbSpoonman: That sounds like a bug, rather than something you should just change for yourself...
22:42:45Strife89FS #10202 updated.
22:43:01Spoonmani filed a bugreport :
22:43:09AlexPIt looks like short press of play is pause and long play is split
22:43:19LambdaCalculus37Strife89: I'll check it out.
22:43:39linuxstbAlexP: Short select is probably "OK" - inherited from the standard context.
22:43:48linuxstbBut I don't know if that's used...
22:43:56AlexPlinuxstb: ah right
22:43:56pixelmaAlexP: long play (slightly differently) is also stop and that's the problem
22:44:05AlexPpixelma: yep
22:44:07Strife89LambdaCalculus37: Oops, I just noticed the iPod is missing one keymap....
22:44:17linuxstbAlexP: But no, it doesn't seem to be...
22:44:22Spoonmanok, i gonatest this, give me aminute
22:44:27Unhelpfulmcuelenaere: if you don't have a WPS with AA, you could also just try the test_core_jpeg plugin by adding it to SOURCES and trying to load a JPEG file with it.
22:44:35mcuelenaereUnhelpful: cover art works fine
22:44:38AlexPlinuxstb: No, so short select would seem to be sensible for split
22:44:43LambdaCalculus37Strife89: Quick, add the fix! :)
22:44:44pixelmashort select is a bit dangerous though if long select is used to go to the menu
22:45:01AlexPpixelma: Are there any other options?
22:45:20Strife89Ummm... "#define ROCKPAINT_QUIT ( ~BUTTON_MAIN )"
22:45:41Strife89I have no idea how to notate that in the manual....
22:45:47pixelmaI don't know - I just remember thinking stuff like that for the c200 map :\
22:45:58 Quit FlynDice (Remote closed the connection)
22:46:15Spoonmanactually long select does nothing at least on my nano
22:46:28AlexPHow do you get to the menu?
22:46:53linuxstb"menu" ?
22:46:53*Strife89 types "ButtonMain".
22:46:57pixelmaah... I think you can't
22:47:12 Part Ubuntuxer
22:47:23*linuxstb is very confused with the recording screen keymap...
22:47:25AlexPlinuxstb: recording settings menu
22:47:57*pixelma remembers talking about Ipod recording screen keymap issues with linuxstb when filling out button tables for the manual... quite some time ago
22:48:06Spoonmanit seems that now i cant get back to the menu at all, but i was able to do this before...
22:48:14linuxstbpixelma: Was I confused then as well?
22:48:34Strife89LambdaCalculus37: I have no idea what the key for exiting Rockpaint is on the iPods...
22:48:35pixelmaI think so
22:48:38mcuelenaeretomers: are you here?
22:48:47AlexPit is /me is confused too
22:48:57Spoonmanok, now it worked by short pressing menu
22:49:00tomersmcuelenaere: Yes, I'm here
22:49:22AlexPSpoonman: Did that not just quit back to the Rockbox menu?
22:49:34mcuelenaeretomers: I never got USB HID working without commenting out the if(active || ..) line in usb_hid.c, do you know why this could be?
22:49:39Strife89Would it be "ButtonHold"?
22:49:42Spoonmanok, on a long press of select u get to recording menue
22:49:53AlexPSpoonman: I meant (and I was very unclear, sorry about that) the recording settings menu
22:50:03Spoonmansorry, i was mistaken...
22:50:05mcuelenaeresorry if(!active || ..)
22:50:06pixelmaStrife89: my *guess* is that you can't quit directly with a button press on them, just through the menu
22:50:09tomersmcuelenaere: I'll check it out
22:50:17AlexPSpoonman: So does short select do anything?
22:50:44Strife89Oy, I wonder how I should put that in....
22:50:47mcuelenaereI verified that it's the !active part of the expression that makes the function return, but I don't see why it's set to false..
22:50:56Spoonmanbut while recording u get nowhere with select
22:51:04mcuelenaeres/why it's/why active is/
22:51:22 Quit Ricky26 ()
22:51:25pixelmalinuxstb: I think I was confused when looking at it too (for the c200). IIRC there were two buttons to start recording (the "split" does it as well) or somesuch
22:53:05mcuelenaeretomers: also do you perhaps know what could cause a HID command get sent infinitly repeated? (this is probably due to my recent USB driver-rework)
22:53:08 Join _lifeless [0] (n=lifeless@
22:53:13tomersmcuelenaere: Is USB HID driver gets enabled in usb_thread() (firmware/usb.c:298) ? If not, I wonder how it is that you can use it... Maybe it is not needed...
22:53:14 Quit __lifeless (Remote closed the connection)
22:53:16LambdaCalculus37Strife89: The code says ( ~BUTTON_MAIN )... I have no idea what button that is.
22:53:51Strife89Can I just \nopt the "Quits the program" text?
22:53:53tomersmcuelenaere: How do you sniff the USB traffic, or maybe you just see the key presses go crazy?
22:53:54tomersmcuelenaere: How do you sniff the USB traffic, or maybe you just see the key presses go crazy?
22:54:18mcuelenaeretomers: it is enabled on usb.c:298
22:54:43LambdaCalculus37Strife89: Yeah, might as well.
22:54:57mcuelenaeretomers: I haven't sniffed the USB traffic yet when it does this, but I see the volume going up to 100% while it didn't do that before (I can sniff it using usbmon+wireshark)
22:54:59tomersThe the active variable should be set by usb_hid_init() or usb_hid_init_connections()
22:55:00Strife89LambdaCalculus37: Testing a build,
22:55:26 Quit lymeca (Read error: 104 (Connection reset by peer))
22:55:31Strife89LambdaCalculus37: That works fine, making a patch.
22:55:31 Quit _lifeless (Remote closed the connection)
22:55:47 Join _lifeless [0] (n=lifeless@
22:56:10 Quit evilnick_230 (" ajax IRC Client")
22:56:14mcuelenaeretomers: they are set (I can see it in the logf debug menu)
22:56:43mcuelenaerethis is really weird, the only thing I can imagine that could cause this is memory corruption
22:56:48Strife89FS #10202 updated.
22:57:02Strife89LambdaCalculus37: Alright, I think it's ready. :)
22:57:10mcuelenaereor a memset/memcpy gone wrong
22:57:37 Quit Jaykay ("ChatZilla 0.9.84 [Firefox 3.0.10/2009042316]")
22:58:01LambdaCalculus37Strife89: Is it up?
22:58:11Strife89LambdaCalculus37: Yep.
22:58:17Strife89Patch version 9.
22:58:20tomersmcuelenaere: The 'active' variable can only be unset by usb_hid_disconnect... Since this is a very simple code, I guess you are right :-( What is the USB driver you use? I'll take a look on it tomorrow
22:58:43Unhelpfulmcuelenaere: i have to get read for work here. i'll try to put together a viewer for pfraw files, that should tell us whether the cache images are wrong, or if the problem is in the renderer.
22:58:45SpoonmanAlexP do you think we can map it to select since it seems to do nothing during recording?
22:58:48mcuelenaeretomers: firmware/target/mips/ingenic_jz47xx/usb-jz4740.c
22:58:49Strife89LambdaCalculus37: I didn't post version 8. ;)
22:59:21mcuelenaereUnhelpful: perhaps I should clear the cache and try again?
22:59:25AlexPSpoonman: Possible - I'm a bit confused about the recording screen mappings
22:59:37LambdaCalculus37Strife89: Okay, let me svn add on my end and test a manual build.
22:59:53Unhelpfulmcuelenaere: it might be worth trying, but the format of the cache files hasn't changed since well before JPEG.
22:59:53Strife89LambdaCalculus37: Alright.
23:00:02tomersmcuelenaere: Have you reviewed my code? Maybe it is there somewhere
23:00:13Spoonmani think they are from recording enhancement pack and made it into the build
23:00:31 Join {phoenix} [0] (
23:00:38tomersmcuelenaere: Have to go. Good night!
23:00:43AlexPStrife89: Maybe the .bmp should be in a \fname
23:01:09AlexPStrife89: Also, the manual uses UK English - Color should be Colour etc
23:01:58 Join Aevum [0] (
23:02:04 Part Aevum ("Leaving")
23:02:22AlexPStrife89: "+Draws and unfilled circle.
23:02:41AlexPI suspect the "and" should be an "an"
23:02:41 Join matsl [0] (
23:02:49LambdaCalculus37AlexP: I'll fix it up a little bit.
23:02:59*LambdaCalculus37 already is svn add'ing the files to his trunk
23:03:01Strife89LambdaCalculus37: I got it. :)
23:03:04LambdaCalculus37Okay. :)
23:03:15Spoonmanmaybe we could do a test build or something?
23:03:49Strife89LambdaCalculus37: Posted.
23:03:50AlexPStrife89: Why is "+Click on a color in the image to change the Foreground color to it." Foreground in capitals?
23:03:54 Quit matsl (Client Quit)
23:03:56mcuelenaereUnhelpful: what else except .rockbox/rocks/pictureflow/* do I need to delete to trigger a pictureflow cache rebuild?
23:04:06 Join matsl [0] (
23:04:30AlexPStrife89: All the other "color" need changing to "colour" too
23:04:33Strife89AlexP: it should be lowercase.
23:04:39Strife89AlexP: Done that.
23:04:48 Join ricky26 [0] (
23:04:53AlexPStrife89: Not in v10 you haven't
23:05:04Strife89AlexP: Whoops....
23:05:06 Join lymeca [0] (
23:05:26pixelmaStrife89: if you don't include the \\ in the \nopt, the Ipod manuals will get a blank line at the end - but I admit I had weird breakage once when trying to \nopt the last line of a table. Worth a try though
23:06:29 Join FlynDice [0] (
23:06:30AlexPSpoonman: That is an option
23:06:51AlexPStrife89: I'll wait for v11 before looking further :)
23:07:00Strife89pixelma: So take out the \\ ?
23:07:08Spoonmanok, could you do the build and sendme per email?
23:07:17AlexPNot right now
23:07:43pixelmaStrife89: no, put them before the } so after "immediately."
23:07:48Spoonmanno prob, when do you think you can?
23:07:55AlexPSpoonman: Maybe later - if you want it sooner you could set up a build environment :)
23:08:11LambdaCalculus37Strife89: There's also a lot of weird unexpected breaks all over the patch for rockpaint.tex; you may want to check your line endings to make sure they're Unix.
23:08:20Strife89pixelma: And leave the existing ones?
23:08:52Spoonmani dont have stress, it just should be done somewhen... my mail adress is
23:09:05Strife89LambdaCalculus37: I'll check with Notepad ++.
23:09:59CIA-38New commit by 03kugel (r21000): Sansa AMS: Centralise mapping of RAM and IRAM in as3525.h via #defines. This will hopefully make ongoing mmu work easier as less places need to be ...
23:12:48Spoonmanok, so im off now, and be back when i have tested and tell you my results... bye...
23:12:54pixelmaStrife89: hmm... the IPOD_1G2G_PAD isn't currently needed as the 1st and 2nd gen Ipods are greyscale but I think you can leave it there
23:14:08mtlinuxstb : Did you see my messages ? (sent about 3 hours ago).
23:14:23 Join Strife1989 [0] (n=michael@
23:15:13 Quit Strife89 (Nick collision from services.)
23:15:21 Nick Strife1989 is now known as Strife89 (n=michael@
23:16:06linuxstbmt: Yes. What code are you talking about? The "librm" parser for use in the codec?
23:16:36mtNo, the test program. I'll modify that first. :)
23:17:31linuxstbBut only part of that needs to use the codec API - the rest will be in core Rockbox, and you'll be reading from a file descriptior.
23:18:05Strife89AlexP, pixelma, LambdaCalculus37: That blank line is still there in the iPod manuals.
23:18:06linuxstbIf possible, you should try and avoid duplication - e.g. read the header and cook extradata in the metadata parser, so the codec parser just needs to parse the actual data packets.
23:18:13pixelmaStrife89: and about your question - don't leave the old ones. \\ mark a newline- if you exclude what's on the line it would be nicer if the newline was excluded too
23:19:08pixelmagrr... putting them inside seems to break building for me
23:20:11mtlinuxstb : 1. Can I fill an instance of RMContext in the metadata parser and send it to cook.c ?, and 2. "You'll be reading from a file descriptor." You mean in cook.c or librm ?
23:20:36Strife89pixelma: It didn't seem to work....
23:21:46Strife89FS #10202 updated.
23:21:52Strife89Revision 11.
23:22:48AlexPStrife89: If you are going to have one .bmp inside \fname you ought to have them all :)
23:22:59AlexPStrife89: Also, I imagine \fname{.bmp}s ends up looking odd
23:23:16 Quit Spoonman (" HydraIRC -> <- s0 d4Mn l33t |t'z 5c4rY!")
23:23:38AlexPStrife89: Maybe swap that for empty images or empty bitmaps?
23:24:27 Quit petur (Remote closed the connection)
23:24:39AlexPStrife89: What does "The NW-most tool" mean?
23:25:07AlexPStrife89: "+the foreground of the Colour Picker to the selected color.
23:25:09Strife89AlexP: NW=top left
23:25:16 Nick p3tur is now known as petur (n=petur@rockbox/developer/petur)
23:25:19AlexPThere is still a color on that line
23:25:45*Strife89 needs to stop and go eat......
23:25:49AlexPStrife89: That is not very clear - I didn't get NW to mean top left. I think it clearer to say top left
23:25:56pixelmaStrife89: you would have to remove the last two \ after the } - but it breaks building of an Ipod Video manual for me. Trying some things
23:26:46Strife89AlexP, pixelma, LambdaCalculus37: I need to take a break and eat...
23:26:53linuxstbmt: 1) No, but you can store some extra data in the "id3" struct. The wma codec does this (it uses the toc field, which is otherwise only used for mp3 files) 2) Neither - the get_metadata() function is in core Rockbox - apps/metadata/
23:26:58AlexPStrife89: Virtually all the "colours" are still "color" - I suggest doing a search as each time you seem to change one or two and leave the rest :)
23:29:23mtlinuxstb : So in modifying the test program, all I actually have to modify is how the audio packets are copied. (I could read the DATA chunk offset into the id3 struct and then advance_buffer(data_offset)).
23:30:27*pixelma feels a bit bad about all this nitpicking, Strife89 already did quite a lot of work ...
23:30:48AlexPIndeed, but it is best to put it in correctly :)
23:31:38 Quit Strife89 (Read error: 60 (Operation timed out))
23:31:52 Quit bertrik ("Leaving")
23:33:03***Saving seen data "./dancer.seen"
23:33:55*LambdaCalculus37 has to go too
23:34:10LambdaCalculus37Strife89 (for the logs): Let me know when you have the final version of the patch up.
23:34:16 Quit LambdaCalculus37 ("Fwump")
23:36:22pixelmawhat? calling the rockpaint menu on c200 is supposed to be on a combo of Select+Power to have the direct quit on a short press of Power
23:36:46*pixelma would have to take the blame for that one, probably
23:37:37pixelmait's not mentioned correctly in the manual patch though
23:38:11 Join GodEater [0] (n=nyeahrig@rockbox/staff/GodEater)
23:39:28pixelmathere are other not so nice things in the keymaps themselves, e.g. the "Toolbar2" won't work on the X5 and M5
23:39:38pixelmaerr... X5 only currently
23:42:06pixelmaspeaking of which - the IAUDIO_X5M5_PAD is still called IAUDIO_X5_PAD in the manual, so those opts will have no effect
23:42:08linuxstbmt: Yes, that should work.
23:50:24 Quit ender1 (" Be wary of strong drink. It can make you shoot at tax collectors and miss. -- Robert A. Heinlein")
23:58:18 Join Strife89 [0] (n=michael@

Previous day | Next day