Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2007-01-03

00:01:28Ajaxincwhat linux compilers you guys recommend?
00:01:30 Quit Lyle_ca (Read error: 104 (Connection reset by peer))
00:04:32 Join topbloke [0] (
00:05:30 Quit lee-qid ("aufwiederbyebientotsayonara")
00:05:54 Quit linuxstb (Remote closed the connection)
00:06:23 Quit topbloke (Read error: 104 (Connection reset by peer))
00:06:26 Join linuxstb [0] (
00:09:48 Quit cynicalliberal (Read error: 104 (Connection reset by peer))
00:12:13 Quit JoeyBorn (Read error: 110 (Connection timed out))
00:13:47FebsGrrr. How can I revert a commit.
00:14:28linuxstbCommit the reverse...
00:14:36dionoeasvn merge svnpath@newrev svnpath@oldrev .
00:14:48linuxstbcvs diff -u file.tex | patch -R
00:14:52linuxstbcvs commit file.tex
00:14:55dionoeaerr ... oops :)
00:15:05puetzkdionoea, not quite yet :-)
00:15:10linuxstbErr, that's not quite right either...
00:15:28linuxstbYou need to specify the revisions with -r
00:15:44dionoeadoesn't cvs have something to specifically revert a commit ?
00:16:07Bagderyes, with -j I believe
00:16:25Bagderi.e put the changes between to revisions into the file
00:18:28maquisdionoea: if you're asking about reverting all the files that were changed in a specific commit, i'm afraid not, unless i've missed something and wasted *way* too much time
00:18:46Bagderright, it can't do it for multiple files
00:19:28 Quit trypt0 (Read error: 54 (Connection reset by peer))
00:20:58 Quit hcs ()
00:24:16 Quit Febs ("Chatzilla 0.9.77 [Firefox]")
00:24:22linuxstbBagder: Could you move this to the download server (ipod bootloader directory) -
00:24:52Lloreanlinuxstb: He said he restored on a windows computer, shouldn't that mean it's properly converted. Is it the /disk1s2 thing? Should that just be /disk1 (don't know macs...)
00:25:07Bagderlinuxstb: done
00:26:27 Join Administrator [0] (
00:26:36preglowBagder: so, are we still up for svn?
00:26:43 Quit ender` (" Anyone who thinks people lack originality should watch them folding roadmaps.")
00:26:59 Nick Administrator is now known as Alonea (
00:27:19Aloneaok, I am confused. registered where?
00:27:21linuxstbBagder: Thanks.
00:27:43AloneaI just have a question on Doom...I can't get it to show up on the plugins list
00:27:46*preglow does the svn dance
00:27:52bluebrotherFebs, in case you read the logs: you can use \wikilink{PageTitle} ...
00:28:04 Quit ed_ ("Leaving")
00:29:45 Join webguest21 [0] (i=180d2205@gateway/web/cgi-irc/
00:30:06*Bagder leaves for sleep
00:30:10webguest21woah, I didn't get it to work on mIRC
00:30:17webguest21is the port 667?
00:30:30Aloneayeah, I just got rockbox and I was trying to get doom to work, but I can't get it on the plugin list.
00:30:31webguest21beok, well, 6667
00:30:39bluebrotherat least usually ;-)
00:30:44webguest21well, anyways, #rockbox shows up blank for me
00:31:01bluebrothermaybe a netsplit
00:31:10linuxstbAlonea: Which device are you running Rockbox on?
00:31:29AloneaToshiba Gigabeat F-40
00:32:19LloreanAlonea: Are you sure it's compiled for your target?
00:32:21linuxstbFor some reason, it's not being compiled for the Gigabeat..
00:33:02 Join Sentertai [0] (
00:33:04Aloneawell, everyone was saying at the mygigabeat site it was for it and its listed in the gigabeat I thought it was
00:33:12*linuxstb looks around for a Gigabeat developer
00:33:20webguest21finally, I got in as Sentertainment
00:33:31Ajaxincim not a developer but i know some infor abotu them cause i use a f10
00:33:34Sentertaieh, it cut off my name
00:33:34Aloneai have all the wads. I got that doom collectors edition disk.
00:33:48 Quit Sentertai (Client Quit)
00:34:02Aloneaand I did the wierd directories thing, but no mas on plugins list...
00:34:17 Join Shawn_K [0] (
00:34:18puetzkit would have been in the plugins list even without the wads
00:34:21Shawn_Kthats better
00:34:23Ajaxincdo you have the latest build?
00:34:26Shawn_KI'll just use my name
00:34:35Ajaxincdoom wasnt supported in some earlier cuilds
00:34:50Shawn_Kso...who are the rockbox devs? I've been dying to talk to them
00:34:56Shawn_KI want to help out with coding
00:35:06linuxstbAjaxinc: It's not enabled for the Gigabeat in the official CVS...
00:35:19linuxstbShawn_K: Just grab the source and dig in...
00:35:21Ajaxincwell it works on mine
00:35:22 Join hcs [0] (n=hcs@rockbox/contributor/hcs)
00:35:27Ajaxincand i use the cvs builds
00:35:31Aloneayeah, I have todays build. its all I could find
00:35:47linuxstbAjaxinc: Maybe you installed an unofficial build in the past, and it's left from that.
00:35:50 Join farbrormelon [0] (
00:35:52Shawn_Kyeah, I grabbed the source...I just was in the forums and had many oppsoing views on my ideas for optimization.
00:36:03Shawn_Knot too many for my ideas
00:36:07Ajaxincdamn you
00:36:11Ajaxincincompatable now
00:36:13Aloneashould I try the jan 1 build instead of jan 2?
00:36:14 Quit Mouser_X3 (Read error: 110 (Connection timed out))
00:36:16Ajaxinci havnt tested doom in a day :]
00:36:27Shawn_Ka few, but you could tell that they weren't experienced with coding
00:36:32Ajaxincyou might want to get the versions before the merge
00:37:07LloreanShawn_K: Many of the people who objected to you were core developers of Rockbox who are *very* experienced with coding.
00:37:16tehsmohcs: sorry, I was asleep earlier. Anyway, I'm pretty sure that itdb2tc isn't working properly
00:37:20farbrormelonHi I have some questions about the diskmode, someone want to help me?
00:37:29tehsmounless it's been updated very recently and my rockbox is out of date (I last updated it friday)
00:37:32Shawn_KLlorean: Hey, yeah...I realize that
00:38:08 Join safetydan [0] (i=cbca159f@gateway/web/cgi-irc/
00:38:08Shawn_KLlroean: I've had alot of people object my ideas, then find out they work after I do them
00:38:08linuxstbThen do them, and prove us all wrong...
00:38:08 Join ctaylorr [0] (
00:38:18LloreanShawn_K: Then do them, and ignore the objections, it's that simple.
00:38:18Shawn_KLlroean: I'm not really nieve, I just think differently at times
00:38:44linuxstbBut a guiding Rockbox principle is also to keep things simple...
00:39:06Lloreanfarbrormelon: The disk mode, on iPods, is not Rockbox software, it's Apple software.
00:39:12Shawn_KYeah, I understand
00:39:52Aloneaso any ideas on the gigabeat? is the doom not on it yet or do I need to try a different build? sorry if you replied and I missed it
00:39:57Shawn_Kbtw, just thought i'd ask...does this server support nick registration?
00:40:01LloreanShawn_K: Yes.
00:40:04farbrormelonsorry... but i just installed rockbox and now the filetransfer is slow
00:40:08bluebrotherembedded programming can be quite different than usual pc applications, and most people coming in with such "optimization" ideas don't know embedded stuff ...
00:40:13farbrormelonis this a rare problem?
00:40:24preglowwhat optimisations are we talking about?
00:40:26goffaso.. Shawn_K what did you have in mind?
00:40:30safetydanMan, real life really saps your ability to work on Rockbox. That and Guitar Hero.
00:40:36Lloreanfarbrormelon: Again, the disks mode is not our software. It's the apple disk mode.
00:40:36preglowsafetydan: long time no see
00:40:37linuxstbAlonea: Doom is not officially in Rockbox for the Gigabeat, no. You'll need to wait for a Gigabeat for the answer why.
00:40:41safetydanyo preglow
00:40:41scorchepreglow: malloc, among others
00:40:50Shawn_Kgoffa: oh, wow...what did I have in mind...alot
00:40:54Ajaxincalonea, go to #gigabeat and ask there
00:40:58farbrormelonSorry again...
00:40:58Lloreansafetydan: Guitar Hero is an evil, addictive thing.
00:41:03Aloneasafety dan: dear kami..guitar hero is addicting. that and DOA4
00:41:08Shawn_Kgoffa: I like to start small. so I at least know what I am getting into
00:41:13Lloreanfarbrormelon: Try booting into the Apple firmware instead of using the hardware disk mode
00:41:20goffaor do you have forum posts
00:41:29goffai'm just curious to hear...
00:41:30farbrormelonAh.. smart... =)
00:41:44Shawn_KI want to optimize the way alot of things are done, and examine the graphics rendering so I could maybe add caching and better drawing methods
00:41:51safetydanShawn_K, did you have a look at profiling to find places to work on?
00:41:52farbrormelonits menu+select then back, right?
00:41:54zehah guitar here is addicting
00:41:57goffaand worst case scenario.. you make your patches .. they work for you
00:42:03 Quit phrozen77 ("Some people say that I must be a horrible person, but that's not true. I have the heart of a young boy... in a jar... on my d)
00:42:05Shawn_Ksafetydan: Nope
00:42:08zei'm sitting here right now planning on playing it as soon as i finish eating
00:42:20Lloreanfarbrormelon: Just shut down your iPod, and then hold Menu while booting.
00:42:29linuxstbfarbrormelon: Best to do a clean shutdown (long press on PLAY), then turn the hold switch on, then insert USB.
00:42:31Aloneathank you ajaxinc and linux
00:42:35linuxstb(assuming you have the newest bootloader)
00:42:45LloreanShawn_K: What exactly were you planning to cache?
00:43:00LloreanThe term "caching" doesn't really specify what you wanted to add a cache for.
00:43:07preglowmalloc is evil
00:43:27Shawn_KLlroean+farbrormelon: I've had some trouble w/ pressing menu on some builds, I've found holding play with a few creates an error 5 to default to apple diskmode
00:43:43LloreanShawn_K: Holding Play tries to boot Linux.
00:43:59LloreanShawn_K: With the newest bootloader you can just boot up with Hold on.
00:44:07Shawn_KLlroean: no wonder, maybe the bootloader I used was bad then >.<
00:44:32safetydanSo, preglow. EQ graphing? :)
00:44:32Shawn_KLlorean: menu doesn't work for me on the patched co-op build I use
00:44:37safetydanI've forgotten where we got to with that.
00:45:02LloreanShawn_K: Menu is all about the bootloader, it doesn't matter what build you're using, just what bootloader you installed.
00:45:03Shawn_KLlorean: I used someone else's build since I was having minor toruble patching the co-op to mine
00:45:03preglowsafetydan: gave up on that
00:45:10preglowsafetydan: precision issues and me sucking at fixed point :/
00:45:18farbrormelonAllright! Thanks Llorean and linuxstb!
00:45:22Shawn_KLlroean: co-op builds required a suppoorted bootloader ;)
00:45:49LloreanI assume you mean COP or coprocessor?
00:45:53Shawn_KI thought that was short for calling it the co-processor
00:46:04preglowsafetydan: the issue was the numerator and denominator in the final divide operation being really equal, but affected by small roundoff errors which would then give huge errors in the result
00:46:06LloreanCoprocessor builds don't need a patched bootloader.
00:46:32LloreanAnd even back when they did, menu still worked on it just fine for me.
00:46:43Shawn_Kmine didn;t run w/ the default rockbox bootloader
00:46:44 Join webguest10 [0] (i=5438997d@gateway/web/cgi-irc/
00:46:46Shawn_Kand I read it wont
00:46:51Shawn_Kwithout a patched one
00:47:08 Quit funky ("leaving")
00:47:15Shawn_KI thought it sounded wierd myself, but when I saw it really didn't work....
00:47:54LloreanThe only versions of the coprocessor patch that have actually had a positive effect have worked with the most recent official bootloader.
00:48:30Shawn_Khmm, I don't know what version was patched into it
00:48:47LloreanAnd you never actually said what you were talking about using "caching" to cache.
00:49:02Shawn_Ki couldnt get my CVS copy patched w/o a few problems so I used psyious builds
00:49:13safetydanpreglow, hrm, doesn't sound like something with an easy fix
00:49:52Shawn_Koh, I did find a minor bug in all the ipod builds
00:50:00linuxstbOnly a minor bug?
00:50:13Shawn_Katm, I havent gotten much time to look at soiurces
00:50:31Shawn_KI've actually found a few major ones revolving around the equalizer
00:50:43preglowsafetydan: who knows, at least not with my fixed point skills
00:51:04Shawn_Kanyways, the -57db and -58db for volume are swapped
00:51:18Shawn_Klike i said, minor bug
00:51:24preglowsafetydan: btw, the url you gave me once which had some fp stuff on it, it's got a cos/sin that's more accurate than the one we have
00:51:37Shawn_Kbrb in 10
00:51:43preglowsafetydan: works the same too, i think
00:51:53safetydanpreglow, yeah I'm actually working on porting that astroinfo program as a plugin
00:51:54LloreanShawn_K: If you've found "major" bugs, why haven't you filed bug reports?
00:52:08safetydanwhich would of course bring in the fixed point stuff
00:52:13preglowsafetydan: might very probably also be faster
00:52:20 Quit bluebrother ("sleep...")
00:54:46Shawn_KLlorean: I've been real busy and I'm still adjusting to helping out
00:55:09safetydanpreglow, it wouldn't surprise me if it was faster. I really had no idea what I was doing when I wrote that :)
00:57:28 Join [-BF42-]Thommy [0] (
00:59:19linuxstbShawn_K: Regarding your volume bug, I assume you have an ipod 5g, in which case if I understand the code correctly, -57dB is the lowest volume, and -58dB is used to represent mute. So what are you experiencing?
00:59:37 Quit hcs ()
00:59:57Shawn_Klinuxstb: doesn't ever mute, and the variances seem like -57 and -58 are swapped
01:00:05Shawn_Kand yes, I have 5.5g
01:01:00 Join combrains [0] (
01:01:02Shawn_Kanyways, I am going to be away from IRC time to time, I am looking at the rockbox sources
01:02:00linuxstbYes, it looks like the muting code is broken...
01:02:40Shawn_Klol, heh ^^
01:02:47 Part webguest21
01:03:13Shawn_Kanyways, why are there so many failsafe checks in the code
01:03:28Shawn_Klike theres a check in the scrollbars to see if the min is the min
01:03:31Soaplinuxstb: For What It's Worth: It appears your mpeg2enc encodes could be optimised (for speed) more by manually lowering the VBV buffer size. Even on your smallest videos I'm seeing a buffer of 224, which is mucho overkill for the job. I don't know how much impact extra VBV buffer has in your decoder, but that's about the only thing I can find which is abnormal.
01:03:47Aloneacan I ask a basic operation question on the gigabeat? I can't seem to figure out how to do insert...
01:04:15 Quit Hyperbit_01 (Read error: 145 (Connection timed out))
01:04:18linuxstbSoap: Don't make me encode them all again.....
01:04:34Soapoh no
01:04:45Soapnot saying that at all.
01:04:49linuxstb:) But I'll try it.
01:04:51***Saving seen data "./dancer.seen"
01:05:04 Join Mouser_X3 [0] (
01:05:08 Nick Mouser_X3 is now known as Mouser_X (
01:05:29SoapAlso, mpeg2enc isn't detecting scene changes and adding I frames at scene changes, this doesn't affect the speed, but does hurt the quality/bitrate ratio.
01:05:34 Join dfkt [0] (
01:05:35Shawn_Kouch...all the GUI still is drawn on a pixel level pretty much
01:06:01Shawn_Kit;s easy and eifficient, but there may be room to improve
01:06:12Shawn_KI'll have to look at the drawing methods
01:06:41goffaimprove how? like less cpu usage to render? or just look nicer?
01:07:05goffacan't argue with that
01:07:21Shawn_Kof course it can be made to look nicer, but when you deal with commands at pixel level...there often is room for improvement
01:07:32Shawn_Kalso, how do the buffers in th players work?
01:07:45Shawn_Kcan you draw to buffer, save it to other memory, etc?
01:07:46linuxstbYou mean the framebuffer?
01:08:11Shawn_Kor do most players have more than one?
01:08:19linuxstbThere's one framebuffer which is as close as possible in format to the LCD hardware. The lcd_update() function transfers the contents to the LCD controller.
01:08:28linuxstbor lcd_update_rect()
01:08:52Shawn_Kupdate selection?
01:08:59linuxstbThat framebuffer is just a normal static buffer.
01:09:01 Join Jsunu [0] (
01:09:10Shawn_K*lcd_update_rect() = update selecdtion?
01:09:28Shawn_Kactually, section is a better word
01:09:46linuxstbYes, it just transfers a rectangular section of the framebuffer to the LCD hardware.
01:09:59linuxstbBut it's probably not used as often as it could be.
01:10:32Shawn_Kyou've got that right, looking at how slow the menus are at times...I can tell it's not fully optimized
01:10:58linuxstbThat's the 5g's LCD driver as much as anything though.
01:11:42Shawn_Kanyways, like I"ve said on the forums...I'm not the greatest coder, but I learn fast
01:11:48linuxstbOn the 5g, lots of smaller updates would actually be slower than big updates.
01:12:04Shawn_Kyeah, that's what I thought
01:12:19Shawn_Kbut someone on the forums told me I should be aiming higher that what I've been
01:12:52linuxstbaiming what higher?
01:13:11Shawn_Kmy optimizations, told me I was looking at too miniscule of optimizations pretty much
01:14:40Shawn_Kanyways, I'm going to come up and say this now...I plan on optimizing most for iPod...if I get other players in process than thats a plus
01:14:56Shawn_Kor....if I get a new mp3 player, I may do those too :P
01:15:58linuxstbThere's already a patch in the patch tracker to improve list scrolling speed - that could be a useful start.
01:16:15linuxstbBut it's quite old and the developer has disappeared...
01:16:19Shawn_Kcould I have the #?
01:16:29Shawn_Kmaybe i can bring it back to life ;)
01:16:44Soapthere are two ipod scolling patches, one of the has been recently synced.
01:17:07Shawn_Koh, not the accelerations...I'm talking strait-up optimizations
01:17:19Shawn_KI tried a few of those scroll demos and don't like them
01:17:31Shawn_KI will probably borrow code from one and start my own :P
01:17:47linuxstbYes, I'm talking about generic list scrolling, not the clickwheel acceleration.
01:18:04linuxstbI don't know the number without searching...
01:18:05Shawn_Kor, if it's not too difficult, I may write one all on my own
01:18:32Shawn_Klinuxstb: give me a good tag to search for it?
01:18:43linuxstbProbably just "scrolling"
01:19:22safetydanI think it's 5591 as puetzk says
01:19:22puetzkthat makes scrolling into a blit and a redraw of exposed lines, instead of a full redraw
01:19:25linuxstbYes, it is...
01:19:39 Quit lamed ("CGI:IRC (Ping timeout)")
01:19:54Shawn_Kalso, I was wondering if anyone thinks a delayed 2nd crossfade would work well to simulate surround sound on headphones?
01:20:12puetzkShawn_K, that effect is called crossfeed, I love it, and it's already implemented :-)
01:20:27Shawn_K*woops, ya, I meant crossfeed
01:20:57Shawn_Klol, I know its implemented, I love it too....I use sterio width and a weak crossfeed
01:21:13Shawn_Kanyways, a slightly delayed 2nd crossfeed could simulate rear speakers
01:21:34 Join barrywardell [0] (n=barrywar@
01:21:46Shawn_KI was asking if it would be too much as far as processing and if the effect would work all that well
01:22:14puetzkyou'd want a different shaped attenuation curve (and it's a much less generic one), but it is possible
01:22:27farbrormelonIs the plugins that are in the builds all plugins availble, or can I find more somewhere, like games and such
01:22:41Shawn_Kok, I'm not up with what the attenuation is
01:22:54Shawn_Kcould anyone offer a quick explanation? ^^
01:22:55puetzkfighters do it to place sounds in 3d to help the pilot's awareness
01:23:35Shawn_Kso rockbox already has the base attenuation coding?
01:23:38puetzkthe crossfeed doesn't blend the signal directly, it quiets the whole thing some, and higher frequencies more
01:23:57puetzkand delays the whole thing a fixed amount
01:23:58Shawn_Kyeah, I know about the cut-off, it wouldnt be needed
01:24:00 Part dfkt
01:24:28puetzkthis is all meant to simulate what would happen to the sound in the loudspeaker setup, as it either conducts through your skull or reflects off walls
01:24:30Shawn_Kanyways, quieting the waves would be wanted to simul;ate rear speakers
01:24:44 Join Mouser_X3 [0] (
01:24:44puetzkboth delay it, and the skull doesn't pass high frequencies much
01:24:58puetzkfor rear you want to simulate the same effects, only front-to-back
01:25:04Shawn_Kyeah, I know a fair amount about sound waves
01:25:14puetzkand the relationship of delays and muffling is quite different than for side to side
01:25:40Shawn_Klol, you're pretty much saying a modified crossfeed just like i said :P
01:26:05puetzkright. But good front-back effects pretty much require calibrating it per-listener
01:26:15puetzkand having the delays vary with frequency too
01:26:18puetzkso the math involved is harder
01:26:21Shawn_Kso you do agree that the idea is there and it can ve implemented effectively? ^^
01:26:38Shawn_Kyes, math is harder, thats why i also metioned CPU usage
01:26:40puetzkyes, it's quite impressive when done well
01:26:55puetzkof course, you really want 4-channel sound to start with too :-)
01:27:03Shawn_Kok, maybe I will try to start a patch for that and see how far I get
01:27:06puetzkor else you're carefully presenting something that you just made up
01:27:26Shawn_Kactually, don't need 4 channel sound if you are just simulating stereo to surround
01:27:35phrontistif I'm in shuffle mode
01:27:46phrontistoh, nevermind
01:27:52Shawn_KI'm speaking of the effect stereo sound has on a surround system in a room
01:27:56Shawn_Kanyways, brb in 10
01:27:58LloreanI'm not sure why you'd want to take 2 channels and simulate 4.
01:27:59puetzksure, but there's not much point to that. Just add reverb if you want made-up surround :-)
01:28:25puetzkLlorean, the only time I've used it with headphones is that my santa cruz (pc soundcard) has sensaura, and can do it in realtime
01:28:26 Quit Mouser_X (Nick collision from services.)
01:28:29 Nick Mouser_X3 is now known as Mouser_X (
01:28:35puetzkand it can be fun to use in games, which place sounds in 3d
01:28:44puetzkso there really is surround data
01:29:07Soapare those games using the older prologic encoding for rear?
01:29:14Lloreanpuetzk: Yeah, but see, that's not taking 2 channels and simulating 4. That's "taking surround sound, and simulating it in two channels"
01:29:22 Join fsadfs [0] (
01:29:26 Quit Jsunu (Read error: 131 (Connection reset by peer))
01:29:28puetzkright. I see little point in that :-)
01:30:25puetzkSoap, they're just playing samples (usually mono even) and telling the driver where each sample is so it can produce a surround mix
01:30:50fsadfsis there any way that rockbox can read the music i put on with itunes
01:31:08puetzkfsadfs, if you turn on the ID3 database it will find it
01:31:19puetzkit's all there, just iTunes makes the file and foldernames pretty useless
01:31:54fsadfshow do i turn on the id3 database
01:32:00puetzkor see to seed it from the itunesdb
01:32:36 Part Llorean
01:32:47puetzksettings>general settings>file view>Database to set it up, then Settings>general settings>file view>show files to pick it
01:33:05puetzkafter you hit initialize database it will take quite a while to be ready
01:33:11 Quit [-BF42-]Thommy (Read error: 110 (Connection timed out))
01:33:13puetzk(depending on how much music you have)
01:38:39Aloneawhat is car adapter mode? it wasn't in the manual.
01:41:21 Quit webguest10 ("CGI:IRC (EOF)")
01:44:36 Join debauched_sloth [0] (
01:47:11 Join zuzu_ [0] (
01:47:55zuzu_how are rockbox devices commonly interoperated with a host OS such as Linux? ala the itunes-ipod workflow.
01:48:36zuzu_specifically I'm wondering if something similar can be done with rhythmbox and rockbox... or is it a "just copy over the MP3s" only thing?
01:49:21 Quit miepchen^schlaf ("Verlassend")
01:49:23safetydanzuzu_, just copy the MP3s generally.
01:49:43safetydanThe database can be built by the device so other software isn't generally needed
01:50:05crwlamarok probably supports rockboxable players too, though i haven't tried the function
01:50:20crwli'm happy using the command line for copying :P
01:50:26 Join senab [0] (
01:50:59 Join webguest95 [0] (i=4742e174@gateway/web/cgi-irc/
01:51:27webguest95can someone help me?
01:51:46senabif you tell someone the problem :P
01:52:17webguest95oh well i have a ipod 5.5 and i installed rockbox but the the battery died
01:52:19webguest95and i can't charge it
01:52:22farbrormelonIs there anyone that can help me with rockdoom?
01:52:43senabhow do you mean the battery died?
01:52:43Aloneawhats your player farbrormelon?
01:52:50webguest95i pluged it into the usb but it...
01:52:57farbrormeloniPod Nano
01:53:00webguest95i was playing doom and it ran out
01:53:05 Part fsadfs
01:53:22 Join webguest64 [0] (i=5438997d@gateway/web/cgi-irc/
01:53:26senabso it won't charge?
01:53:48webguest95I plugged it into the usb but it won't charge or turn on
01:53:51farbrormelonNo... i have put the files in games/doom/ folder
01:54:04Aloneafarbrormelon: so are you just having trouble installing it?
01:54:06senabtry holding menu + select for about 5 seconds
01:54:20Aloneais rockdoom or doom under the plugins?
01:54:33farbrormelonI get this message "No base wad"
01:54:43senabnp :)
01:54:51 Join fsadfs [0] (
01:55:20fsadfshow can i stop my ipod from going out of disk mode
01:55:28 Quit webguest95 (Client Quit)
01:55:39senabyou can't if rockbox booted into it
01:56:13 Quit Thundercloud_ (Remote closed the connection)
01:56:20senabif you've gone into disk mode manually, the disconnect screen will stay on until you press select
01:57:41Aloneaand you have the other doom wads correct? either from the internet somewhere or from an actual game cd?
01:58:15 Quit senab (Remote closed the connection)
01:58:40farbrormelonI got it from a site... sorry to bother =/ I see now that the file was damaged
01:59:43linuxstbfsadfs: Have you enabled the "Use ipod as a disk" option in itunes?
01:59:49Aloneaokies. I happened to have the doom collectors edition disk. do you need help getting a file?
02:00:05 Join Digamma [0] (
02:00:43linuxstbAlonea: I assume you're not going to discuss copyright violations here?
02:01:03Aloneaoh, Iwasn't going to post any sites on here at all.
02:01:31Aloneaif he needs help I was going to do the pm thing. and yes I know it is copyright violations
02:03:17 Quit fsadfs ()
02:03:30 Join miepchen^schlaf [0] (
02:04:17farbrormelon<Alonea: That would be nice =)
02:06:13Aloneayeah, lemme figure out how to register real quick so I can pm
02:06:16linuxstbAlonea: You just did discuss copyright violations and it will forever be in the logs...
02:07:13webguest64linuxstb: Where do you wanna put the fill up procedure for mpegplayer?? Just before the "if" with te audio buffer overflow?
02:07:22webguest64nick Angry
02:07:24 Quit BHSPitMonkey (Remote closed the connection)
02:07:44webguest64i meant the fill up for the audiobuffer
02:07:51linuxstbI haven't thought about it yet.
02:08:07webguest64ok... sry^^
02:08:33linuxstbBut before we can rebuffer, we need to mark the processed data as being read. Plus also handle the buffer wraparound point...
02:08:53webguest64that sounds logic
02:09:08linuxstbBut if you want to tackle it, please go ahead...
02:09:54webguest64i think you're better in that... hust learning now from the code i see and its very interesting to follow your progress
02:10:04 Part pixelma
02:10:25Alonealinuxstb: ok, what do you mean by discussing copyrigh violations? As in, dont discuss it at all or what? If there is a rule somewhere about discussing it entirely, sorry.
02:10:37 Join jeffb [0] (
02:11:23webguest64i think he meant that you could be talking about sharing copyrighted software in this forum
02:11:44linuxstbReplace "could" with "shouldn't"
02:11:53webguest64Thats what i meant
02:12:28 Join fsadfs [0] (
02:12:48 Join TrueJournals [0] (
02:13:06 Join BHSPitMonkey [0] (
02:13:07 Quit webguest64 ("CGI:IRC")
02:13:17fsadfsok i need help im doing id3 converting i have done everything that is instructed where will my music be
02:13:17TrueJournalsHey, I have a small coding question...
02:13:25 Join Angry [0] (
02:13:29TrueJournalsis it possible to set attributes on files (hide/unhide)?
02:13:59Aloneaah, well, sorry. trying to keep within rules here...i finally got the pm thing to work.
02:15:47AngryTruejournals: No its not possible
02:16:14AngryI assume that you mean the problems with the OF
02:16:20BHSPitMonkeyTrueJournals, you can change what you can view in the file browser, though
02:16:49AngrySo i would recommend renaming the folders also if you wanna switch. Thats what id do to avoid the hiding.
02:17:16 Quit Alonea (Remote closed the connection)
02:17:16TrueJournalsBHSPitMonkey: Angry is right. People on misticriver don't like that the original firmware sets the music directory to hidden. I was wondering if there was a way to make it automatically un-hide the music folder when the player starts up.
02:17:27farbrormelon"Private messages are dissabled"
02:17:33fsadfsok i need help im doing id3 converting i have done everything that is instructed where will my itunesmusic be
02:17:46TrueJournalsAngry: I actually don't have this problem myself :-p I use the ID3 database... works great :-\
02:17:47farbrormelonPrivate messages from unregistered users are currently blocked due to spam problems,
02:18:20 Join Alonea [0] (
02:18:42farbrormelonHow do I install a patch in rockbox?
02:19:13Angryyou dont install it... you need to apply a patch to the source and afterwards compile your build
02:19:16BHSPitMonkeyTrueJournals, how is it a problem? Decent OS's disregard that hiding practice, and the rest has settings for it
02:19:33 Join gotthardt [0] (i=86868804@gateway/web/cgi-irc/
02:19:41BHSPitMonkeyfarbrormelon, you get the source code to rockbox, apply the patch to it, and recompile it yourself, using the cross-compiler for your player
02:19:48BHSPitMonkeyfarbrormelon, it's documented in the wiki.
02:19:56AngryLook here:
02:20:05farbrormeloni looked in the PDF
02:20:08gotthardtAny devs here that can grant commit rights?
02:20:32BHSPitMonkeynot many devs that will hand out commit rights liberally
02:20:39 Part fsadfs
02:20:43TrueJournalsBHSPitMonkey: People like to have their settings set to "supported" so that they don't see hidden directories (and all their clutter) when they use rockbox. However, with the H10 original firmware (guess I should have specified), it sets the Music folder to hidden. So people either have to connect the player to a computer to fix it, deal with the All setting, or use the ID3 database.
02:20:49Ajaxincits becasue markun left and we dont have anyone to commit thigns for the gigabeat
02:20:51Shawn_Kwhere can I find lcd.c in the sources?
02:21:12Ajaxincwe are leaderless currently in the gigabeat department
02:21:18Shawn_Keh, no fast response, guess I'll do search
02:21:46farbrormelonAlonea: I am blocked to pm you
02:21:49TrueJournalsShawn_K: IIRC, every player has it's own lcd.c... which one are you looking for?
02:22:01 Join Llorean [0] (n=Llorean@rockbox/administrator/Llorean)
02:22:41Shawn_Ki only came up with one lcd.c in the search
02:22:59Shawn_Kthis is before the compilable sources have been made
02:23:05farbrormelonAlonea: still blocked...
02:23:23Alonea /msg nickserv set unfiltered on
02:23:27 Join fsadfs [0] (
02:23:27TrueJournalsShawn_K: You can find all the lcd.c's you need: (or the /firmware/drivers folder if you downloaded the source)
02:23:50Shawn_Kyeah, I downloaded the full CVS source
02:24:01Aloneahave you registered? go here
02:24:06Aloneaand follow instructions
02:24:11Shawn_Kbut i found it in apps/plugins/rockboy
02:24:24TrueJournalsah, a plugin's file :-p
02:24:33TrueJournalssource is so confusing sometimes
02:24:41Shawn_Kah, found it
02:24:44fsadfsi converted the intunesdb is there any way of seeing whole albums instead of single songs
02:25:30Shawn_Kare the lcd-16bit.c and ipod lcd source both used in an iPod compile?
02:26:02 Nick Angry is now known as Angryman (
02:26:19TrueJournalsThat I don't know...
02:26:44 Nick Angryman is now known as Angry (
02:27:01Shawn_Klol, ok....I just started reading, it seems the 16bit is just for 16bit formats and basic rendering
02:27:08Shawn_Kboth are most likely used
02:27:09 Part Angry
02:27:19Arathisbarrywardell: is the BL for H10 at dualboot now? at least it's been updated lately
02:27:41 Join Angry [0] (
02:28:38linuxstbShawn_K: Rockbox LCD drivers are split into the high-level code that manipulates the framebuffer (e.g. lcd-16bit.c) and the low-level code that transfers that framebuffer to the LCD (amongst other things) - lcd-ipodvideo.c for your ipod.
02:29:36Shawn_Kok, I see that now...
02:29:48Shawn_Kanyways, this is...well, sad
02:30:00linuxstbThe GraphicsAPI wiki page may help understand the high-level code.
02:30:03Shawn_Kthe whole system is designed for a 2bit color
02:30:07Shawn_Kas for drawing
02:30:15Shawn_Ki mean 1bit
02:30:18fsadfsi converted the intunesdb is there any way of seeing whole albums instead of single songs
02:30:24Shawn_K1bit = 2 colors
02:30:37barrywardellArathis: yes, it was update after the dual boot changes
02:30:51Arathisokay, thanks
02:31:27Arathisbtw: did I see right that mpeg with sound is basically supported now?
02:31:38 Part Angry
02:31:52linuxstbArathis: Yes - I've just updated the PluginMpegplayer wiki page with information.
02:32:04Arathiscool :D
02:33:17*myzar|away licks scorche
02:33:18Shawn_Kwould anyone be interested in recoding the lcd drivers for 16bit players (mainly iPods) with me?
02:33:35 Nick myzar|away is now known as myzar (n=myzar@
02:33:54TrueJournalsShawn_K: umm... why? (And, no, not me... I'm not a coder)
02:34:23linuxstbYou have ideas to speed them up?
02:34:30scorchemyzar: tricksy...
02:34:35Shawn_KTrueJournals: the whole system is made for 2bits, containing little room for expansion and optimization
02:34:51 Join D_C [0] (
02:35:14Shawn_KI like "pretty" and fast GUIs ^^
02:35:35LloreanShawn_K: Remember that the whole system has to maintain compatibility with 1-bit displays.
02:35:56Shawn_KIt would be a dream if I could fit some PNG support w/caching in there
02:36:10 Quit muesli__ ("ich will Kühe!!!")
02:36:27Shawn_Kit would be possible to fit it in therem but it would require two drawing methods
02:36:39 Join Audioengineer [0] (
02:36:45Shawn_Kone for the old 1bit support and one for the new 16bit methods
02:37:15Shawn_Kif done right, shouldn't make a difference in speed...just a larger rockbox file
02:37:19D_Cdoes the latest rockbox for iriver h3xx support video play back? i swear i saw a video play back page except i kind of lost it
02:38:02 Quit TrueJournals ("See ya")
02:38:33linuxstbWhat do you want to do that the current API doesn't let you?
02:39:01 Quit fsadfs ()
02:39:04myzarlinuxstb :)
02:39:07 Quit combrains (Read error: 145 (Connection timed out))
02:39:07myzarhi =]
02:39:13linuxstbGood morning.
02:39:16LloreanShawn_K: A larger rockbox binary means less buffer space. You can ifdef two methods if you feel it needs new code for the color screens.
02:40:02 Quit Shawn_K ()
02:41:41 Join Shawn_K [0] (
02:41:52Shawn_Know registered my nick ^^
02:42:01Shawn_KI AM....SHAWN_K
02:42:20Arathisbarrywardell: is the h10 mic mono or stereo?
02:42:34Shawn_Kprobably mono
02:42:41Shawn_Kmay record the mono > stereo
02:42:42linuxstbShawn_K: What do you want to do that the current API doesn't let you?
02:43:04barrywardellArathis: mono, i think
02:43:32Shawn_KI want to br able to have an optimized graphical GUI
02:44:07Shawn_Kusing current API would make a mess
02:44:42 Quit netmasta10bt (Read error: 60 (Operation timed out))
02:45:43linuxstbCan you be more specific?
02:45:44 Quit Digamma ("Leaving")
02:46:16Shawn_KIt would be nice to be able to customize menus with graphics, etc
02:46:22Shawn_Kthere's already some support
02:46:32Shawn_Kbut not to what I would like to see
02:47:52safetydanShawn_K, remember that you need to support multiple targets, some of those don't even have a bitmap display
02:48:58linuxstbAnd some have two displays (a remote LCD)
02:50:13linuxstbAnd the Voice UI, and multiple languages...
02:51:17 Quit farbrormelon ()
02:51:33BHSPitMonkeytheoretically, you could code an entire new interface that makes sense for your wishes and your player, and have it run on top of rb.
02:51:52BHSPitMonkeyand/or patch the heck out of it
02:52:26Shawn_Ksafetydan: even bitmaps can be converted to 1bit ;)
02:52:37scorcheand then it would be a lonely patch on the tracker that would never be committed
02:52:45safetydanShawn_K, yes, but they can't be converted to a charcell display
02:52:47Shawn_Kalso, like I said...if the support for the extra features is optional
02:52:49safetydanwell, not easily
02:53:08Shawn_Kit would be more of a skin feature :)
02:53:26safetydanadding skinning is hardly optimising though
02:54:15 Quit ctaylorr (Read error: 60 (Operation timed out))
02:54:58D_Clinuxboy how do you add the mpegplayer in? or is it already in with the latest rockbox?
02:56:29safetydanD_C, it's already part of Rockbox
02:56:40D_Cah ok
02:57:11 Join Ron_ [0] (
03:00:06 Join Febs [0] (
03:00:36 Quit gotthardt ("CGI:IRC (Ping timeout)")
03:00:44Ron_msg / NickServ HELP IDENITY
03:02:37 Nick Ron_ is now known as rkostynu (
03:03:14D_Calso what do i do with this coldfire patch to speed up file playback ?
03:04:39 Join ctaylorr [0] (
03:04:44scorcheD_C: see the wiki page SimpleGuideToCompiling
03:04:53***Saving seen data "./dancer.seen"
03:04:56D_Cthanks :)
03:06:44 Join jdong [0] (n=jdong@ubuntu/member/jdong)
03:07:56jdongit seems like on the iPod (video 30GB 5.5G) disk cache can be "defeated"
03:08:07jdongi.e. load up a playlist and play in shuffle mode
03:08:15jdongit'll cache up pretty nicely and spin down the disk
03:08:26jdongwhen cache is empty, it'll refill it and then shut off the HD
03:08:44jdonghowever, if I quickly skip 5+ tracks....
03:08:55jdongthe Disk Active notification will blink on and off
03:09:08jdongand next song won't display
03:09:19jdongwhich seems like the disk isn't caching but just reading in realtime...
03:09:26jdongif I pause for a few seconds then play again, it returns to normal
03:09:42jdongI'd expect such behavior to have a significant impact on battery life?
03:12:09jdongit seems like after two or so songs (5mins?) it's able to recover
03:16:06 Join Gon [0] (
03:24:22D_Cscorche i am suppose to use the source archive right?
03:24:58D_Cthe latest source archive link is dead ;(
03:25:26D_Cil just use yesterdays
03:30:39 Quit Alonea ("Chatzilla 0.9.77 [Firefox]")
03:35:44 Quit rkostynu ("Chatzilla 0.9.77 [Firefox]")
03:40:22 Join dewdude [0] (
03:40:36 Quit Mouser_X (Read error: 110 (Connection timed out))
03:41:59 Join Nico_P [0] (
03:43:20 Join jackfusion [0] (i=cfa11562@gateway/web/cgi-irc/
03:43:39jackfusionhi all
03:46:00jackfusionhas any one had problem with bootpartition.bin not opening
03:48:14jackfusionI get this error cannot open loader image file bootpartition.bin
03:49:29 Join Techni [0] (
03:49:36jackfusioncannot open loader image file bootpartition-video.bin
03:50:03Technicould someone send me the gigabeat f bootloader? I have no method/idea of building it myself
03:50:30Lloreanjackfusion: There is no file "bootpartition-video.bin" as part of the install process.
03:50:55jackfusionthat is part of the bootloader install
03:51:05LloreanYes, but the file for that is bootloader-video.bin
03:51:15LloreanSo if you're typing bootpartition-video.bin, clearly it won't be there.
03:51:24LloreanAlso, try the new install instructions, as it's a better method.
03:51:44jackfusionbut what about bootpartition.bin
03:52:05Lloreanbootpartition.bin is bootpartition.bin, not bootpartition-video.bin
03:52:39D_CWARNING: The compiler you must use (m68k-elf-gcc) is not in your path! is that bad :S ive ticked it during the install so its definitely there
03:53:11jackfusionyes I fix that but I still get the error I get this error cannot open loader image file bootpartition.bin
03:53:58LloreanIs the file bootpartiton.bin in the folder you're in?
03:54:33LloreanIt shows up when you type "dir"?
03:55:04 Join webguest84 [0] (i=57afc73e@gateway/web/cgi-irc/
03:55:28LloreanEither try the new install method, or paste the results of a "dir" and the error message to pastebin.
03:55:34LloreanDO NOT post them in this channel, by the way.
03:55:58jackfusionok thank u
03:56:56webguest84hello, i have a question. i patched my ipod nano with the kernel_on_cop_6.diff. but the video is still unsync with the audio. (sry for my bad english)
03:57:10webguest84what can i do ?
03:57:16Soapwebguest84: there is no syncing code in the mpegplayer yet
03:57:30Lloreanwebguest84: Try enabling frameskip and see if that helps
03:57:38Soapall that kernel_on_cpo patch does is allow the splitting of tasks between the two cores
03:57:49 Join Matu [0] (
03:58:08 Part Matu
03:59:41webguest84oh.. i thought the new kernel will start the audio parallel to the video
04:00:01 Quit barrywardell ()
04:00:14Lloreanwebguest84: Only if you have a video that can playback faster than its frame rate, and you have frame limiting on.
04:02:16webguest84that means ? do i have to encode the video in another way ?
04:02:50webguest84sry but i have no experiences with those things
04:03:04webguest84but i want to look videos on my ipod :)
04:04:10LloreanJust wait for the plugin to be finished then.
04:04:25LloreanAt the moment it can't play videos longer than the buffer length anyway
04:05:36jackfusionHow do u restart ipod g5?
04:05:38Shawn_Khey guys...I think I may have found some low-level LCD driver optimizations, but moving the files to my iPod and starting it up over and over is a pain
04:05:48Shawn_Ki saw there were emulators or something in CVS
04:05:56Shawn_Khow can I get win32 versions of those?
04:06:08scorchesimulator != emulator
04:06:20Shawn_Kok, bad then
04:06:22LloreanShawn_K: Firstly, they're simulators. They're never called emulators in any official documentation because they don't emulate any hardware, they just simulate the user interface.
04:06:33Shawn_Kok, thank you
04:06:41Shawn_Kanyone know of any emulators?
04:07:09LloreanI don't believe there are any iPod Emulators.
04:07:44Technidoes anyone have the gigabeat bootloader? its not included with rockbox for some odd reason
04:07:49LloreanThere's really very little cause for running iPod code on a PC outside of us and iPL, and putting together a functional emulator for fairly undocumented hardware would probably be more of a hassle than it's worth.
04:07:51jackfusiongoogle search ipod emulators
04:08:01SoapThough a P4 emulating an iPod emulating a Sinclair would be cool.
04:08:40dewdudex86 emulating PPC emulating Atari
04:08:42LloreanTechni: Instructions for compiling your own can be found here:
04:08:44dewdudei was bored one day.
04:09:15TechniLlorean: The instructions are for linux though
04:09:50LloreanTechni: Or Cygwin or the VMWare toolkit that we provide.
04:10:35Techniso you dont have a precompiled bootloader? that sounds kind of odd, why provide precompiled anything if you're not including everything?
04:10:50LloreanBecause it hasn't been made available yet
04:10:59LloreanDon't assume that it'll never be up just because it isn't now.
04:11:18LloreanBut you want to install it before it's been made officially available, and I'm telling you the current process.
04:11:33jackfusionsorry about this How do u restart ipod g5?
04:11:47Lloreanjackfusion: Turn it off, and then turn it on...
04:12:03Lloreanjackfusion: Have you read the manual for Rockbox, specifically the basic controls portions?
04:13:07 Join midkay_ [0] (
04:13:13 Quit Arathis ("[rl_bot quit]")
04:13:44jdongjackfusion: that's covered on every iPod related site and even on the front page of apple's ipod support :)
04:14:05jdongtoggle HOLD on then off, then hold MENU and CENTER for 10 seconds
04:14:12Lloreanjdong: That's a hard reboot, and not the suggested method unless it's actually frozen.
04:14:21LloreanOr unless you're in the Apple Firmware, which doesn't have a power off.
04:14:32jdongLlorean: I thought that's what he was asking for
04:14:40jdongI know it's not exactly healthy to rockbox to do that
04:16:39LloreanIt's not exactly "unhealthy" just... well, you don't get the proper shutdown procedure so you can end up with reset settings or other things
04:17:37 Join duende_inside [0] (
04:17:50jdongwell half-written playlists and changed settings doesn't sound very healthy either :D
04:18:15LloreanYeah, but both of those are just really minor irritations overall.
04:18:17LloreanAt least, to me
04:18:25dewdudemy settings reset every time i update rockbox...then again, i'm probably doing it wrong. i'm lazy, i just extract the zip directly to my unit
04:18:48jdongdewdude: that's how I do it, and my settings don't seem to reset from it
04:19:08dewdudei did it yesterday...and lost all my settings
04:19:08jdongI just have a shellscript that wgets the latest CVS build and unzips it with force overwrite
04:19:19jackfusionthis is cool I finally got something working on my ipod
04:19:29jdongI'd imagine some kind of updates would cause that to happen, dewdude
04:19:44dewdudemaybe..i don't know
04:19:44jdongthe wiki does recommend to save a .cfg file for those cases
04:19:50dewdudeyeah, i should do that
04:19:57dewdudei just...i can't figure out the text entry
04:20:09jdongbtw, major kudos to everyone involved with rockbox
04:20:12dewdudeat all..i read the manual, and i still can't get it to work. it's not like i made a LOT of custom settings
04:20:19dewdudejust EQ and display color
04:20:25jdongit's made my newly acquired iPod a lot more enjoyable
04:21:44Lloreandewdude: All you have to do is choose "Write config file" and then press the confirm button to save it.
04:22:59 Part debauched_sloth
04:23:29 Quit midkay (Read error: 110 (Connection timed out))
04:23:50dewdudeyeah...that's the one thing that's constantly confuzzled me. you have to scroll to the file..then click play to save it. like i said, i read the manual, i rarely use it, i never change EQ settings unless i do something like change headphones
04:24:36midkay_jackfusion: so now it works and your request can be closed? (solution seems to be that you read the instructions wrong.)
04:24:44midkay_er, bug report.
04:24:53jdongthe text editing widget kind of takes a bit to get used to
04:25:08jdongand sadly due to iPod while-playing UI lag, it isn't all that easy to use
04:25:27jdongis there any work towards alleviating some of the UI lagginess while playing back music?
04:25:41dewdudejdong: sure. don't use mp3
04:25:52jdongis only MP3 affected?
04:25:54 Join barrywardell [0] (n=barrywar@
04:25:57dewdudeogg too
04:26:02dewdudethey haven't optimized the decoders
04:26:06jdongthen what format isn't affected??
04:26:15dewdudei use FLAC and musepack
04:26:40dewdudei think the frequency scaling patch will help, but i've not looked
04:26:45Techni um, i ran configure, and its not giving me the option of making a build file for the gigabeat
04:26:47Lloreanjdong: The choice right now is basically "UI Lag" or "Skipping during playback"
04:26:47dewdudei use musepack exclusively
04:26:49goffahey dewdude save your .cfg file
04:26:52goffawhen you update
04:27:05dewdudehey, goffa, just did.
04:27:14goffai like to put it out in root
04:27:22jdongLlorean: yeah I kinda imagined
04:27:24goffathen load it afterwards
04:27:35Lloreanjdong: The dual core patch might offer some improvement in that area, though not immediately
04:27:56jdongdo we currently only use one core?
04:27:57LloreanTechni: What is it giving you?
04:28:04dewdudejdong: yeah.
04:28:08Technioptions 1 - 12, mostly archos systems
04:28:09Shawn_Kyeah, like I said...I am working on optimizations ^^
04:28:11Lloreanjdong: Well until a couple days ago, we couldn't get the second one to work properly.
04:28:20jdongI see
04:28:23Techniarchos, iriver, iaudio
04:28:28jdonganyway, I'm splended to hear that you guys are making progress
04:28:33jdongkeep up the great work
04:28:51LloreanTechni: No Sandisk even? Is your source up to date?
04:28:53dewdudejdong: just be patient. they'll fix stuff.
04:29:02Shawn_KI'm new, I just started today w/ the code....none of them know if I'm doing anything usefull YET, lol
04:29:10jdongdewdude: I have full confidence in that :)
04:29:15Technii just downloaded the source from a few minutes ago
04:29:26Technithe one that links to
04:29:32 Quit webguest84 ("CGI:IRC")
04:29:34jdongdewdude: I'm far from a whiny brat; I do quite a bit of open source work myself :)
04:29:52LloreanTechni: Why did you download 2.5?
04:30:01LloreanThat's a very very very ancient source archive
04:30:04Technicause thats the one the site links to
04:30:05jdongdewdude: I am an admin over @ and I also maintain the Ubuntu Backports repositories
04:30:06LloreanThat's the last release version
04:30:26Shawn_Kjdong:, I like kde :P
04:30:29LloreanTechni: That page says it's dated 2005-09-22 <−− Notice the YEAR on that?
04:30:32dewdudejdong, ubuntu is my fav linux distro...mostly because it was the one i was using when everything clicked
04:30:37dewdudeShawn_K, kubuntu FTW!
04:30:40jdongShawn_K: Kubuntu, lol
04:30:48LloreanTechni: Use the source archive from the CVS builds page, or just use CVS to check out a copy
04:30:51Shawn_Kjdong: too lazy to set up kde, so ya....I use that one
04:30:59jdongShawn_K: ubuntu/kubuntu are one :)
04:31:01Shawn_Kjdong: that's probably my fav linux distro
04:31:01Techni page cannot be found
04:31:04scorcheenough of the ubuntu talk...
04:31:08Technicvs is not working, asks for a password
04:31:18LloreanTechni: Remember that link I gave you that tells you how to make the bootloader? It even specifically tells you how to use CVS to check out sources
04:31:34Techniand that links instructions did not work, thats why i downloaded the source directly
04:31:34Shawn_Kjdong: the core is pretty much the same, but one for gnome, one for kde...apps and some optimizations are different
04:31:47scorcheTechni: they work just fine
04:31:48LloreanTechni: When it asks for a password, just hit enter
04:31:52Technii logged into linux on my ps3, and copy/pasted the first line in, and it asked for a password
04:32:02Technii hit enter it goes back to the prompt
04:32:05LloreanThen it worked
04:32:14Shawn_Koooh, linux on does that run?
04:32:14Shawn_Ki been dying to know
04:32:25LloreanShawn_K: Keep off-topic messages to PMs when the channel is actually active on relevant things
04:32:26scorcheand the general linux talk...
04:32:27Techniyellow dog is pretty, but fedora core seems set up better
04:32:40LloreanTechni: Keep off-topic to PMs when the channel is active.
04:33:27LloreanTechni: Once you've hit enter after the password prompt, then you run the second line
04:33:30Technisorry, i was typing that before you said it
04:33:34Shawn_Kyellow dog is made for mac, did they update it for intel/amd since intelmac?
04:33:54Techniah, i expected some sort of feedback
04:34:06scorcheShawn_K: we told you before...
04:34:08LloreanShawn_K: Keep OFF TOPIC to PMs
04:34:55 Join rkostynu [0] (
04:35:18Shawn_Kscorche: what did you tell me?
04:35:26Shawn_Kand Llorean: ok, sorry
04:35:50scorcheShawn_K: see the one who ended my sentence =)
04:35:51 Join hcs [0] (n=hcs@rockbox/contributor/hcs)
04:36:20Shawn_Kok, i see
04:36:25Shawn_Ksorry guys
04:36:30Shawn_KI get carried away easily :P
04:42:52Shawn_Kwell, I made a few minor changes to the lcd drivers for ipods
04:43:01Shawn_Kgonna see if they work as expected ^^
04:43:07Technioh come on *cries* it wants me to install/configure gcc-elf
04:43:14Shawn_Koh, that
04:43:19Technithis is why i hate linux
04:43:35Shawn_Kah, running on your own linux box?
04:43:40Technion the ps3
04:45:06Technino rule to make target zip stop
04:47:09LloreanTechni: Well, we told you to use Cygwin or VMware.
04:47:25LloreanTechni: If you must do it on your PS3, here:
04:47:58Technii assumed since the instructions were given for linux it would work on a linux box, my bad
04:48:25LloreanIt will work on a linux box.
04:48:29LloreanIf you install the proper software.
04:48:41LloreanThat being the toolchain.
04:48:51goffaluckily a nice setup script is in the source download
04:48:59LloreanAh, there is that.
04:49:07 Join justlevine [0] (i=4355f22a@gateway/web/cgi-irc/
04:49:08 Quit rkostynu ("Chatzilla 0.9.77 [Firefox]")
04:49:11LloreanWhat was it called again? 's in the tools folder, but I can never remember the name
04:49:32LloreanSee, I always want to call it "" for some reason
04:49:36Shawn_Kwow, I started with a fresh source and suddenly it wont compile????
04:49:37Shawn_Kaction.c:24:18: error: lang.h: No such file or directory
04:49:41LloreanTechni: can do most of the work for you.
04:49:49LloreanShawn_K: Did you check out with CVS?
04:50:06 Quit justlevine (Client Quit)
04:50:26Shawn_Kit compiled just fine for me last night
04:50:44Shawn_KI mean I had trouble with patches, but I just did a remove of ones that didnt go through
04:50:52Shawn_Kand even thewn it work
04:50:58Shawn_Know, no patches and it doesnt work
04:51:21LloreanAnd a cvs update shows all the files are fine, no Ms?
04:51:31Technihow do i run the .sh? I type its filename while in the tools directory and it says command not found, even though it is there when i type dir
04:51:51Llorean ./
04:51:57Shawn_Kit was a copy of a CVS checkout i did earlier
04:52:04Shawn_Kand it worked before
04:52:06LloreanShawn_K: That does not answer my question.
04:52:41 Part jackfusion
04:53:45dewdudehas anyone ever tried to play content higher than 44.1 or 48 khz/16-bit on the ipod?
04:53:56Shawn_KLlroean: yes
04:54:05Shawn_Kit did check fine
04:55:32LloreanShawn_K: So, a cvs update doesn't update any files, and it doesn't say any are merged, and yet a new, fresh build directory doesn't work after a configure and a make?
04:55:39Shawn_Ki am trying a clean build directory, before I had just used "make clean", now I am trying an entirely new directory
04:57:20Shawn_Ki dont like how it keeps telling me it's using a ccache that doesnt seem to exist
04:57:36Shawn_Kbut if it works this time, then I'll just ignore it
04:59:49Shawn_Kyeah, that was it, since it wasnt was ignoring some makes
04:59:50 Join Mouser_X3 [0] (
05:00:00Shawn_Kit just generated a new lang.h
05:00:08 Nick Mouser_X3 is now known as Mouser_X (
05:02:02Shawn_Kmy first goal will be to get my iPod to last over 5hrs w/ playing MP3s, lol
05:02:22hcsI got mine to 6:30, once
05:02:25myzarmp3s are overrated
05:02:27Shawn_Ki have a 5.5g
05:02:34myzarbe cool like linuxstb and scorche
05:02:39myzarand get your 5.5g to play FLAC!
05:02:40Shawn_Kwell, I don't exactly feel like reformatting all my mjusic
05:02:59Shawn_KI'm a music nut, the re-code and audio lost would drive me nuts
05:03:14Shawn_KI would notice every loss from compression >.<
05:03:52Shawn_KI'm already ready to kick myself in the ass for encoding 5 albums or so into WMAs
05:04:02scorchemyzar: i never had a 5.5g...
05:04:25hcsShawn_K: why use mp3 if you notice such losses?
05:04:37Shawn_KLAME + high bitrate = good
05:04:39scorcheand there is no "audio lost" in FLAC
05:04:49Shawn_Khow large is FLAC?
05:04:57***Saving seen data "./dancer.seen"
05:04:59myzarshhhhh scorche
05:05:04zuzu_.5 from WAV/AIFF
05:05:10myzarShawn_K, FLAC is a wav file with no loss
05:05:11jdonga lot larger than a typical mp3 :)
05:05:15myzaryou can remake it into wavs
05:05:23scorchemyzar: wavs dont have loss either...
05:05:30myzarnobody likes wavs scorche
05:05:32myzarshhh :P
05:05:34zuzu_"lossless" is the word
05:05:49myzarnobody likes APE or wavpack
05:05:51Shawn_Kso a 5mb MP3 would be like a 10mb FLAC?
05:05:51myzarso FLAC is a more proper term
05:05:59myzarbecause nobody uses any other lossless formats >:(
05:06:00zuzu_will ipods with rockbox do line-in recording?
05:06:03Shawn_Kor is the compression worse?
05:06:15jdongShawn_K: MP3 is like 10% original filesize, FLAC is like 50% orig filesize
05:06:20zuzu_yeah, FLAC or ALAC (the latter only excused because of iTunes use)
05:06:33myzarALAC is worthless
05:06:35myzarstick to FLAC
05:06:49hcsI've got too much in APE, gotta see about converting it all to FLAC some day
05:06:51Shawn_KFLAC = 5x MP3!?
05:07:00scorcheShawn_K: around 20-60 per song
05:07:03Shawn_Khell no, I'll stick to a high birate MP3, lol
05:07:06zuzu_MP3 is lossy, FLAC is lossless
05:07:14*scorche shrugs
05:07:17SoapShawn_K: FLAC = +-700kbps
05:07:35Shawn_Klol, ya, I get that much
05:07:54Soapso if you are one of the silly boys who encodes 320 MP3, it isn't nearly 5x
05:07:54myzarKBPs in lossless doesn't matter
05:07:55myzarthere's no loss
05:08:09myzarthe bitrate in lossless music tells you an estimation on how big the file is
05:08:20myzaronly the actual recording would make it any better sounding
05:08:22LloreanShawn_K: FLAC is archival, you can reproduce the original CD from them.
05:08:22scorchewhich is why he said that...
05:08:22myzarif it's 16 bit or not
05:08:33myzarbut it's not like that in lossy music
05:08:33SoapA - it is "b" as in bits. B - it matters as it is the measure of bits spent per second. As in the file size.
05:08:38myzarwhere bitrate is both size AND quality
05:08:42zuzu_um, bits per second matter when you're worried about filling the hard drive on your player
05:09:00myzaryou mean filesize wise?
05:09:09myzaryou wouldn't be storing lossless on a small drive anyways
05:09:09Shawn_Kzuzu_: yep, when I have over 1k tracks
05:09:21myzarget an 80 gig and wait for rockbox to work on it
05:09:23zuzu_bits per second * length of song == filesize
05:09:24myzarthen you'll have plenty of tracks
05:09:44 Join rkostynu [0] (
05:09:54Shawn_Kjust checked, I have over 2.7k tracks
05:10:16Shawn_Kso, yeah...FLAC is not for my iPod, lol
05:10:23Shawn_Kmy 30gb ipoid
05:10:51Soapyou should have plenty of space left.
05:11:24Shawn_Kcurrently, yes...not much if I go FLAC according to what you gusy said
05:11:24zuzu_so, line-in recording with rockbox on ipod, yes or no?
05:11:42zuzu_Soap: thanks
05:11:46myzaryou can't make your lossy music into FLAC
05:11:54myzaryou would have to get it all again, album by album
05:12:31hcszuzu_: ooh, since when?
05:12:33scorchesure you would just be a bigger version of the mp3 file
05:12:58Shawn_Kmyzar: yes, I know that
05:13:02zuzu_hcs: when what?
05:13:10myzarwhy would you scorche
05:13:14myzargo from lossy to lossless
05:13:14Shawn_Kscorche: bigger, yes....but I would have almost no space left
05:13:20hcszuzu_: sorry, I meant to direct that question at Soap
05:13:30myzarbut your ears would thank you
05:13:33Soaphcs: at least two weeks.
05:13:35myzargod's own orchestra would play on your ipod
05:13:42myzarcaressing your ears with angelic sound quality
05:13:45hcsSoap: sweet!
05:13:51scorchemyzar: i just said you could...not that you would want to
05:13:58Shawn_Keh, I still need new headphones after my last Sony LP50 broke
05:13:58*myzar licks scorche
05:14:00Soapmyzar: can you ABX LAME v3 and FLAC?
05:14:05Shawn_KI am using generic Apple
05:14:28Shawn_Knot LP50...LP51
05:14:32Shawn_Klol, my bad
05:14:35scorchewith my new headphones, 192 mp3s arent transparent anymore =(
05:14:39myzari don't get your question Soap
05:14:45Soapmy point myzar
05:14:54myzari don't get your point.
05:14:55Shawn_KI loved those, fairlt cheap and good sound
05:16:24Shawn_Kok, wish me loading my new compile onto my iPod
05:16:37Shawn_Klets see if this small change made a speed dif
05:17:15 Join jackfusion [0] (i=cfa11562@gateway/web/cgi-irc/ way, I think the screan drawing for menus actually improved
05:20:26jackfusionhi all
05:21:14 Join Mouser_X3 [0] (
05:21:14Shawn_KYES! it worked, now the vu_meter plugin is 2x as fast on the ipod
05:21:41 Quit Mouser_X (Nick collision from services.)
05:21:46Shawn_KI guess I do know what I am doing :P
05:21:46 Nick Mouser_X3 is now known as Mouser_X (
05:22:35 Join Alonea [0] (
05:23:34hcsI'm getting the impression here that recording and the cop patch aren't too friendly
05:24:29SoapI haven't tried that combo yet.
05:24:48Soapdo you have a line-in adapter hcs?
05:24:50hcsseems to freeze, and leaves me with a /recordings directory pointing back to root
05:25:02hcsI do, but I don't have it connected yet
05:25:48 Quit ctaylorr (Read error: 60 (Operation timed out))
05:25:49LloreanShawn_K: If you have a performance increasing patch, feel free to post it to the tracker
05:27:21Shawn_Kok, I will be posting stuff...this one is really so small its not worth it yet
05:27:56scorchewe would much rather have a bunch of small patches doing one thing on the tracker, than one big patch that does a lot of things
05:28:05LloreanWe *vastly* prefer small, specific purpose patches
05:28:20LloreanEasier to track it down if a patch causes unexpected side effects
05:28:36LloreanAnd "optimization" does not qualify as "one thing", just to clarify
05:29:07Shawn_Kanyways, got a question, it lcd-ipod.c even used for the ipod video or is it just ipod-video.c?
05:29:37jackfusionIs there themes for rockbox
05:30:35Lloreanjackfusion: Yes. That's what the "Browse Themes" option is for in the main menu....
05:32:14jdongLlorean: do you have a link to the patch or discussion about the 2nd core of the ipod?
05:32:27Shawn_KThe patches section on site has one
05:32:38Shawn_Kif u need it I can get it for you
05:32:41scorchejdong: there is plenty about that in the IRC logs
05:32:50Shawn_Klol, nm then
05:32:51jdongShawn_K: I've tried searching using several different methods
05:32:57jdongI guess I suck at searching :D
05:33:06Shawn_Klemme me pull it up, I have bookmarks of some patches
05:33:12Lloreanjdong: Type "coprocessor" into the search on the tracker
05:33:14hcsperhaps you only suck at finding?
05:33:24jdonghcs: I suck at both :)
05:33:48Shawn_Kthat's it
05:34:00jdongLlorean: thanks, found it
05:34:08 Part safetydan
05:34:41scorcheShawn_K: and no lcd-ipod.c is not used at all for the 5g
05:35:06Shawn_KI'm looking at the line drawing function for 16bit players...I just know it can be optimized, but I just don't see it yet >.<
05:35:14 Join BobJonkma1 [0] (
05:35:18scorcheactually, wait
05:35:29Shawn_Kthnx schrche isnt
05:36:23*Llorean still thinks he should post what he's already got, if it does improve performance.
05:37:43Shawn_K*Llorean starts thinking he should.
05:38:17scorchewhy would he post your patch?
05:39:35 Quit jhMikeS (Nick collision from services.)
05:39:35 Join _jhMikeS_ [0] (n=jethead7@rockbox/developer/jhMikeS)
05:40:48Shawn_Knot mine
05:41:14Shawn_Kanyways, I came across an idea (kinda sketchy) to simulate a contrast setting
05:41:56scorchewhy do we need a contrast setting on a color lcd?
05:41:56LloreanOut of curiosity, why is one needed?
05:42:05scorcheLlorean: slow today? =P
05:42:09Shawn_Kset up a color set which colors are "pointed" to and, well...when contrast is set, change color values
05:42:17 Join dantheman [0] (n=danielma@
05:42:22 Quit BobJonkman (Read error: 145 (Connection timed out))
05:42:26Lloreanscorche: Busy finishing Half Life 2 finally. :-P
05:42:35Shawn_Kanyways....I just thought I would bring that up since the sounce has a space for it to be implemented
05:42:39LloreanShawn_K: We didn't ask "how" we asked "why"
05:43:05Shawn_K-/* TODO: Implement lcd_set_contrast() */
05:43:41Shawn_KI never said it would be needed...just saw the todo
05:43:52Shawn_Kanyways, could be good to have settings for playing videos and such
05:45:11 Join docmur [0] (
05:45:16docmurhello all
05:45:24LloreanI'll bet that's left over from whatever file was used as a template for the one you're in
05:45:27jackfusionIs there a restiction on how big the notes files can?
05:45:45docmurI have an iaudio x5
05:45:56docmurand if I put rockbox on will I be able to watch video
05:46:51Lloreanjackfusion: Notes files?
05:46:57 Part argonel ("~")
05:46:59Lloreandocmur: Yes.
05:47:10docmurwhat codec support will I get
05:47:44 Join BobJonkman [0] (
05:47:46 Quit barrywardell ()
05:47:54Lloreandocmur: So, what I'm getting is "I haven't visited the site and looked around yet"
05:48:38docmurI did but I didn't see the codec support
05:49:01jackfusionin the orignal firmware u could only have 4k text files. Is that restriction in place in rockbox?
05:49:07Lloreandocmur: which is linked from the front page
05:49:18Lloreanjackfusion: No.
05:49:32dewdudedocmur: then you didn't look hard enough. i found that on my sidekick while standing in best buy thinking about buying my player.
05:49:33 Quit BobJonkma1 (Read error: 145 (Connection timed out))
05:50:01jackfusionso it dose not matter the size of the text file and how many u can have?
05:51:07Lloreanjackfusion: It shouldn't
05:52:26jackfusionok cool
05:52:30 Join barrywardell [0] (n=barrywar@
05:53:08 Quit barrywardell (Client Quit)
05:55:36 Quit maxkelley (Read error: 104 (Connection reset by peer))
05:57:42 Join maxkelley [0] (
05:58:18 Quit docmur (Read error: 104 (Connection reset by peer))
06:02:35Shawn_Ksince i'm just changing lcd drivers, it rockbox.ipod the only file effected with new compiles?
06:03:08scorcheyes...rockbox.ipod is the actual firmware
06:05:27 Part phrontist
06:05:47Shawn_Khmmm, this is odd...I made 2 changes and one shouln't have slowed it down if the other code was written properly...
06:06:19LloreanAh, but if you think it shouldn't have, that doesn't mean you're right for sure.
06:09:20 Quit Strath (Read error: 131 (Connection reset by peer))
06:09:52 Join webguest64 [0] (i=46748840@gateway/web/cgi-irc/
06:10:25Shawn_Kwell, if I can find the source to what's calling it when it shouldnt be...then its all good :P
06:10:28webguest64Can anyone tell me how to install the mpegplugin to my current build
06:10:34Shawn_Kdouble the optimization, lol
06:10:38 Quit Nico_P (Remote closed the connection)
06:11:11Aloneadoes rockbox handle text files well enough to read a story?
06:11:23Shawn_Ki would think so, lol
06:11:32webguest64Can anyone tell me?
06:11:47 Quit JoeXBorn (Read error: 145 (Connection timed out))
06:12:09Lloreanwebguest64: It comes with updated official builds.
06:12:15LloreanOn the targets that support it
06:12:22Shawn_Kwow, u must get that alot
06:12:47webguest64So do i just put an Mpeg file into my main player and itll play?
06:12:51AloneaI was just wondering. Splitting up a story I have into one chapter per file since there isn't a "bookmark" thing for text I dont think
06:13:15Lloreanwebguest64: See the PluginMpegplayer page in our wiki.
06:13:34*scorche ponders reading a story on the nano
06:13:38*scorche shudders
06:14:07scorcheAlonea: and i do recall something about a bookmark for text...
06:14:18 Quit webguest64 (Client Quit)
06:14:20Lloreanscorche: I know it was disabled at some point, I'm not sure if it ever got re-enabled
06:14:32scorcheLlorean: yeah...i remember that discussion
06:14:34Shawn_Kwhy would it be disabled?
06:14:37Technithe shellscript finally finished, i ran make in the build directory and it says "make: *** No targets specified and no makefile found. Stop. "
06:14:56scorcheTechni: follow the instructions
06:15:06scorcheit says to run tools/configure first
06:15:18Technidid that. ill try again
06:15:30Aloneareally? wow. I will check it out once I get all these files, or at least a good batch of them, done. How do you open them? like you would open a video? guessing there is a text viewer somewhere on the open with... ?
06:15:33hcs../tools/configure from the build directory
06:15:46LloreanShawn_K: Some change probably broke it, and nobody has fixed it, would be my guess
06:15:46 Join LoL [0] (
06:17:27Techniredid ../tools/configure, redid make, same error
06:17:55 Join BobJonkma1 [0] (
06:18:05 Quit BobJonkman (Read error: 145 (Connection timed out))
06:18:31LloreanIs configure giving any error messages?
06:18:57Techni../tools/configure: line 1358: arm-elf-gcc: command not found ah, the shell script didnt do everything it was supposed to
06:19:39hcsshell script doesn't add it to PATH for you, nor does configure
06:20:02Techniah boo
06:20:43Technihow would i do that?
06:21:11hcswhat shell are you using?
06:21:27Techniyellow dog linux/enlightenment 17
06:21:54hcsexport PATH=$PATH:/usr/local/arm-elf/bin
06:21:56scorcheshell...not environment
06:22:56hcs(thus I assume bash)
06:24:31 Join Strath [0] (
06:26:11Shawn_Ki finally found an optimization for the line drawing function (dual coordinate one), but I decided to go back to it then forgot what it
06:27:20TechniLlorean: I sent you the messages it gave me via PM, since it's like 6 lines long and I don't wanna clog/spam up the channel
06:27:27Techniand thank you hcs
06:27:54hcsTechni: put them on and everyone can see them
06:28:41LloreanTechni: I assume you're not registered, because I didn't get the PM. We usually use pastebin so that more than just one person can look at it
06:30:03 Join JoeyBorn [0] (
06:30:16LloreanTechni: Why are you still using the 2.5 source?
06:30:44LloreanActually, arm-elf-gcc shouldn't be used with that at all
06:30:47Techniodd, i built the new version...
06:30:54LloreanSo what's the deal, why's the folder named 2.5?
06:31:07Technicause i used the folder i made before :(
06:31:41hcsa) what directory are you running in b) what directory did to run tools/configure in?
06:31:42Shawn_Kok, i still don't understand what the "cpu boost counter" means
06:31:46hcsthey should be the same
06:32:03Techniodd, the config is the newer one since it has gigabeat, where 2.5 didnt...
06:32:08LoLso its taken more then 15 minutes for my nick D_C to time out :(
06:32:09Shawn_Kdoes the boost counter mean the CPU is more free, or "overclocked"?
06:32:16 Quit Strath (Read error: 104 (Connection reset by peer))
06:32:18 Quit D_C (Read error: 110 (Connection timed out))
06:32:22LoLthere we go lol
06:32:23 Nick LoL is now known as D_C (
06:32:34LloreanD_C: if you register your nick then you can have nickserv kill it
06:32:46Mouser_XIt's what I have to do...
06:32:55hcsShawn_K: it goes up when something needs boosting, back down when it no longer needs boosting, so if it is over 1 the cpu is boosted (as I understand it)
06:32:57LloreanShawn_K: If the boost counter is greater than 0 the CPU is running at full speed rather than a slower speed.
06:33:01hcs*over 0
06:33:09LloreanShawn_K: Basically the number is "How many things still want faster speeds"
06:33:42LloreanTechni: You should probably have done things cleanly, as the fact that it's using the 2.5 source code directory for some strange reason could very well be the cause of your problems
06:34:38 Quit [TEHb]_ (Read error: 113 (No route to host))
06:35:35Shawn_Kwell, this is very good then...with all this crap running on rockbox at once, befair it went between 30 and 75mil, not it's going from 24 to 30
06:35:44Shawn_Kand with lower boost
06:36:13hcsgimmie dat patch
06:36:21Technithe problem was cause i use windows, and made a folder called "My documents" to feel like home, and it didnt like the space,
06:36:30Shawn_Kwell, the numbers jump...just means its much close to 30
06:37:00Shawn_KI'm pretty much stripping out and optimizing code in the graphics libraries
06:37:47Shawn_Ksome failsafes I'm seeing are unecessary and a waste of processing, and it may not be much, but when called 10x or more per adds up
06:40:56LloreanTechni: You didn't check out the full CVS, which means you don't have the ability to make bootloaders.
06:41:02Aloneahey, is there a way to make folder hidden and have them not show up on rockbox?
06:41:39LloreanTechni: cvs -z3 co rockbox-devel should've been used instead of cvs -z3 co rockbox
06:41:45hcsAlonea: rockbox respects the hidden file property, and also hides files beginning with ., unless you choose to show "all"
06:41:49Technii moved the stuff out of the 2.5 directory. when i CVSed the new copy, it put it inside it
06:41:51LloreanAlonea: In "Supported" view mode, hidden files are hidden
06:42:27hcsLlorean: should I just let you handle these, then?
06:42:52Aloneaok, so in windows I make the folder hidden then? and where is this show all option you are talking about?
06:43:26Lloreanhcs: I'm only half around, sorry if I'm repeating things you've said
06:43:54hcsyou probably know better than I, anyway
06:44:04LloreanYeah, but odds are good I'll miss things.
06:44:08LloreanI'm only barely paying attention
06:44:46hcsAlonea: what device are you using rockbox on?
06:45:10Aloneahcs: I have a Toshiba Gigabeat F-40
06:46:24 Join dafty [0] (i=562a4ad6@gateway/web/cgi-irc/
06:47:36hcsthe file view settings are described here:
06:48:26 Quit dafty (Client Quit)
06:48:38 Join Rob222241 [0] (
06:50:30 Join Jsunu [0] (
06:50:47Shawn_Kas I prepare patches, I will be releasing my optimized builds
06:50:56Shawn_Ki just posted my first one on the site ^^
06:51:24Shawn_Kso if anyone is interested, or even wants to tell me if it's an improvement...that would be awesome
06:51:44tehsmoso, what does it optimize, mostly?
06:51:54tehsmoyou said display, so would it improve scrolling on the ipod?
06:52:03Shawn_Kthe GUI drawing at the moment
06:52:12Shawn_Kand so far it seems to improve scrolling on mine
06:52:36LloreanShawn_K: If you're going to distribute a build, you're required to distribute your modified sources as well
06:53:00Shawn_Koh, really? not all on the site had been
06:53:25Shawn_Kthen I might as well keep them to myself until I'm ready?
06:53:27Ajaxincdoes anyone here compile on a *nix enviroment that can help me set up my compiling enviroment for rockbox
06:53:34tehsmoalso patches would be nice, for those of us who would want to add other patches to builds
06:53:54tehsmoAjaxinc: have you used the script?
06:53:56LloreanShawn_K: It's a term of the GPL.
06:53:57Shawn_KI said I will be releasing patches after I get the system down
06:54:08Ajaxinci didntknow where was such a script
06:54:09jdongShawn_K: that's pretty standard of the GPL
06:54:11Ajaxincwhere can i dl?
06:54:15jdongyou need to publish your sources
06:54:16 Quit jeffb (Read error: 113 (No route to host))
06:54:16tehsmoit's in the source
06:54:17LloreanIf you distribute a GPL licensed binary, you have to provide accompanying sources.
06:54:37puetzkShawn_K, quilt is a nice tool for managing a set of patches as you work on changes
06:54:38puetzkI use it a lot
06:54:51Shawn_Ki will try that
06:54:58puetzkhelps me keep my own work straight, and the patch files are right there when I want them
06:55:10Shawn_Kanyways, link :
06:55:16Shawn_Ki willadd sources soon
06:55:16tehsmoAjaxinc: - this describes how to set up a cross-compiler
06:55:46LloreanShawn_K: Soon had best mean "within the next two hours"
06:56:04LloreanThe GPL only requires you to provide sources when requested, which is why the others aren't being hassled about it, but you've now had a request.
06:56:05tehsmoheh. .quilt is a remarkably hard application to google for. :)
06:56:21Shawn_Kyeah, I am rar'ing it atm
06:56:36LloreanI don't really understand why you don't want to post a patch to the tracker though
06:56:37puetzkit's also packaged in cygwin for any using that
06:56:42tehsmoah, thanks
06:56:48LloreanThe best way to get developers to test it, and that's who you want, is to post it to the tracker
06:57:03puetzk(and in debian, darwinports, and I'd assume most anywhere else with package management)
06:57:07LloreanPosting it to the unsupported builds forum gets hundreds of relatively clueless people asking you to include Patch X as well
06:57:54puetzkthe basic premis is that you're working on a base folder, and a stack of patches
06:58:02puetzkyou can use quilt push and quilt pop to move up and down the stack
06:58:06Technimake[1]: *** No targets specified and no makefile found. Stop.
06:58:18puetzkquilt import [-p n ] to bring in a patch file from somwehere else
06:58:23puetzkquilt new <name> to make a new one
06:58:36puetzkat any point you're working on the patch at the top of the stack
06:58:41Techniodd, i can see Makefile in that directory
06:58:47puetzkyou use quilt add <path/to/file> to add it to the patch
06:58:49puetzkthen hack away
06:58:55tehsmopuetzk: cool stuff. that sounds much easier than what I've been doing
06:58:56Shawn_K;( the sources are over 11mb
06:58:59puetzkand use quilt refresh when you want the patch file regenerated
06:59:10puetzkit keeps track of all the relevant 'original' vesions to diff against
06:59:13puetzkit's very slick
06:59:20tehsmosounds like it'll be a lot easier to handle cvs updates with it
06:59:24Shawn_Kmy web server...well, if too many DL...well, I have limited BW and dont like files over 1mb
06:59:29puetzkyes, you just do quilt pop -a (all)
06:59:31scorcheShawn_K: what do you think a diff is for?
06:59:37puetzkthen cvs up and quilt push -a to put them all back
06:59:43 Join Criamos [0] (
07:00:01Shawn_K3min left for source upload
07:00:14puetzkit grew out of andrew morton (linux 2.6-mm maintainer)'s patch scripts
07:00:16scorcheShawn_K: why worry about size and a diff when you can just put the diff on the tracker?
07:00:31Shawn_Klol, because I havent dont diff before
07:00:32scorches/a diff/your webserver
07:00:39puetzkand he handles lots of disjointed patches and new updates :-)
07:00:44scorchethen read how to on SimpleGuideToCompiling
07:00:50 Join strabes [0] (
07:00:50Shawn_Klol, I did
07:00:54Shawn_Ki get lazy
07:01:02Shawn_Kjust give mne time man
07:01:09Ajaxincwhere can i get the latest rockbox source?
07:01:30tehsmoAjaxinc: the docsindex is your friend:
07:01:30scorcheAjaxinc: from cvs
07:01:55Aloneaok, I like the text so far, now, if there was a book mark command there, how would I get to it?
07:02:00Technii keep getting zip warning: name not matched: FWIMG01.DAT
07:03:01AjaxincShawn_K: I havnt registered my nick so it wont let me talk in pms
07:03:09Shawn_Kok, sry
07:03:15Shawn_Kwell, yes?
07:03:17AjaxincShawn_K: i dont program ajax, its just been my nickname since i was 2
07:03:22Shawn_Koh, lol
07:03:34Shawn_Kk, j/w I wrote quite the ajax script
07:03:50hcs(stronger than dirt)
07:03:54Shawn_Kwoohooo, sources done uploading
07:04:04Shawn_Know to fix post to have them
07:04:34Shawn_Kanyone try my build yet?
07:04:34Techniand FWIMG01.DAT is the only file I need
07:04:41tehsmoShawn_K: that's good enough to satisfy the GPL, but you really should learn to make a diff eventually
07:04:48Shawn_Koh, I will
07:04:50Shawn_Kno doubt
07:04:55Shawn_Kjust nopt tonight, lol
07:05:00tehsmook, that's cool
07:05:02***Saving seen data "./dancer.seen"
07:05:23LloreanTechni: Did you ever co rockbox-devel like I told you to?
07:05:33Shawn_Kplz, I love 2nd opinions, I wanna know if it works well
07:05:51Technicvs -z3 co rockbox-devel
07:05:57 Quit Rob2222 (Connection timed out)
07:06:20scorcheShawn_K: how hard is it really to just type cvs diff u > blahblah.patch ?!?
07:06:28tehsmoShawn_K: I'm waiting for the patch, easier to deal with for me
07:06:40Shawn_Kok, lol
07:08:13Aloneaok, it doesn't appear to have bookmark, but apparently it does. When I went to "quit" while the text was elsewhere, and I went abck to the file and opened it, it opened right where it left it.
07:08:16LloreanTechni: Do an LS of your rockbox-devel folder, then cd to your build folder, configure and make, and show all of that in a pastebin
07:08:18Techniat the make stage I get make[1]: *** No targets specified and no makefile found. Stop.
07:10:13Techniah. my bad again, i was in rockbox, not rockbox-devel
07:10:16Technitrying make again
07:12:10LloreanMake fullzip isn't used with the bootloader...
07:12:29 Quit strabes (Remote closed the connection)
07:13:05Technijust zip?
07:14:58LloreanYou just need the one file
07:15:01LloreanYou don't need to make a .zip
07:16:41Aloneahey, how do I get my volume buttons back on the gigabeat? they no longer work and someone else mentioned it too on the gigabeat irc page. The only way to change volume is with the remote control or at the volume screen under sound
07:18:08jackfusionIs there a gui theme for rockbox?
07:18:50jackfusionFor when u are going around viewing menu and stuff?
07:19:19Mode"#rockbox +o scorche " by ChanServ (ChanServ@services.)
07:20:03Topic"Developers, please visit" by scorche (n=scorche@rockbox/administrator/scorche)
07:20:12Mode"#rockbox -o scorche " by ChanServ (ChanServ@services.)
07:20:22jdongmind if I ask a few quick ipod questions?
07:20:33jdongI'm still a relatively new rockbox users
07:21:01jdong(1) will Apple's firmware updater still happily upgrade the firmware of an iPod with rockbox?
07:21:12jdong(2) I'd assume after said upgrade I'd have to reinstall the bootloader?
07:21:18scorcheyes, although you will have to reinstall the bootloader
07:21:55 Join cynicalliberal [0] (
07:25:18 Quit Vyrus001 ("Leaving")
07:25:23Shawn_Khow do i tell configure/make to only compile rockbox.ipod?
07:25:40Shawn_Kor must that be done manually?
07:25:46 Join strabes [0] (
07:25:48tehsmoperhaps 'make rockbox.ipod' instead of make would do it
07:26:01scorcheShawn_K: you dont
07:26:14scorchealthough, if that was the only thing modified...
07:26:14tehsmoshould work, if the makefile is set up like most are
07:26:23 Join Ghoulunatic [0] (
07:26:34Shawn_Kyes, only thing modified
07:26:40Shawn_Ki;m not that dumb :P
07:28:22 Join Hsystem [0] (i=4126c14e@gateway/web/cgi-irc/
07:29:24 Quit mathgl ("Quitte")
07:29:48jackfusionIs there any tutorials for making themes
07:30:19jdongwhat's the story with rockbox charging over USB
07:30:26 Quit Jsunu ("Chatzilla 0.9.77 [Firefox]")
07:30:27jdongI've read so many reports that it doesn't work
07:30:37jdongbut on my laptop it seems to charge just fine?
07:30:38scorchejackfusion: yes...see the manual. the wikipage CustomWPS, and one other wiki for it
07:30:42jdongeven while playing back music
07:30:53scorchejdong: it works fine
07:31:05jdongscorche: really?
07:31:22jdongwhat about threads like
07:31:27jdongis the problem no longer relevant?
07:31:32jackfusionScorche: I want to make a gui for the menus and stuff
07:31:49scorchejackfusion: then that isnt our type of theme
07:32:04Techniack, the gigabeat shuts off while being turned on :(
07:32:07scorchethat involves delving into the code
07:32:28jackfusionthats right it is a suite that I want to make.
07:32:46Lloreanjdong: Fine as in "it works consistently" though it charges much slower than the retail firmware.
07:32:50Technigoes to the charging screen
07:32:56Technican not start it to go to usb mode
07:32:59Ajaxinchow do i run a .sh file in linux?
07:33:19Hsystemyou have to make it executable, and then use ./
07:33:28jackfusionwant language is rockbox writen in?
07:33:29jdongLlorean: ah, ok that's the kind of answer I'm looking for :)
07:33:34jdongLlorean: how well does disk mode charge?
07:33:35Ajaxincjack: c
07:33:38jdongstill slower than retail?
07:33:40jdongsame as rockbox?
07:33:42jackfusionthank u
07:34:02Technii think it bricked my gigabeat :(
07:34:11Lloreanjdong: Disk mode if their software, not ours.
07:34:16scorchejdong: we dont have any control over disk mode, so it hasnt changed from when you last heard about it
07:34:25AjaxincHsystem: how do i make it an executable?
07:34:36jdongI understand that disk mode has nothing to do with rockbox
07:34:37scorcheAjaxinc: man chmod
07:34:44jdongbut I can't find any definitive answer about charging rate in disk mode
07:34:48LloreanTechni: Well, I told you that you should wait until we provide one. If you did, you get no sympathy from me. Does the Gigabeat have a recovery process?
07:35:01Lloreanjdong: It's the same as retail, for me at least
07:35:02scorcheLlorean: it does
07:35:19jdongLlorean: ok, that's cool to hear
07:35:30Hsystemchmod u+x or something like that ...
07:35:34jdongone last thing... what's the typical lifespan of the backlight for the ipod?
07:35:44jdongshould I be concerned about leaving it to charge overnight in disk mode
07:35:51jdongwhere the backlight is dimly lit at all times
07:36:01scorcheit is an LED...they last a looooong time
07:36:34jdongoh wow it's a LED?
07:36:46jdongnot the typical flourescent thingies used in laptop screens?
07:38:30jackfusionWhere would I put .ipod for rockbox firmware tool to see?
07:38:47scorcherockbox firmware tool?
07:39:12 Quit Criamos ("( :: NoNameScript 4.03 :: )")
07:39:17Aloneathe gigabeat should have a recovery process. I remember reading about it somewhere. I have never done it though. Hope you get yours going again Techni
07:39:27 Quit dantheman (Read error: 104 (Connection reset by peer))
07:39:49scorcheTechni: for more info about it, join #gigabeat
07:41:02 Quit strabes (Remote closed the connection)
07:41:43 Join decayedcell [0] (i=792cceed@gateway/web/cgi-irc/
07:41:56decayedcellcheers linuxstb
07:41:58decayedcellmpg working now
07:42:19decayedcellframeskip on bug fixed
07:42:33jackfusionwhen ur in the file browser the last item is rockbox and I read that u can use that to see other firmwares without changing from the orginal.
07:42:43 Quit Ghoulunatic (Read error: 60 (Operation timed out))
07:43:43scorchejackfusion: that is ROLO, and that is not quite the best description for it...see the wiki pages about it
07:43:52decayedcellit appears that videos larger than 320x240 on the 5G Video overfill the buffer or something... mpegplayer freezes on videos larger than that size
07:43:58jackfusionthank u
07:44:01scorcheROLO also does not work correctly on the ipods yet
07:44:42Hsystemmpg still buggy ... too slow in frames.
07:44:53 Join World_Entity [0] (n=chatzill@
07:45:03World_Entitysup you guys...
07:45:08World_Entitygot a ? real quick
07:45:17World_Entityon a mac here
07:45:46World_Entity./ipodpatcher : permission denied
07:45:57Lloreandecayedcell: Well, we do tell you to convert them first
07:46:01World_Entitycan't run the ipodpatcher...i've done it before
07:46:59decayedcellis it possible for mpegplayer to detect the size of the video?
07:47:16decayedcellthat would prevent anyone from getting into the same hole that I dug myself in
07:47:19decayedcelllol =/
07:47:50scorchedecayedcell: as Llorean said, we do tell you to convert them first...
07:47:53jdongWorld_Entity: is it chmod +x?
07:48:13Hsystemdoes that works on a mac? chmod lol iknow it is for linux.
07:48:19jdongLlorean: Is there any hope for taking advantage of the HW video decoder on the ipod?
07:48:26puetzkHsystem, works the same on a mac
07:48:33jdongdocumentation is probably scarce/nonexistant, I expect
07:48:37puetzkthey're BSD unix machines underneath :-)
07:48:39scorcheHsystem: OSX is UNIX based
07:48:49jackfusionwhere can u find .mod files?
07:48:53World_Entityjdong: not sure...
07:48:56Lloreanjdong: There's no docs, so hope is slim.
07:49:00Hsystemah ok.
07:49:06World_Entitychmod has to do with permission settings...what about that sudo shit
07:49:14World_Entitykind of a nub...sorry man
07:49:17jdongpuetzk: their kinda partially bsd derived ;-)
07:49:37jdongWorld_Entity: you may or may not need sudo
07:49:39decayedcellsudo sets permissions
07:49:42decayedcellof users
07:49:42jdongchmod marks files as executable or not
07:49:51jdongsudo grants your command adminsitrator access to your system
07:50:00jdongwhich may or may not be necessary to low-level access your iPod
07:50:09Ajaxinchow do i use a .sh file in linux?
07:50:09Hsystemor you could use −−> su if you know the password for the root.
07:50:13jdongI don't use OSX so I wouldn't know
07:50:17Ajaxincparticularly the
07:50:18World_Entitywell i did this a few days ago and had the same problem...after trying the same thing again and simple worked..i shit you not
07:50:23jdongAjaxinc: same; chmod +x it, then ./
07:50:28jdongor sh
07:50:40scorcheWorld_Entity: the cussing is not needed
07:50:57jdongWorld_Entity: you must've hammered the command so much that Apple figured out you wanted to mark it executable :D
07:51:01World_Entitymy bad
07:51:10World_Entityha ha...
07:51:21World_Entitycool maybe ill do the same
07:51:38jdongbut you gotta chmod +x exectuables before they'll run
07:51:47World_Entitycool lemme try that
07:51:48Aloneanight all. Hope someone fixes whatever disabled the volume buttons on the gigabeat.
07:51:49puetzkjdong, which is all any of the modern *BSDs can claim :-)
07:52:02 Quit Alonea ("Chatzilla 0.9.77 [Firefox]")
07:52:13Technii managed to force recovery mode, and restore the original boot loader
07:52:13jdongpuetzk: what I was saying is OSX's bsd ancestry is much far-stretched compared to any of the modern BSD decendants
07:52:14World_Entityyou the man jdong
07:52:24World_Entitythanks bro...that was the ticket
07:52:28 Join ender` [0] (n=ender@
07:52:31Techniin case someone else needs it
07:52:31scorchelets not get into a BSD discussion...
07:52:31jdongnp, World_Entity, any time :)
07:52:32World_Entitysorry for the cursing schorce
07:52:57*jdong angrily exports tracks from gtkpod
07:53:03jdongstupid filename mangling :)
07:53:11jdongmakes it so impossible to identify songs
07:53:16puetzkscorche, aww :-P
07:53:22jdongand no, I didn't think far enough to tag them!
07:53:36 Quit World_Entity ("Chatzilla 0.9.77 [Firefox]")
07:53:59decayedcellanyone know is there a windows based .fnt font editor
07:55:27puetzkjdong, gtkpod makes an itunesdb, no?
07:55:36jdongpuetzk: yep
07:55:41puetzkIIRC there's a program in the wiki that can convert that to rockbox tagcache form
07:55:47puetzkso you'd get titles
07:56:04jdongyeah, there is
07:56:11jdongbut I'd rather switch over to flat-files
07:56:13scorchedecayedcell: we use our own font format...edit BDFs then convert them
07:56:32jdongit'd make my ipod a lot more useful as a song storage device for computer playback
07:56:37puetzkbut it could be a start; might not be too hard to hack it to store that data as id3 tags instead of as a generated tag cache
07:56:41 Quit Mouser_X (Read error: 110 (Connection timed out))
07:56:48jdongeven if rockbox understands gpod23243.mp3, I don't :D
07:56:50puetzk(or gtkpod may already have such a feature)
07:57:03jdongpuetzk: I've got a pretty semi-auto solution hacked out
07:57:09puetzkrun with it then
07:57:21jdonggtkpod can export on %a - %t.mp3 files
07:57:23jdongfrom the itunesDB
07:57:34jdongand then a tagging script can easily turn that into ID3 tags
07:57:44scorchewe dont care about gtkpod...take it to PMs please
07:58:15decayedcellis there a windows based bdf editor then lol
07:58:26decayedcellI couldn't get fontforge to work under cygwin =/
07:58:27scorchedecayedcell: see the wiki page CreateFonts
08:00:56 Join Mouser_X3 [0] (
08:00:56 Quit Hsystem ("CGI:IRC (EOF)")
08:01:03 Nick Mouser_X3 is now known as Mouser_X (
08:01:44 Quit hcs (Read error: 131 (Connection reset by peer))
08:06:13 Join matsl [0] (
08:10:28Technilooks like I'm forced to wait then, any progress report on when your boot loader should be done?
08:11:24 Join pondlife [0] (n=Miranda@rockbox/developer/pondlife)
08:18:19LloreanWhy are you posting that link here?
08:18:33 Quit Mouser_X (Read error: 145 (Connection timed out))
08:19:10 Join Exclamation [0] (
08:19:36Exclamationi just have a few q's before i put this on my ipod
08:19:53Exclamationwill the interface change?
08:20:06scorcheof course
08:20:06LloreanVery much so. Have you read the manual?
08:20:24scorcheplease do =)
08:20:33Exclamationhow much different is it? better? worse?
08:20:42scorchethat is for you to decide
08:21:03LloreanMost of us think it's better.
08:21:12LloreanBut, y'know, you're asking in #Rockbox, you can't call us unbiased
08:21:22Exclamationbtw, i just got a 30g video, thats supported right?
08:21:29decayedcelli have one
08:21:29scorcheor neutral as we like to call it ;)
08:21:36 Join Vyrus001 [0] (
08:24:22ExclamationIs there any advantage with rockbox in terms of playing videos?
08:25:13Technithe link worked
08:25:23Techniwas very easy, no compiling
08:25:26Technijust copy/paste
08:25:28*myzar licks scorche
08:25:34myzarscorche, are you married?
08:25:48scorchewe have discussed this before, and i said no...
08:25:53*scorche shoos myzar away
08:26:04*myzar creeps up behind scorche
08:26:41*scorche readies the boot stick in case someone were to creep up on him from behind with more off-topic-ness
08:27:23LloreanTechni: Yes, and that link he posted is very likely just what he got when he compiled the bootloader.
08:29:31 Join GodEaterWeb [0] (i=c2cbc95c@gateway/web/cgi-irc/
08:29:42Ajaxinchow do i "Make your PATH include :/usr/local/arm-elf/bin
08:29:47Ajaxincin linux
08:30:46pondlifeHmm, it seems that "Whoops" is gaining ground in CVS comments. What's wrong with the traditional "Oops" ? ;-)
08:31:04pondlifep.s. Happy New Year (at last)
08:31:22 Join _Veseliq_ [0] (n=veseliq@
08:31:30 Quit jackfusion ("CGI:IRC (Ping timeout)")
08:32:05scorchepondlife: they felt that "Oops" was just too short (and see the topic)
08:32:38*pondlife likes short words
08:33:29pondlifescorche: I did have a look, will add some info if I get time, but that's a bit unlikely at the mo.
08:33:40scorchewhen you get the time =)
08:33:46Technialright, now that i have rockbox working, how would i go into usb mode?
08:33:53pondlifeI wish I didn't have to use that phrase.
08:35:20AjaxincXbox ~ # dir
08:35:20Ajaxinc-bash: dir: command not found
08:35:25Ajaxincthats not good right?
08:35:38Techninvm, the dock isnt supported yet
08:40:06 Quit Shawn_K ()
08:42:20 Join juxtap [0] (
08:43:18Ajaxincwhen using what is the directory that it wants me to add the the PATH again?
08:43:24 Join norbusan [0] (
08:44:31Techniwhy dont you guys just offer the precompiled file that site links to, rather than making ppl jump through hoops and possibly brick the thing?
08:45:01LloreanTechni: We can compile bootloaders now, without a problem
08:45:06LloreanWe just HAVEN'T POSTED IT YET.
08:45:07LloreanAs I told you
08:45:22decayedcellreading the manual helps
08:45:27decayedcella lot :)
08:45:52Technibut i got it to compile too, but it compiled incorrectly
08:46:05LloreanYou also did about a million things wrong along the way
08:46:19Ajaxincwhen using what is the directory that it wants me to add the the PATH again?
08:46:30LloreanThere will be a precompiled bootloader posted for downloading at some point
08:46:35LloreanYou were impatient, and you had problems.
08:46:41Technithats no excuse when it gives me the file, and gives no errors
08:46:48Ajaxincwhen using what is the directory that it wants me to add the the PATH again?
08:47:03LloreanTechni: You can compile ANYTHING and have it name the file the file you got.
08:48:06LloreanTechni: Error messages can only tell you if the compile failed, not if the file you got will do what you expect it to do.
08:48:44decayedcellit should do it automatically Ajaxinc
08:48:53decayedcellbut read this
08:49:00decayedcellnear the bottom if you still have problems
08:49:27Ajaxincits jsut that when ever i try to compile my hellworld script it has a ton of errors
08:49:42Ajaxincand its the exact hellworld script that is on the plugins tutorial on the rockbox site
08:49:57decayedcellwell whats the first error? ;p
08:51:07Ajaxincone sec
08:51:11Ajaxinci closed my ssh session
08:51:16Ajaxinchave to try re compiling
08:51:43Ajaxinchelloworld.c:1:20: plugin.h: No such file or directory
08:52:21 Join webguest99 [0] (i=c27f0812@gateway/web/cgi-irc/
08:52:49decayedcellis the file there? lol
08:53:00Ajaxincis it exposed to be in that folder?
08:53:55decayedcellwell the problem could be that you have an incomplete source to start with
08:54:15Ajaxincwell my source isnt even on my linux box that im compiling on
08:54:22Ajaxincit should probaly be on there shouldnt it?
08:54:38Ajaxincany directory?
08:54:44Ajaxincspecific directory
08:55:01decayedcellhelloworld script lets see
08:55:12Ajaxincfor plugin
08:56:30decayedcellit would appear
08:56:32 Quit webguest99 (Client Quit)
08:56:40decayedcellthat you need the full rockbox source to start off with
08:56:51decayedcella quick search on my end reveals that plugin.h
08:56:56decayedcellis in rockbox-devel/apps
08:57:07decayedcellrockbox-devel being the source off CVS
08:57:18decayedcelli'm not an expert on compiling
08:57:31decayedcellbut i'd say that you need the rockbox source as well as that helloworld thing
08:57:39decayedcell(correct me if I'm wrong anyone)
08:58:06Ajaxincwow, i just put that file in the directory then re complied
08:58:11Ajaxincgot about 30 lines of error codes
08:59:11decayedcellmissing files again?
08:59:27Ajaxincill try again
08:59:27Ajaxincone sec
08:59:43decayedcellperhaps your compiling environment isn't set up properly
08:59:58decayedcellyou should ask the other guys for help lol i'm stabbing in the dark here
09:04:17 Join petur [0] (i=d4efd6a6@gateway/web/cgi-irc/
09:05:06***Saving seen data "./dancer.seen"
09:07:35 Join [23]Fuzzipelz [0] (i=54b1e023@gateway/web/cgi-irc/
09:09:59 Quit [23]Fuzzipelz (Client Quit)
09:13:11 Quit Exclamation ("Leaving")
09:20:53 Quit pill (Remote closed the connection)
09:21:24 Quit scorche (Read error: 104 (Connection reset by peer))
09:22:03 Join scorche [0] (n=scorche@rockbox/administrator/scorche)
09:23:26 Part BobJonkma1
09:32:01 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
09:38:27safetydanlinuxstb, for the simulator, is it sufficient that SDL resample the audio when you do pcm_set_frequency? I think that will be easier than reiniting the entire SDL audio subsytem.
09:42:48 Nick myzar is now known as myzar|away (n=myzar@
09:43:07GodEaterWebI'd admire that Shawn_K's enthusiasm, but he sounds like he thinks he's more intelligent than he is....
09:45:17safetydanGodEaterWeb, he's definitely enthusiastic. It will be interesting to see what patches he produces.
09:45:56Lloreansafetydan: This is what he's done so far:
09:47:02safetydanuh... that just seems to remove some safety checks and swaps tabs for spaces
09:47:43Lloreansafetydan: He also moves a bit of stuff outside of a loop near the end, and I'm not sure what the switch case he's removed does at all.
09:48:18safetydanWell I'm sure he' measured it to be n frames per second faster :)
09:48:40scorche"vu_meter is 2x faster now"
09:49:10 Join BHSPitMonkey_ [0] (
09:50:46GodEaterWeband he couldn't manage to add it to Flyspray...
09:51:16scorchehe said he will tackle patches tomorrow
09:51:49GodEaterWebI've not noticed pasting something to pastebin to be a *massive* amount more effort than doing the same to Flyspray
09:52:01LloreanGodEaterWeb: I made the diff.
09:52:05LloreanHe posted his whole source tree
09:52:09GodEaterWebyeah I saw that
09:52:19GodEaterWebI couldn't be bothered to download his entire tree
09:52:33GodEaterWebnice work on ragging him on the GPL though :)
09:52:57LloreanWell, it was true. :-P
09:53:14LloreanEven if it was merely to satisfy my curiosity
09:53:41GodEaterWebI liked the "soon better mean within the next couple of hours" and the unspoken threat of Llorean's lawyers :)
09:54:07LloreanNah, it was the spoken thread of "I can delete your thread"
09:54:29GodEaterWebyou clearly type "thread" too many times a day
09:54:34LloreanWell, it would've been the spoken threat, but I didn't need to speak it. :)
09:54:36petur"and ban you for ever"
09:54:44 Quit evilnick (Read error: 110 (Connection timed out))
09:54:56LloreanI actually have to make use of the ban hammer far less often than I feared I might.
09:54:58safetydanAm I the only one not getting rockbox CVS emails?
09:55:15LloreanWe have a surprisingly good ratio of "People who'll listen to reason" vs "LOL u gunna b4n me?" types.
09:55:21GodEaterWebwell I'm not getting them either, but I'm not subscribed to them either :P
09:55:33Lloreansafetydan: The script is broken what with the single-coreness of the cvs server right now
09:55:42LloreanI believe
09:55:45GodEaterWebyeah - what the hell is the deal with people putting "LOL" in front / behind everything they type ?
09:55:58GodEaterWebthe world must be full of very cheerful (but stupid) people
09:56:09peturlol yeah :p
09:56:22safetydanLlorean, ok so is that same issue affecting flyspray and other emails then?
09:56:31*safetydan gets lonely without his regular Rockbox related emails
09:56:56*petur just recovered from another diskcrash :/
09:57:10pondlifeAnyone object if I close the "volume label is deleted" bug - ? It's not been reported in a long time.
09:57:28peturI'm not sure it has been fixed at all
09:57:37peturI'm even sure it is still there
09:57:49pondlifeAre yousure it's a Rockbox bug?
09:58:03peturgood question
09:58:11pondlifeI doubt it myself.
09:58:18pondlifeToo specific IMHO
09:58:26Lloreansafetydan: I don't know. I still get the occasional flyspray email. I think it's a case of "some emails go out, and some don't"
09:58:44pondlifeI would expect files/directories to go missing too, and CHKDSK to give errors at least.
09:58:55pondlifeThis would be reported more often.
09:58:59GodEaterWebLlorean: Bagder said sendmail basically throttles back when the server is above a certain load
09:59:03peturpondlife: I once went through that code but only found other bugs ;)
09:59:40pondlifeOK, I'll leave it alone if you think there's a chance it's a genuine Rockbox bug.
10:00:07 Quit bun-bun ()
10:00:17peturwell I've had it once and I there's been no 'gotcha' commit for that one
10:01:33peturmaybe a compatibility issue between RB and windows regarding FAT/DIR handling
10:04:27 Quit BHSPitMonkey (Read error: 110 (Connection timed out))
10:06:30 Quit decayedcell ("CGI:IRC (Ping timeout)")
10:07:25 Join Mouser_X3 [0] (
10:08:12pondlifeHmm, are the mailing lists down too?
10:08:32*pondlife must stop Hmm-ing for 2007
10:09:58 Join Mouser_X [0] (
10:10:12 Quit cynicalliberal ("Chatzilla 0.9.77 [Firefox]")
10:11:43safetydanpondlife, if what Llorean is saying is correct, then everything being sent is affected
10:12:07peturyummie yummie ->
10:12:56peturseems they plan 1.8" versions :D
10:15:01 Nick BHSPitMonkey_ is now known as BHSPitMonkey (
10:16:11 Join doktoreas [0] (n=geko@
10:16:36 Join My_Sic [0] (
10:17:45doktoreashi to all!
10:18:02doktoreasi have succesfull installed rockbox on my ipod and all is fine
10:18:22doktoreasi wonder if is there a plugin that create a sort of "collection"
10:18:55doktoreasinstead of moving inside folders
10:19:28peturcheck 'database' (formerly known as tag db)
10:20:50doktoreaspetur, in the plugin folder?
10:21:18peturcheck the manual first to see it's what you want ;)
10:21:23 Join softi_42 [0] (
10:22:02 Quit softi-42 (Read error: 110 (Connection timed out))
10:26:04doktoreaspetur, cool is exactly what i want :D
10:26:08doktoreaslike the amarok database
10:27:00 Quit Mouser_X3 (Read error: 110 (Connection timed out))
10:28:17doktoreaspetur, the database update will recognize also mp3 was inside ipod before updating to rockbox?
10:29:09GodEaterWebdoktoreas: yes it will
10:29:22 Quit Audioengineer (Read error: 145 (Connection timed out))
10:29:35doktoreasi have selected initialize database
10:29:44doktoreasit says updating in background
10:31:44doktoreasnow all is fine thx
10:38:41GodEaterWebdo you know - I've lost count now of the number of times I've hit the "next" link in the forums assuming it'll take me to the next page, and end up on a completely different post.
10:40:16 Part norbusan
10:43:17 Part Llorean
10:45:07 Join bluebrother [0] (i=DludPOhK@rockbox/staff/bluebrother)
10:50:06Techniim not sure if it's possible, but could you guys make a dual bootloader? maybe if we hold down on the touchpad itll load the backup copy of the original bootloader?
10:50:29Bagderit is dual-boot on most targets already
10:50:30 Quit GodEaterWeb ("CGI:IRC (EOF)")
10:51:09Techninot on the gigabeat to my less than informed knowledge
10:54:32BagderI would be surprised if they wouldn't want that to work as well
10:54:51 Quit pondlife ("disconnected has pondlife")
10:56:48 Join GodEaterWeb [0] (i=c2cbc95c@gateway/web/cgi-irc/
10:57:08Techniwould be nice, RBox isnt exactly as pretty/functional as the gigabeats OS
10:57:32GodEaterWebthat would be because it's not even close to being finished on the gigabeat
10:57:44peturgive it some time willya - it's in early stage
10:57:56GodEaterWebwhich one would assume you knew - given there's not even an official manual for it yet
10:58:31Bagderthere is a manual for it
10:58:58GodEaterWebbut that doesn't include the bootloader install yet does it ?
10:59:07Bagderno, I don't think it does
10:59:22BagderI haven't got any bootloader binary to host on the download server either
10:59:47GodEaterWebI think the only place to get the bootloader install guide is from the forums currently
10:59:57GodEaterWebthat says "not even close to being finished yet" in quite big letters to me
11:00:01Bagderwe need to get the core gigabeaters to join us here
11:00:23 Join MarcoPolo [0] (
11:02:09Techniyeah, being a homebrew developer for PSP I know how hard it is, well, I imagine it's harder for these guys since PSP is much more powerful and has a pretty robust OS.
11:02:32 Join tj2master [0] (i=a@
11:02:59Bagderyeah, getting Rockbox to a new target is a lot of hard work
11:04:17 Quit D_C (Read error: 54 (Connection reset by peer))
11:05:10***Saving seen data "./dancer.seen"
11:06:51 Join NuclearDog [0] (
11:08:28 Join funky [0] (n=repulse@unaffiliated/funky)
11:09:04GodEaterWebBagder, I'm just reading through some of the docs on the iPL pages in an effort to understand the hardware a bit more - any idea what "bic'ed" means in the context of hardware initialisation ?
11:10:30GodEaterWebI'll have a hunt to see if they have a glossary then :)
11:10:45Bagderit's not a term I use in my every day :-)
11:10:46amiconnI guess that relates to the arm 'bic' instruction
11:10:56Bagderyeah, must be
11:11:02amiconnBIC == 'bit clear', i.e. reverse AND
11:11:28 Join Rob2222 [0] (
11:12:52 Quit matsl (Read error: 110 (Connection timed out))
11:14:45linuxstbI guess it's the ARM bic (bit-clear?) instruction?
11:16:22GodEaterWebthere's an echo in there...
11:16:38peturcho in here...
11:16:45Bagderin here...
11:17:37GodEaterWebwow - you lot must be really bored :)
11:18:17peturshhh don't tell my boss
11:18:51 Join spiorf [0] (
11:19:02 Join adam1302 [0] (i=cbab4195@gateway/web/cgi-irc/
11:20:49BagderI think I have all the daily builds working properly again now
11:21:17peturah there was a problem?
11:21:26BagderI wiped the entire directory
11:21:41Bagderme not a very smart person ;-)
11:22:33Bagderso I rewrote the scripts
11:23:04Slasherithere was no backups?
11:23:20Bagderthere is, but during the holidays it would take quite some time to get the tapes
11:23:24bluebrotherhmm, the manual still isn't build daily?
11:24:06Bagderbluebrother: it should be, I'll investigate
11:26:47Bagderfound the reason
11:28:34bluebrotherbtw, I was thinking if it would be better to name the html output files more like the ouput folder, like rockbox-h100.html
11:28:43*amiconn likes tape libraries
11:28:48bluebrotherI guess changing this will break the daily builds, won't it?
11:28:49 Quit Rob222241 (Read error: 110 (Connection timed out))
11:29:06 Join dan_a [0] (
11:29:10Bagderbluebrother: well, I'd adjust to such a move easily
11:29:34bluebrotherok, I'll notify you when I find the time to do this :)
11:32:30BagderI'll also make sure there's a static URL for the latest PDF
11:36:47 Join Angry [0] (
11:39:37dan_aYay! I've got another MP3 player on the way from ebay :D
11:39:58Bagderdan_a: did you get the US one I sent you?
11:39:59adam1302what is it
11:40:21dan_aBagder: I don't know yet - I'm getting home tonight
11:41:11dan_aadam1302: A Samsung YH-820 - PortalPlayer based, so a port shouldn't be *too* difficult. And it means I can charge up my YH-920, and see if I can do a port to that, too
11:41:44dan_aI've also got a microSD card for my Sansa, so I'm trying to see if I can get that going
11:42:13*Bagder adds dan_a to the list of insane people ;-)
11:42:38bluebrotheroh, there is a list? ;-)
11:42:40 Quit GodEaterWeb ("CGI:IRC (EOF)")
11:43:18Bagderthe secret society of listing insane people maintains it
11:43:34*dan_a grins
11:43:39 Join ctaylorr [0] (
11:44:41 Quit ctaylorr (Client Quit)
11:45:04 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
11:45:12 Join matsl [0] (
11:47:04 Join GodEaterWeb [0] (i=c2cbc95c@gateway/web/cgi-irc/
11:47:17GodEaterWebisn't the Nano 2G based on a Samsung built chip ?
11:48:12dan_aGodEaterWeb: I believe so
11:48:38linuxstbYes, Apple/Samsung made a big announcement about new ipods using Samsung ARM chips.
11:48:46adam1302dan_a: sounds awsome
11:48:47GodEaterWebso why would they build devices based on portalplayer stuff if they had their own chip to use ?
11:48:55linuxstb(causing PortalPlayer's stock to plummet....)
11:48:57 Join yossa [0] (
11:51:12 Quit adam1302 ("CGI:IRC (EOF)")
11:54:38 Quit My_Sic (Read error: 104 (Connection reset by peer))
11:54:45 Join My_Sic [0] (
12:09:07 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
12:09:12 Join miepchen^schlaf [0] (
12:09:36Bagderfresh new set of daily manuals fixed
12:12:33Bagderfixed URLs for the PDFs now working
12:18:04 Quit Techni (Read error: 145 (Connection timed out))
12:18:18BagderGodEaterWeb: I think it is a matter of time
12:18:29Bagderthe PP ones were better earlier, and the samsung ones are more recent
12:18:57Bagderat least the PP targets all debuted long before we saw these samsung-based ones
12:19:40Bagderfood time
12:19:43linuxstbsafetydan: (regarding audio samplerate in the sim) - do whatever you think is easiest. If you make SDL resample, then you don't have to worry about the samplerates supported by the user's soundcard.
12:21:15dan_aDarn - MicroSD support for the Sansa is more complicated than just changing a GPIO output
12:21:43linuxstbWhen was anything easy?
12:25:58linuxstbdan_a: How should I be removing threads from the COP? Should the thread kill itself, or can I remove it from the main CPU?
12:27:08 Quit Mouser_X (Read error: 110 (Connection timed out))
12:28:16dan_alinuxstb: The safest way will be for the thread to kill itsself, but I'm pretty sure I implemented remove_thread_on_core(core, threadnum)
12:30:27dan_alinuxstb: Argh, no, there isn't a remove_thread_on_core, so at the moment a thread can only kill itsself
12:30:41linuxstbAh, seems I'm just using remove_thread(thread)....
12:30:49linuxstbWhich oddly seems to work...
12:31:11linuxstbBut could explain why it doesn't work a second time...
12:31:29*linuxstb should stop using ....
12:31:44amiconnThe thread ids are unique across both cores, aren't they?
12:32:30linuxstbIt's a pointer to a struct thread_entry, so yet.
12:32:57amiconnThen there should be no need for a separate remove_thread_on_core()
12:33:00dan_aThe thread numbers aren't unique, but thread entries will be
12:35:01linuxstbI'm assuming a remove_thread(NULL) call will never return?
12:37:31 Quit _jhMikeS_ (Nick collision from services.)
12:37:37 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
12:38:31safetydanlinuxstb, what's an easy way to test if pcm_set_frequency is working? Play an MPEG video file with !44100 Hz audio?
12:39:06linuxstbProbably try one of the plugins - e.g. Doom - that sets the samplerate to something other than 44.1KHz. mpegplayer doesn't do that.
12:39:08jhMikeSsafetydan: I have a test plugin for that
12:39:11linuxstbOr Rockboy.
12:39:34linuxstbBut you'll probably need to enable sound in the sim for those plugins - the pcm functions weren't exported in the past.
12:39:46safetydanah okay, I just assumed it was related to the mpegplayer as the sim change was comitted after you made some changes there
12:40:35linuxstbI exported all the pcm* functions in the plugin API so that mpegplayer would compile for the sim - I don't think there was a reason why they weren't. I then got the error that pcm_set_frequency() wasn't defined in the sim.
12:40:53 Join Tim_Zuiker [0] (
12:40:58jhMikeSMaybe I should add it to CVS but not as part of the SOURCES?
12:41:17jhMikeSsort of like test_fps and such
12:41:20linuxstbjhMikeS: Yes, it would help when fixing the ipod audio drivers to work with multiple samplerates.
12:41:46jhMikeSIt should automatically have a menu of all samplerates defined in the config
12:41:52jhMikeSI'll do that then
12:42:05 Join trypt0 [0] (
12:42:39Tim_ZuikerWhy doesn't RockBox support Creative players?
12:42:54linuxstbBecause no owners of Creative players have ported Rockbox to them.
12:43:49 Quit _Veseliq_ (Read error: 110 (Connection timed out))
12:44:35Tim_ZuikerSo there is now other Firmware for Creative, then there original Firmware from Creative?
12:45:10linuxstbNot that I know of.
12:46:07Tim_ZuikerDamn, I'm missing some functions that I'd like to use, but I'm stuck with there original firmware :S
12:47:33 Quit Rob2222 ()
12:48:23Tim_ZuikerI've a litte experience in C++, but I don't know where to start if I want to make it
12:48:45 Quit GodEaterWeb ("CGI:IRC (Ping timeout)")
12:50:44safetydanwell I'll have to pick this up tomorrow
12:50:45 Quit safetydan ("Ex-Chat")
12:51:28 Quit rkostynu ("Chatzilla 0.9.77 [Firefox]")
12:54:26 Join lee-qid [0] (
12:54:35 Join _Veseliq_ [0] (n=veseliq@
12:58:20 Part Tim_Zuiker
12:59:30 Join nls_web [0] (i=d9d1bcda@gateway/web/cgi-irc/
13:02:17 Nick Everybody|away is now known as Everybody (
13:03:10 Quit linuxboy (Read error: 110 (Connection timed out))
13:05:13***Saving seen data "./dancer.seen"
13:08:41 Join GodEaterWeb [0] (i=c2cbc95c@gateway/web/cgi-irc/
13:19:44 Part tj2master
13:20:25 Join jba [0] (
13:21:04Bagdergwhite added as new cvs committer
13:21:21linuxstbWho is gwhite?
13:21:29BagderGreg White, gigabeat hacker
13:21:38peturcheck your mails ;)
13:21:57linuxstbI can't atm...
13:22:37 Join Mouser_X3 [0] (
13:22:51 Nick Mouser_X3 is now known as Mouser_X (
13:23:18linuxstbdan_a: Changing mpegplayer to use remove_thread(NULL) works well. I can now repeatedly start mpegplayer on my Photo, and the video thread seems to start reliably.
13:23:34amiconnBagder: Do you think the // comment in the config-*.h files are still bad in spite of the special case?
13:23:49linuxstbI prefer them in that situation.
13:24:12*amiconn wondered about Bagder shortly popping in for just telling that
13:24:23linuxstbYes, I saw that
13:25:00 Quit markun (Read error: 104 (Connection reset by peer))
13:25:07 Join markun [0] (n=markun@rockbox/developer/markun)
13:25:27linuxstbdan_a: Forget that, it just failed to start...
13:25:32 Join phrozen77 [0] (
13:28:46linuxstbdan_a: That seems to have been a one-off failure... It hasn't failed since then.
13:29:47 Quit trypt0 ("Leaving")
13:33:45 Join Rob2222 [0] (
13:36:23 Join Ricky [0] (i=cbad3b96@gateway/web/cgi-irc/
13:36:34Rickyhello all:)
13:36:50Rickyi would like some help with rockbox is anyone could help
13:37:26 Join aliask [0] (
13:38:04nls_webRicky: what's your problem?
13:38:47Rickywould you be able to help me install a theme
13:39:32Rickyi have the theme now there a four seperate folder 1.fonts - 2.themes 3.backdrops and 4.wps
13:40:24nls_webyes, in your .rockbox directory there are folders with the same names
13:40:44nls_webput files from fonts into your fonts directory etc
13:41:12*petur starts reading the darwin awards :)
13:41:19Rickyok thanks for you help
13:41:30 Quit _Veseliq_ (Remote closed the connection)
13:41:49 Quit Febs (Read error: 145 (Connection timed out))
13:42:13 Join _Veseliq_ [0] (n=veseliq@
13:42:55 Quit Ricky ("CGI:IRC (EOF)")
13:48:10 Join fleytin [0] (
13:48:29 Join tendersmile [0] (
13:48:56 Quit tendersmile (Client Quit)
13:50:49 Part fleytin ("ERC Version 5.1.4 (IRC client for Emacs)")
13:52:18 Join Ricky [0] (i=cbad3b96@gateway/web/cgi-irc/
13:52:35Rickycan rockboz play videos?? if it how how can i do this?
13:52:49 Join Mouser_X3 [0] (
13:54:42 Join Wiwie [0] (
13:58:50 Quit jba (Read error: 110 (Connection timed out))
13:58:58 Quit Wiwie (Client Quit)
13:59:06 Join Wiwie [0] (
14:01:13 Join pondlife [0] (n=Miranda@rockbox/developer/pondlife)
14:01:22 Nick Everybody is now known as Everybody|away (
14:04:24 Quit Ricky ("CGI:IRC (EOF)")
14:04:37 Quit jhMikeS (Nick collision from services.)
14:04:43 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
14:05:44 Quit _Veseliq_ (Remote closed the connection)
14:06:17pondlifejhMikeS: Hi, any further plans on the playback stuff?
14:10:40 Quit Wiwie ("Nettalk6 -")
14:11:01 Join Wiwie [0] (
14:13:01 Quit Mouser_X (Read error: 110 (Connection timed out))
14:15:02 Quit GodEaterWeb ("CGI:IRC")
14:15:09 Join GodEaterWeb [0] (i=c2cbc95c@gateway/web/cgi-irc/
14:15:40 Quit dan_a (Read error: 110 (Connection timed out))
14:16:25jhMikeSpondlife: big time. first stage: organize it :)
14:17:09pondlifePerhaps we need a Wiki to get it all documented before any re-coding starts? Aims: MAS/SW unification, simplification, new metadata handling, video integration etc...?
14:17:33pondlifei.e. set the scope too
14:19:40pondlifeIt was the "non-buffering video audio", and a crash I had this morning that got me going ;)
14:19:48jhMikeSI'll unify anything that can be done. Playback thought UI will be done throught the playlist interface alone and a control callback to feed it tracks. Metadata handling: definitely a 2nd iteration thing. Video: could be at any point but would like to complete the audio stuff first. Simplification: mostly from the heirarchy of UI->playlist->playback->pcm buffer->pcm
14:20:23jhMikeSWell also gain 512KB of ram since ther'll be only one swap buffer
14:20:47pondlifeHopefully we can reduce the number of interfaces too. Lots more routines to be made static...
14:21:37 Quit aliask (Read error: 104 (Connection reset by peer))
14:21:50*linuxstb likes it when two lines of code solve a big issue
14:21:59jhMikeSdoing that. playback.c will have no awareness of directories and multiple callbacks can be registered for important events like buffering and track switching (things that scrobbler needs for instance)
14:22:06linuxstbA/V sync is now very close in mpegplayer.
14:22:59jhMikeSlinuxstb: should frames be time stamped for display maybe?
14:23:38preglowjhMikeS: wouldn't it be better to synthesise the tones so that the pitch stays the same?
14:23:54linuxstbI just count the number of frames displayed, and keep track of how long each frame should be displayed for (a constant in most files - i.e. the framerate).
14:24:13jhMikeSpreglow: I suppose but that's what I had to work with and needed a quick checker when doing the cf stuff
14:24:17 Quit matsl (Read error: 110 (Connection timed out))
14:26:03AngryIs there any updated mpegplayer.c outta there??? Cant wait to see^^
14:26:03jhMikeSlinuxstb: I'm thinking decoding should happen or something right after the frame is drawn, just wait till the proper time to draw it. Just an idea. Then you can drop them if behind.
14:26:29linuxstbmpegplayer has always done that.
14:27:05linuxstbIt skips the display of frames, but doesn't skip the decoding (but that might be possible to some extent).
14:27:40preglowshould be, i think
14:28:00preglowthat is, if you mean the actual resynthesesing
14:28:05preglowyou always have to decode, i think
14:28:10jhMikeSlinuxstb: how do you keep from accumulating errors then?
14:28:22linuxstbWhat errors?
14:28:32jhMikeSwhat if you only decode I frames when things get bad?
14:29:44jhMikeSlinuxstb: It seems like showing the frame for a certain amount of time but not using it's absolute time in the stream would accumulate small errors. I'm not perfectly clear on what you're doing though. :\
14:30:17 Quit yossa ("Chatzilla 0.9.77 [Firefox]")
14:31:25linuxstbI just keep count of the number of frames decoded, and calculated the ETA for that frame based on the framerate.
14:32:05linuxstbAnd I also keep count of the number of audio samples played through the DAC, which I now use to keep the master clock in sync.
14:32:48 Join Kittt0s [0] (i=Kaa@
14:32:49jhMikeSso you compensate it with that? The DACs gonna be the best clock, not ticks for sure :)
14:32:54 Quit Kittt0s (Client Quit)
14:33:16 Join dan_a [0] (
14:33:45linuxstbI have a "tick_offset" variable (updated in the audio get_more() function), which I subtract from current_tick to get the elapsed time.
14:34:05linuxstbAnd that elapsed time is used to decide when to show video frames.
14:35:22linuxstbSo basically, video is synced to audio.
14:35:27jhMikeShmmm...recording doesn't even use ticks at all for ET. Just the number of samples and samplerate.
14:35:46dan_alinuxstb: I'm going to do some serious work on the dual core support when I get home - the reports I'm hearing seem to be that it's a bit unstable with the way we are doing things at the minute
14:36:06linuxstbBut I need to display more than one video frame during the time of one 32KB DMA chunk.
14:36:07 Quit jaebird ("Leaving")
14:36:13 Join matsl [0] (
14:36:20 Join jaebird [0] (
14:36:47 Join Febs [0] (n=chatzill@
14:37:51linuxstbdan_a: Yes, I'm still getting problems with the video thread not always starting. That could be my fault as well though, but at least it seems much better now I'm correctly removing the thread.
14:39:27 Join jba [0] (
14:39:53linuxstbdan_a: Do you have a target capable of running mpegplayer?
14:40:01linuxstb(a dual-core target)
14:41:03dan_alinuxstb: No, not at the moment (no sound on Sansa, no graphics on the 3G) I don't think I can borrow one from anywhere either.
14:42:20jhMikeSlinuxstb: isn't pcm_get_bytes_waiting available? should be enough I suppose.
14:42:47dan_alinuxstb: Do things work better if you remove the "COP_CTL = PROC_SLEEP" from thread.c?
14:43:25linuxstbjhMikeS: I didn't know about that... That tells you how much of the 32KB has not yet been played?
14:44:04linuxstbdan_a: I'll try it.
14:44:55jhMikeSlinuxstb: yes
14:45:27linuxstbThat should be better, I can then ignore current_tick completely...
14:56:59 Quit My_Sic ("Miranda IM! Smaller, Faster, Easier.")
14:58:54 Quit jba (Read error: 110 (Connection timed out))
15:00:00 Quit XavierGr (Read error: 113 (No route to host))
15:01:12jhMikeSyeah .01s is a bit gritty with frames lasting about .033s :)
15:02:52amiconnrvf playback uses a really simple synchronisation which works surprisingly well
15:03:55amiconnNothing complex like separate threads, time stamps etc involved. Just playing the video frames based on the user timer, and feeding the audio to the mas
15:04:02linuxstbdan_a: It seems worse without it.
15:04:36linuxstbBut in a way the mas can be thought of as a separate thread.
15:04:55linuxstbIs the audio feeding in a DMA interrupt though?
15:05:09dan_alinuxstb: Really? I wouldn't have expected that!
15:05:19***Saving seen data "./dancer.seen"
15:05:23amiconnThe only reason for time drift on rercorders would be slightly off-standard clocks for cpu and/or mas
15:05:30amiconn(both have separate xtals)
15:06:14linuxstbCan the MAS tell you how many samples it's played?
15:06:38linuxstbSo you would have to decode the mpeg audio frames yourself?
15:06:41amiconnIt can only tell how many mpeg frames it has played, but in a very unhelpful manner
15:06:58amiconnNo, we just feed the compressed frames to the mas
15:07:34amiconnThe mas frame counter is signed 20 bit (the standard internal data format of the mas) - and it saturates (!) at 0x7ffff
15:08:44linuxstbdan_a: Well, it's hard to judge. There seems to be a little randomness built-in to the PP5020....
15:10:19amiconnlinuxstb: (1) the SH1 has way too little power to decode mpeg audio even near realtime (2) Why do you think this would be necessary?
15:11:08linuxstbSorry, I didn't mean fully decode, just parse the stream to keep track of the number of samples played.
15:11:48linuxstb... and then keep the video in sync that way.
15:12:33 Nick Everybody|away is now known as Everybody (
15:15:20*linuxstb wonders what the difference is between set_sound() and sound_set() in the plugin API...
15:15:49amiconnWe don't keep track of anything during normal playback. We just start and let it run
15:16:42amiconnIt doesn't run apart too far even when running for hours (unless there is a bug, which was the case in the earlier stages)
15:17:10linuxstbI know, but you said there was a small time drift due to off-standard clocks. But it sounds like it's good enough.
15:17:32jhMikeSpreglow: is fsincos up for doing a nice synth of a sine wave for constant pitch as different samplerates? The other waveforms are just straight lines so easy with line drawing.
15:17:35amiconnIt doesn't need to be sample exact; for lip-sync appearance it's sufficient to keep it within +/- 50msec - and that's nearly 2 mpeg audio frames
15:18:10*jhMikeS like high precision if you can get it :)
15:19:10amiconnOn Ondio there's another source of error. RVF specifies the video frame duration in mutliples of the recorder's CPU clock (11.0592 MHz), but the Ondio runs at 12.00 MHz
15:19:53amiconnWe just calculate the new frame duration using the ratio of 625/576
15:20:19linuxstbSo errors accumulate?
15:20:52amiconnYes, but even that doesn't become visible during a standard movie
15:21:13*jhMikeS likes rock solid sync that never drifts :)
15:21:24 Quit gantrixx ("Ex-Chat")
15:21:37*amiconn prefers simplicity if complexity isn't really necessary
15:21:40*jhMikeS is a picky bastard though
15:22:13linuxstbHow long does the movie need to be for a/v sync to drift?
15:22:52amiconnI never found noticeable drift on recorder, even with lord of the rings (~3 hours)
15:23:05amiconnOn Ondio, I didn't try lord of the rings yet
15:23:29amiconn(reqires at least an 1GB MMC - part 1 is around 850MB)
15:24:36nls_webwow a 3 hour movie on a monochrome 128x96... "fun" ;-)
15:25:58jhMikeShow will keeping the frames tightly synced to audio make anything really complicated? doesn't seem to lead to that if no ticks are involved. just schedule a frame and wait for the proper time to display it. keep track of lateness and compensate. hmmm
15:26:21 Join dpro [0] (
15:27:20 Part Angry
15:27:35pixelmanls_web: the Archos displays are 112x64... but you should first take a look and then decide how much fun it is (for my tastes it's better than I'd imagined) ;)
15:29:01jhMikeSyou might notice that recording has none of the "estimates" as it did previously...but that's the way I do stuff. It was actually simpler than all the estimation.
15:30:10 Join Arathis [0] (
15:30:34amiconnjhMikeS: That means you need to keep track of time a fine-grained manner
15:31:24amiconnOf course video.rock keeps track of time, in order to support seeking, played / remaining time etc
15:31:44amiconnThen, rvf playback is very different from mpeg video
15:32:32amiconnNo cpu intensive decoding, just feeding the frames to the lcd (which still causes ~50% cpu load on archos, due to the serial lcd i/f)
15:33:03amiconnOtoh no frame should be skipped or doubled there, because it would cause a visible glitch
15:33:09 Quit matsl (Read error: 60 (Operation timed out))
15:33:46amiconn(rvf uses the same temporal dithering method as the grayscale lib. In fact video.rock was the first using this method, and I developed the grayscale lib working from that)
15:34:40jhMikeSamiconn: yes, using the samples and sample rate
15:35:41amiconnWith rvf rockbox cannot count samples
15:35:51amiconn(on hwcodec)
15:36:18jhMikeSif the encoder rate doesn't match the hardware rate (possible in spdif) the clock will move slower or faster since it gives the playback time of what's recorded.
15:37:12jhMikeSon SWCODEC using the sample clock is easy with the interrupt handler and pcm_get_bytes_waiting
15:40:22jhMikeScan't really imagine watching something too much longer than a music video on a little screen though. LOTR is out of the question :)
15:40:27 Join lowlight [0] (i=c730180a@gateway/web/cgi-irc/
15:40:51 Join BigBambi [0] (
15:41:29 Join matsl [0] (
15:44:08jhMikeSamiconn: Is RVF a format you made for Archos and video.rock?
15:44:29amiconn[IDC]Dragon made it, to be precise
15:45:02jhMikeScompressed video?
15:45:54jhMikeSthough as I understand there's not a lot of CPU there to do that :\
15:46:00amiconnIt's basically video frames in native LCD resolution and frequency (112x64x1, 67Hz on archos), interleaved with audio blocks containing bit-swapped mpeg audio data (mp2 or mp3, all standard sample rates)
15:46:04 Quit Wiwie ("Miranda IM! Smaller, Faster, Easier.")
15:52:19*jhMikeS has been pondering a possible way to remove the "wave" on the screen in the grayscale overlay. not sure how to pull it off yet.
15:55:52 Join linuxboy [0] (
15:55:58amiconnjhMikeS: You can't
15:56:41amiconnThe 'wave' comes from the fact that the frame rate we're writing at doesn't match the lcd's internal frame rate exactly
15:56:59jhMikeSI'm not sure why not. Remove components of the noise at 1/n intervals of the framerate (widely notched) when n = an integer from 1 -> inf theoretically.
15:57:40amiconn...and that match can't be made exact because (1) the lcd uses an RC oscillator with a fairly high temperature drift and (2) the lcd doesn't tell the scan position
15:58:18jhMikeSthat's why the pixel switches should occur at times not related to the lcd refresh rate
15:58:30amiconnThe temporal pattern for each pixel isn't random, as that would cause far too much graininess
15:59:12jhMikeSuse blue or purple noise in the spatial domain though. that's the plan anyway.
15:59:15amiconnOnly the start position within the pattern is chosen at random whenever the pixel is changed. It stays fixed as long as the pixel doesn't change
16:00:52 Quit Mouser_X3 (Read error: 60 (Operation timed out))
16:01:21amiconnYou can't switch single pixels at arbitrary times, only the complete frame at once (technically the smallest possible unit for switching would be one pixel block, but with way more overhead than switching frames)
16:02:34jhMikeSare you saying a screen with only a solid (undithered) color will exhibit this wave as well?
16:02:43preglowjhMikeS: sure
16:03:14jhMikeSpreglow: "sure" to the fsincos thing?
16:03:18amiconnBut the only solid colours are pure black and pure white
16:03:21preglowjhMikeS: yea
16:03:36preglowjhMikeS: just use a straight unsigned long phase accumulator as argument, the increment should be 0xffffffff/samplerate*freq
16:03:43jhMikeSamiconn: but on H120 they're not
16:03:50amiconnthey are
16:04:19amiconnYou can't mix programmed temporal dithering with any of the lcd's greylevels
16:04:41preglowjhMikeS: just make sure to use fairly low frequencies for the other waveforms, or frequencies which fit a whole number of samples in each period, to reduce aliasing artifacts
16:04:56amiconnThe lcd uses temporal dithering itself; in combination with programmed temporal dithering it flickers like hell
16:05:26preglowfairly low frequencies are most suited to detecting sample rate error artifacts anyway
16:06:18jhMikeSpreglow: ok, this notation of 0xffffffff/something means what exactly? fixed point for 1.0/something?
16:07:16jhMikeSamiconn: ok, to get the grays on the H120 display as well then except black and white?
16:07:16 Quit spiorf (Remote closed the connection)
16:07:25preglowjhMikeS: 0xffffffff represents something very close to 2pi, so 0xffffffff/sfreq would give you one full sine cycle using sfreq samples, multiplying it by freq will just scale the frequency up from 1 to whatever you want
16:07:44preglowi guess it should be calculates as 0x100000000/sfreq*freq
16:08:44 Quit doktoreas ("Leaving")
16:09:10jhMikeSpreglow: ok, you're doing radians with that. got it
16:09:40 Quit Everybody (Read error: 110 (Connection timed out))
16:11:40preglowone way of seeing it, yes
16:12:09preglowit's not really radians, since 2*pi = 0x100000000, but the principle is the same
16:13:43 Quit MarcoPolo (Remote closed the connection)
16:15:12jhMikeSthen any fraction of that is a fraction of 2*pi. your giving me angular velocity (w) if I read correctly
16:19:36 Quit dan_a (Read error: 110 (Connection timed out))
16:25:04Juice^anyone here tried to disassemble sansa's mi4 in IDA?
16:25:23 Join debauched_sloth [0] (n=gwhite@
16:26:14 Join _Veseliq_ [0] (n=veseliq@
16:29:37BagderJuice^: ida and other tools, sure
16:30:30jhMikeSamiconn: was just saying with 1/n*(refresh/2) frequencies and near neighbors filtered out of the what I presume is white noise, pixels won't switch in near time with the refresh rate since those components won't exist. The whole frame would still be updated in bulk. I'll just get to proving my point or feeling like a schmuck when I fail ;) I'll still learn stuff nonetheless.
16:32:12 Join Arathis_ [0] (
16:32:23amiconnThe temporal pattern for each individual pixel is not white noise, it's deterministic
16:33:09amiconnAnd leaving out the high frequencies is undesirable, as it would increase flickering
16:33:17 Join Angry [0] (
16:33:59amiconnThe only thing which is random is the starting point within the deterministic pattern
16:34:21 Join netmasta10bt [0] (
16:34:21amiconn(pseudo-random, of course)
16:34:25jhMikeSI wouldn't leave out high frequencies, in fact, I want them just certain frequencies
16:35:51amiconnOne part of the wave is also caused by the fact that we sometimes write to the part of the lcd's internal buffer where the lcd controller is reading just at the same time
16:36:13amiconnThat effect can't be suppressed at all
16:36:50jhMikeSthat happens on the x5 too
16:37:09jhMikeSjust with normal screen updates I can see a change
16:37:16jhMikeSnot with grayscale of course
16:37:33amiconnYes, but that's not 70 times per second under normal circumstances
16:37:57amiconnThe epson lcd controller reacts with a brief switching to white in that case, the one in the archos produces small black 'fishes'
16:38:13 Join tj2master [0] (i=a@
16:43:00jhMikeSthe effect of updates on x5 isn't too bad. Just a slightly dark band that moves horitontally if near the refresh rate.
16:44:49*amiconn wonders whether it would be possible to calculate those frames 'live' using simple prng
16:45:05amiconnThen the grayscale lib would no longer be limited to 33 shades, and use less memory
16:45:21jhMikeSan adjustable gamma for grayscale would be nice though since my display is quite light and dark switches to light rather suddenly
16:45:39amiconnThere already is gamma correction
16:46:09 Join bluey- [0] (
16:46:26 Quit Arathis (Read error: 110 (Connection timed out))
16:46:27jhMikeSyes, but I can't set it to my liking. very light/dark with no real grays
16:46:35amiconnIt's in fact more complicated than just gamma, the curve is a superposition of standard gamma with an S-shaped transfer function
16:47:27amiconnThe problem is that the transfer function depends on the 'contrast' setting, the individual lcd _and_ temperature...
16:47:28Juice^Bagder: ok nice. what setting did you use to get it readable? i tried binary/raw and speficied it as a ARM710a cpu witch was the closest....
16:47:31jhMikeSI would say it appears quite s-shaped with a sudden ramp up from dark to light towards the middle
16:47:37Juice^Bagder: using IDA
16:48:13amiconnjhMikeS: The s-shape is compensated for in the library...
16:48:28BagderJuice^: it is ARM7 so that works
16:48:59amiconnI guess you need to specify arm720t or another 't' model because plain arm7 doesn't support thumb
16:49:13Bagderbut most of the mi4 is not thumb, iirc
16:50:05jhMikeSMe eyes don't deceive me :) Perhaps a plugin to adjust it to perfection on your own player for the finicky like me is in order. I guess I'd need a photo to show what I mean though.
16:50:09Juice^there isnt any other arm7 cpus to choose of
16:52:28BagderI recall dealing with thumb with IDA anyway
16:52:42Bagdernot that I achieved much with it, but still
16:59:03jhMikeSamiconn: the gradients in grayscale.rock are just the 33 shades of gray right? The lower 25% looks pretty much solid black and the upper 25% or so pretty much white some grayness in the mid 50%. I must have a freakish player then. :)
16:59:57 Join RedBreva [0] (
17:01:56lowlightJuice^: ARM works fine. It will just load the binary though, you need to find the entry point though.
17:03:25lowlightand sometimes you need to do the switch to thumb mode
17:03:29amiconnjhMikeS: Yes, the gradients in grayscale.rock are uncompensated
17:03:49amiconnThe library allows you to choose between raw and compensated mapping
17:04:18 Quit Kitt0s (Read error: 110 (Connection timed out))
17:04:22amiconnJPEG, doom etc use compensation, grayscale.rock and mandelbrot don't
17:04:28*jhMikeS notices that navi is not the fire button in doom but rec is. hrm.
17:05:24***Saving seen data "./dancer.seen"
17:05:25jhMikeSwhat's needed in grayscale.rock to use compensation? would like to see the test screen with it.
17:05:55 Quit GodEaterWeb ("CGI:IRC")
17:07:15 Join barrywardell [0] (n=barrywar@
17:07:36amiconnYou need to change 'gamma' in the gray_init() call (second last parameter) to something else than 0
17:08:14amiconnThis parameter is fixed point with 8 bits fractional part
17:08:52 Join spiorf [0] (
17:08:54amiconn0x100 will give you pure linear mapping, 0x200 will sort-of resemble standard computer displays
17:10:37amiconnThe latter is also what jpeg.rock uses (2 << 8)
17:13:49jhMikeSwhat about adjusting the amount of "S"? this gives no control over that. wish I could just set a linear gradient globally and force it on everything that uses the lib.
17:14:28amiconnNo, the 's' compensation is done using a table
17:19:03amiconnThe option to choose whether to compensate or not is for use with low depths
17:19:26amiconnmandelbrot uses 8 shades, and cube (on archos) uses 4 shades
17:19:28jhMikeS0x200 is way too light
17:19:59amiconnThe compensation curve was taken with contrast set to 28
17:20:32amiconn27 is _still_ too light... but with 28 your "fixed" greylevels are too dark...
17:21:41jhMikeSWell, I said I'd revert that and do a better fix cause it's about perfect on mine and preglow's but obviously isn't going over well everywhere.
17:22:51 Nick maquis is now known as ack (n=erins@
17:23:09 Nick ack is now known as maquis (n=erins@
17:23:50jhMikeSYou said you're not using any of the gray levels anyway in the lib so how could it affect that?
17:25:32amiconnThe contrast setting influences all greylevels
17:26:01amiconnBlack, native greylevels, and software-pwm greylevels
17:26:14 Quit tj2master (Read error: 110 (Connection timed out))
17:26:25 Part Angry
17:32:13jhMikeSEven back at 28 with the pattern as it is now it's ok. Just to the other side of correct. 28 still looks essentially the same as 27 for grayscale.rock
17:34:35 Quit lowlight ("CGI:IRC (EOF)")
17:37:06 Quit bluebrother ("leaving")
17:38:23 Join mako_ [0] (
17:42:27 Join lowlight [0] (i=c730190a@gateway/web/cgi-irc/
17:43:31 Quit barrywardell ()
17:43:33 Join Ribs [0] (n=ribs@
17:47:06 Join hcs [0] (n=hcs@rockbox/contributor/hcs)
17:50:25 Quit BigBambi ("Leaving")
17:51:20 Join Mouser_X3 [0] (
17:55:00 Quit mako (Read error: 110 (Connection timed out))
17:55:39hcsthere's a gigabeat MEGF40S on, can I presume by the F in the name that rockbox will be usable on it?
17:58:13peturtricky because of the S at the back - better learn more about the naming
17:58:21amiconnYes, afaik this type is an gigabeat F, 40GB, Silver
17:58:38amiconnThe gigabeat S has the S before the capacity
17:58:40hcsit is indeed silver
17:59:18hcsnow I just have to find someone to buy my ipod :)
18:00:26hcsis its better support due to it having more open hardware?
18:00:54Ajaxincits more powerful also
18:00:54 Quit pondlife ("disconnected has pondlife")
18:01:08Ajaxincand my F20 is super durable
18:01:19Ajaxincwhile my girlfriend has broken 3 ipods
18:01:33hcsI've been happy with the durability of my ipod, 60 GB photo, built like a brick
18:03:00lowlightjhMikeS: for playback reworking...have you considered Slasheri's idea of a separate for flexible buffer api & tread to handle buffer operations instead of mixing them with playback
18:03:11hcshmm, how's the sansa port coming along? I see it isn't on the chart yet but I haven't been paying attention
18:04:21hcsbest buy has one for $90
18:04:26jhMikeSlowlight: not sure what that means exactly. you mean for the pcm buffer?
18:06:37 Quit petur ("plop")
18:08:40 Join Kitt0s [0] (i=Kaa@
18:08:45pixelmahcs: the rockbox sansa port deals with the e-series
18:09:07 Quit bluey- ("This computer has gone to sleep")
18:09:15hcspixelma: doh, these characters look too similar, thanks
18:10:04lowlightjhMikeS: no the "filebuf" in playback.c. It would be useful to have a set of functions to read/write/fill/etc from the buffer for audio, video, & metadata
18:12:19hcsok, how about the e250, then, which they have for $109?
18:12:35hcse250R, in case that's important
18:12:44jdonghcs: from what I can see on the wiki page, far from being a usable audio player :)
18:12:47jhMikeSlowlight: ok...sounds like a good idea as long as overhead doesn't grow. metadata on the buffer is tricky in some ways since it gets destroyed (say in a big file that wraps). If it's needed later it must be offloaded somewhere.
18:15:02hcsjdong: ah, yes, "no sound"
18:15:31jhMikeSlowlight: right now I'm just combing through to not have it hard coded to playlists and such. anything should be able to serve as a track source really and that can worry about the details of what content should be played. cleaning that up will ease other changes.
18:15:51Ajaxinccurrently I'd say the gigabeat has the most power and is one of the players being worked on the most
18:16:04jdonghcs: right :)
18:16:15pixelmahcs: I think I read that the "R" (so-called Rhapsody) players aren't supported at all - you could check out the forum thread (as that's where my info comes from too ;) )
18:16:21jdonghcs: but I too greatly anticipate the sansa e200 port :)
18:16:40 Quit matsl ("Leaving")
18:16:41hcsif I do get a new one it'll be because it is better supported than the ipod, not less, so the gigabeat it'll likely be
18:17:19hcsit's just that people got me gift cards to Best Buy for Christmas...
18:17:30jdongAjaxinc: would it be the most usable port at the moment?
18:17:35jdongor the one that holds the most future hope
18:17:41jhMikeSI mean, is the buffering handled in mpeg.c in a similar manner or completely differently? Haven't really looked much.
18:19:08 Join Criamos [0] (
18:21:26Ajaxincjdong: holds the most future hope I would say
18:21:35Ajaxincand it is currently very usable
18:21:46Ajaxincoff to school, peace peeps
18:23:12lowlightjhMikeS: can't say I know anything about mpeg.c
18:23:21 Join bluey- [0] (
18:26:52 Join Angry [0] (
18:27:35 Join gotthardt [0] (
18:32:51linuxstbAm I right in thinking that remove_thread() doesn't work in the sim?
18:34:00*linuxstb is happy to see pcm_get_bytes_waiting() is implemented in the sim
18:36:52 Join spiorf_ [0] (
18:38:57 Quit blue_lizard (Read error: 60 (Operation timed out))
18:39:31 Join blue_lizard [0] (
18:40:10jhMikeSThinking about it, I can't say atm how much awareness of content a buffer api should have if any. So much of playback.c is about buffering and knowing just exactly what and where things are written. Will think hard. Maybe a flash of insight...maybe smoke from the ears :)
18:41:29 Quit Mouser_X3 (Read error: 110 (Connection timed out))
18:42:40 Quit Ribs ("eh eh ehhhh!")
18:42:48linuxstbWhat about having codec-specific loaders?
18:42:55 Quit spiorf (Read error: 60 (Operation timed out))
18:44:41jhMikeSlinuxstb: in the codec?
18:44:55linuxstbNo, in the core...
18:45:06linuxstbOr at least, loadable loaders...
18:45:14 Join mathgl [0] (
18:45:42 Join GodEater [0] ( bout relocatable executable code on the buffer to do it?
18:46:58amiconnjhMikeS: Buffering in mpeg.c is much simpler than in playback.c
18:47:01linuxstbThis has been talked about in the past for the non-streaming codecs (MOD, midi etc) to allow more flexible use of memory.
18:47:12amiconnThe one in mpeg.c I can understand even though I didn't write it...
18:47:16linuxstbBut could also open up things like supporting any codec in any container.
18:47:43hcsmaybe useful for stuff like ADX where you know there will be seeking back in the file (for looping)
18:48:06hcsright now I have it force loading the whole file at once
18:48:11jhMikeSamiconn: and it does all the seeking stuff that playback.c does?
18:48:11amiconnI still think it would be easier to write a new engine from scratch than fixing the one in playback.c to work with metadata on buffer, support hwcodec etc
18:48:32amiconnWhat do you mean with all the seeking
18:48:39amiconnIt does handle seeking of course
18:49:12jhMikeSwell, same basic operations and multiple tracks right?
18:49:33amiconnFrom a functional view, yes
18:49:40amiconnFrom implementation, I don't know
18:49:51amiconnI tried to follow the playback.c logic once, and failed
18:49:59linuxstbMy only other thought about the playback engine is that a "push" system may be simpler - i.e. the playback engine calling a decode_frame(inputbuffer,outputbuffer) function to decode the next frame of data. Seeking would be done by the file loaders, rather than the codec itself.
18:50:11hcsamiconn: yeah, but if you read the file as more than one buffer it may have to rebuffer the first bit if you seek back (the buffer is discarded when you read a new one), and most ADX files always repeat
18:50:55amiconnThe hwcodec engine always rebuffers when seeking backwards
18:51:19amiconn...even if the part seeked into is not yet flushed
18:51:39hcsthus the forced single buffer loading I do to avoid rebuffering
18:52:30jhMikeSI know what plaback.c is up to. It's just that no comment is given consistently on the WHY of it's implementation.
18:52:48preglowamiconn: sounds highly annoying
18:52:54amiconnpreglow: why?
18:53:00preglowwhy do you think?
18:53:03preglowa rebuffer makes me wait
18:53:13amiconnyes, sure
18:53:31preglowkiss isn't always the way to go
18:53:32amiconnBut keeping track of what is not yet overwritten means extra work, and extra code
18:53:38jhMikeSI don't think it's need to essentially throw away buffered data to seek backwards. You just have to know what still exists and if it's gone, reload.
18:53:43amiconn...and seeking backwards is a rare operation
18:54:05linuxstbWhat does the swcodec engine do?
18:54:17hcsthus why I brought this up in the custom loader discussion, with some files it may not be as rare as all that
18:54:18*linuxstb can't remember ever seeking backwards...
18:54:19jhMikeSI'm thinking of a virtual, essentially infinite buffer whose location never wrap but that needs some big integers
18:54:26preglowswcodec works as it should, most of the time
18:54:33preglownot rebuffering if the data is still there
18:54:41amiconnAnyway, chances are quite high that the requested part is indeed already gone on hwcodec... with only 2MB of RAM
18:54:48preglowtrue enough
18:55:25jhMikeSThe real available interval of RAM would be a window.
18:55:37jhMikeSAnything outside the window is lost
18:56:00lowlightlinuxstb: I agree with the "push" idea if you split out the buffering part
18:56:02jhMikeSor doesn't exist yet.
18:56:05amiconnhcs: Non-streaming codecs are a complete different beast. With streaming codecs, seeking (either forward or backward) only happens when triggered manually
18:56:30hcsamiconn: There are exceptions, and one of them is in rockbox
18:56:37jhMikeSWould 64 bit integers be enough? I don't think most of the math would need it, just the virtual addresses
18:56:41hcsADX is streaming but looping
18:56:55amiconnhcs: Please read again.
18:56:57lowlightlinuxstb: a push system would probably make video sync easier
18:57:15amiconnHmm, I would define ADX as non-streaming if it loops
18:57:42amiconnjhMikeS: 32 bit ints for the position are enough
18:57:50amiconnMore isn't supported by fat32 anyway
18:58:28 Quit spiorf_ ("Read error: 104 (Connection reset by peer)")
18:59:06 Join spiorf [0] (
19:00:09 Nick _pill is now known as pill (
19:02:20jhMikeSamiconn: they'd only be for a vitual window that moves only forward as playback progresses, all loading would always occur at an ever higher address and the window slid over by that amount but be translated to real addresses when given to a codec or whatever.
19:03:40jhMikeSas long as say, the ADX file is entirely in the window, you just just jump the cp back. no aliasing problems either
19:05:26amiconnI don't think such a simple approach would work
19:05:28***Saving seen data "./dancer.seen"
19:05:40amiconn...because it doesn't take buffer wrap into account
19:06:15amiconnI don't know about adx, but e.g. mod does not only require the entire file to be in ram at once, but also requires it to be contiguous, without wrap
19:07:46hcsmmhmm, as does NSF, but they're much smaller than adx streams
19:08:16hcsadx is just adpcm, only really needs 18 bytes at a time
19:10:15 Quit lowlight ("CGI:IRC (EOF)")
19:10:20jhMikeSwrap would be taken into account. why would anything absolutely require a non-wrapped buffer anyway?
19:11:08 Join lowlight [0] (i=c730190a@gateway/web/cgi-irc/
19:11:22hcsnsf and sid are programs, though I suppose you could just statically allocate the maximum possible size of the memory space
19:11:39amiconnFor sid, perhaps
19:11:42hcsfor nsf that could be several megabytes
19:11:53amiconnNot for large-ish mod formats like xm or it though
19:12:12preglowjhMikeS: we'd have to modify absolutely all mod players to account for wrapped samples, then
19:12:25preglowjhMikeS: every array access would have to be bounds checked, slow
19:15:49jhMikeSpreglow: still not certain of the difficulty. I'm not even clear on "mod players" :) Surely they can take larger chunks and process locally if needed. :\
19:16:37hcsas they are typically written, they need direct access to all samples in flat arrays
19:16:59preglowjhMikeS: most of the file is basically just samples which the player is told how and when to play back, which makes memory access a big bottleneck. if the samples can be wrapped in memory at any place, then all memory accesses need to be bounds checked, which will slow stuff down very nicely
19:17:36 Join WilliamC [0] (n=chatzill@
19:18:26WilliamCWhat is the status of the iFP port, or is the wiki up-to-date?
19:18:51preglowslowly progressing, still just one guy at work on it
19:22:16jhMikeSah, a wav file with instructions. hmmm...yes depends on the codec design whether it will need frequent bounds checks. the buffer could have the ability to actually rotate if needs to accomodate that. can any CPU alias the buffer twice at two contiguous address ranges? (ok, crazy thought)
19:22:52preglowsome probably, with constraints, but certainly not all
19:23:09WilliamCWIsh I could help
19:23:16jhMikeSMaybe in 64k multiples
19:23:28WilliamCI'm tired of my iFPs poor implimentation of Ogg
19:23:31preglowi really think nonstreamable files shouldn't wrap anyway
19:23:37preglowthis is definitely a good place to apply kiss
19:23:49WilliamCYou're dating?:P
19:24:02preglowWilliamC: i'm coming on to you
19:24:05GodEaterkiss = keep it simple stupid
19:24:13WilliamCI know
19:24:16WilliamCLearned about it in C++
19:24:17jhMikeSfor cf anyway if a second address range can be the same mem
19:24:37preglowyes, but this sounds like a severe bout of making stuff harder than it has to be
19:24:49 Quit Angry ("Miranda IM! Smaller, Faster, Easier.") alias would make it easy since the guard buffer would be the buffer itself
19:26:36preglowyeah, but it's not always possible
19:26:41amiconnjhMikeS: Not just "wav with instructions". MOD formats use an array of samples, which can be played at varying sample rates (i.e. varying pitch) and varying volume, in a number of channels which is normally 4 or greater
19:26:42 Quit Gon ("( :: NoNameScript 4.02 :: )")
19:26:43preglowthis is embedded stuff after all
19:28:13 Quit GodEater (" Like's GUI? Then try HydraIRC -> <-")
19:28:33amiconnThese formats go back to the amiga and its audio hardware, but the formats have evolved and some newer formats support up to 128 channels (afaik) and arbitrary panning of channels, as well as 16 bit samples
19:29:11preglowthe only formats we need to worry about are max 32 channels
19:29:16preglowand we probably won't be able to play even them
19:29:21amiconnThe formats playable on amiga without tricks have 4 channels, 2 of which are mapped to the left and 2 to the right stereo channel, and 8 bit samples
19:29:22jhMikeSI think the codecs should have to handle the dsp and mixing in that case
19:29:27preglowjhMikeS: sure
19:29:31WilliamCMOD leads itself to techno.
19:29:38WilliamCerr lends
19:29:50preglowbut it gets complicated if the sample data isn't guaranteed to be stored linearly
19:30:00*amiconn hopes that at least the simpler mod formats wll be playable on archos with the pcm codec
19:30:07 Join MarcoPolo [0] (
19:30:08WilliamCI have a question though, what is bidi?
19:30:16preglowWilliamC: bidirectional text
19:30:22preglowWilliamC: like arabic, etc
19:30:38WilliamCI meant bidirectional looping
19:30:38preglowthe no idea
19:30:38WilliamCHow the heck
19:30:38jhMikeSI'm hoping I can cook something up simple taking CPU into account that makes it think it's linear
19:30:59preglowjhMikeS: i really wouldn't like a scheme that requires certain cpu features
19:31:15amiconnjhMikeS: I don't think it's possible without an MMU, and the only target featuring an MMU is the gigabeat
19:31:32preglowyou very seldom see mmu's in embedded work
19:31:50 Quit bluey- ("This computer has gone to sleep")
19:32:20amiconnYes, as they increase all sorts of cost (of the chip itself because of increased die size, power consumption etc)
19:32:24linuxstbThe current playback engine (with a 32KB guard buffer) works well IMO. The only thing that needs changing is to load non-streaming codecs into a contiguous area.
19:32:40linuxstb(in terms of buffer organisation)
19:32:48preglowi think i agree
19:32:56preglowbut i don't know that much about it
19:33:08preglowanother scheme is ok, but i think the current one isn't very advanced as it is
19:33:28linuxstb32KB is chosen as the maximum chunk of input data a codec needs, so the playback code is guaranteed to always give the next 32KB contiguously.
19:33:42jhMikeSrotating the buffer around wouldn't be very difficult
19:33:43linuxstbWhich I assume needs a little memcpy'ing.
19:34:06jhMikeSyes, recording does the same fudge at the end. 20k
19:34:11*amiconn thinks that the guard buffer and the requirement to load non-streaming tracks contiguously are in fact related
19:34:33preglowi still think that non-streaming codecs should be able to do their own loading too
19:34:39preglowbut i don't know how that can be done nicely
19:34:41jhMikeSyes, if it fits in the guard buffer, np
19:34:42amiconnEach codec could define its required guard buffer size
19:35:00amiconnFor wav, aiff and such it would be zero
19:35:27WilliamCIs the iFP firmware even useable?
19:35:36amiconnFor other streaming formats it would be whatever the format requires (e.g. the maximum possible frame size in case of mpeg audio)
19:35:36jhMikeSI don't know about zero
19:36:05amiconnFor non-streaming formats, guard_buffer_size = file_size
19:36:28amiconnSo if it doesn't fit at the end, it would wrap
19:37:34 Join barrywardell [0] (n=barrywar@
19:37:43amiconnThere's one more special case that needs handling: if the non-streaming track doesn't fit into the buffer at all. In that case rockbox should just refuse to even buffer the track
19:38:08amiconn(i.e. if guard_buffer > audio_buffer)
19:38:40 Quit hcs ()
19:39:19jhMikeSay dios mio...if we don't give a darn about gapless with a non-steaming format. just give it the whole thing when it's ready to play and don't buffer beyond such tracks.
19:39:39linuxstbIt's not gapless, it's minimising disk spinups.
19:39:45amiconnThat would cause unnecessary spinups
19:40:25jhMikeSthen buffer as many as will fit entirely from the start if there's a train of them.
19:40:41amiconnEven archos could buffer several "typical" ProTracker mods at once
19:41:45Juice^1GB Sandisk MICRO SD 1 G GB MEMORY CARD TransFlash - thats what the sansa e200 uses right?
19:42:03Juice^i want to buy a card from ebay
19:42:10Juice^of ebay*
19:42:40barrywardellyes, it uses micro sd. there isn't support for it in rockbox yet, but I think dan_a is working on it
19:42:56Juice^i know.. yes i noticed
19:43:09 Join GodEater [0] (
19:43:22jhMikeSsounds like a general property that buffering should handle. they could be wrapped until say the previous mp3 finishes then unwrapped before playback.
19:44:12amiconnThat sounds more complicated than a dynamic guard buffer
19:44:31amiconnWhat if the previous track is shorter than what is required to unwrap the non-streaming track?
19:45:30Juice^barrywardell: that card is also known as Memory Stick Micro™ (M2™)?
19:45:45jhMikeSthen it's not fully loaded yet or can't fit anyway if I get what you're saying
19:45:56amiconnIt can
19:46:08barrywardellmemory stick micro sounds different
19:46:44barrywardellin theory we might be able to use some kind of mmc and microsd in Rockbox
19:47:01amiconnImagine the following start> nst2b | st3 | st1 | nst2a <end
19:47:04Juice^probably, because my sony ericsson uses the micro (m2) and i cant make it fit into my sansa. but the size seems to be about correct
19:47:16 Join Everybody [0] (
19:47:19amiconn(st == streaming track, nst == non-streaming track, numbers are in order of playback)
19:47:29barrywardellmemory stick is different to sd
19:47:43barrywardelli doubt they're compatible
19:47:47 Join BobJonkman [0] (
19:47:50amiconnIf nst2a+nst2b > st1+nst2a, it wouldn't work
19:47:52jhMikeSprevious track will be dumped in preparation so how could I not unwrap the track if it is small enough? All previous tracks will be dumped.
19:47:52barrywardellonly sd and mmc are compatible iirc
19:48:22amiconnjhMikeS: Yes, the previous ones, but there might be subsequent tracks as well
19:48:30 Join Thundercloud [0] (
19:48:49amiconn...which can be located before the previous tracks, memory-wise
19:49:43 Join muesli__ [0] (n=muesli_t@
19:49:45 Quit phrozen77 ("Some people say that I must be a horrible person, but that's not true. I have the heart of a young boy... in a jar... on my d)
19:49:49jhMikeSsure...the stuff before will be wrapped to the end like a giant rol instuction
19:50:59 Quit GodEater (" HydraIRC -> <- IRC has never been so good")
19:53:41amiconnjhMikeS: There are some pitfalls in that concept. Moving buffered tracks around requires updating metadata pointers with metadata on buffer (another desired feature)
19:54:14jhMikeSI'd use offsets not pointer
19:54:36linuxstbWon't the metadata have to move anyway? e.g. for tracks larger than the buffer?
19:54:37amiconnThe other big question is how to implement that giant rol to ensure nothing gets overwritten, without having (much) temporary storage available
19:54:53 Join bluey- [0] (
19:55:06amiconnlinuxstb: In that case, exactly one metadata entry has to move, in a predefined position
19:55:35jhMikeSavoiding temp space if say the previous file is tiny but the wrapped bit is larger
19:56:23linuxstbBut I thought the idea of metadata-on-buffer was to do exactly that. Are you proposing a static buffer still?
19:56:38amiconnImho the easiest way for having metadata on buffer and support tracks larger than the buffer is having one static metadata slot, where the metadata of the current track is copied into as soon as the track starts
19:56:59amiconnThe metadata on buffer could be a linked list
19:57:03jhMikeSunwrapped offsets would be correct after a move
19:57:19 Quit gotthardt ("Chatzilla 0.9.77 [Firefox]")
19:58:22linuxstbBut a single static buffer would have to be large if we are to support all the things wanted - album art, lyrics, cuesheet/chapter/sub-track metadata etc
19:59:51jhMikeSI'm thinking a callback for when the user of the data must do something with it or lose it
19:59:56*amiconn would drop all that except lyrics, but that's jsut me
20:00:58 Join BigBambi [0] (
20:01:00amiconnlinuxstb:The album art needs to be buffered anyway, so that's no loss
20:01:10amiconnIt's just another place to buffer it
20:01:33 Join bun-bun [0] (n=bun@unaffiliated/bun-bun)
20:01:54amiconnBesides, album art as metadata of tracks can't be called album art ;)
20:02:19linuxstbNo, but potentially each track in the buffer has its own artwork.
20:03:01amiconnLet's call it track art, then ;)
20:03:03jhMikeSthe callback would be in the metadata header on the buffer, along with the size and it would just pass it there.
20:03:30*amiconn still wonders what album art has to do with listening to music
20:03:35linuxstbamiconn: But then people will want support for reading it from the track metadata (i.e. ID3...)
20:03:44*amiconn had never seen a radio which displays album art
20:03:53jhMikeSisn't some of that stuff at the end of the file?
20:04:24 Nick kclaf2K7 is now known as kclaf (
20:05:18jhMikeSamiconn: it's pretty to look at? gives one a reason to have the backlight on all the time. :)
20:05:20 Join WilliamC_ [0] (n=chatzill@
20:06:13linuxstbI like having something visual to associate with a track - I can remember album covers better than names sometimes.
20:06:40 Quit WilliamC (Read error: 104 (Connection reset by peer))
20:06:55 Nick WilliamC_ is now known as WilliamC (n=chatzill@
20:09:25 Nick mako_ is now known as mako (
20:09:28jhMikeSI wouldn't want album art being reloaded for every track if there's just one image for the entire directory and nothing in the files themselves
20:09:29 Quit _Veseliq_ (Remote closed the connection)
20:09:42amiconnjhMikeS: Your callback idea would achieve exactly the same as copying at start of track, only with more code
20:10:18amiconnThe owner of the data would still need a buffer to copy the data into. WPS needs the metadata all the time
20:11:35amiconnAnd we need 2 static metadata buffers if we want the next track info to be always available
20:11:50jhMikeSA bit more, yes. If it's gonna all be for WPS then it just tells WPS but I'm not going to hard code a call to it if the WPS isn't involved.
20:12:48jhMikeSIf the current track is too large for the buffer then the next track info never goes on the buffer :\
20:13:19amiconnyes it does, but only later in track 1
20:13:38amiconnThe wps might not be running at the time the metadata on buffer gets overwritten, but running again before the track ends
20:13:40jhMikeSSeems pointless to load it again if you can just copy it from the next track info to the current
20:13:45amiconnSo the data still needs to be copied
20:14:29amiconnYes it can be copied from the next track info - if that is stored in a second static buffer
20:14:37 Quit RedBreva (Remote closed the connection)
20:14:47jhMikeSThat would be if the normal playback control is being used but not nescessarily if a plugin is substituting for the playlist engine which will be technically possilble.
20:14:50amiconnI think this should be optional, and disabled on low-mem targets
20:15:03 Join RedBreva [0] (
20:15:10amiconn(archos and iFP atm)
20:16:02jhMikeSI suppose what data is used is optional but implement two entirely different methods alltogether you're saying?
20:16:21amiconnThey're not entirely different
20:16:24amiconnBtw, why should a plugin substitute the playlist engine?
20:17:16amiconnIt's just that next track info could either have a second static buffer, or not
20:17:24amiconnAll subsequent metadata would be on buffer
20:17:25jhMikeSwhy not? I don't really want that level tied exclusively to higher ones. It's just a consequence of separating things. Plugins could atm substitute for the recording screen and do it all.
20:17:52amiconnWhat real davantage would that bring?
20:18:38 Join fasmaie [0] (
20:18:40jhMikeSa clean separation of levels and playback not having to deal with things like directories and such as it does now. that should be irrelevant to it.
20:19:03amiconnSince when does playback deal with directories?
20:19:38jhMikeSIt does via calling back to playlist functions and even has function calls to initiate directory changes. bad.
20:20:16amiconnThe dirchange functions should be taken out of playback.c. They are playlist stuff, or even above that
20:20:28jhMikeSyes...did that in my cvs
20:20:46jhMikeSplaylist controls playback. wps controls playlist.
20:20:48amiconnBut how would you avoid the playback engine having to call back playlist?
20:20:55 Join PaulPosition [0] (
20:20:58amiconnSure the engine needs to get the next track?
20:21:33jhMikeSIt only queries what it should do through an abstracted control function. Anything could register as the controller.
20:21:35amiconnNo, wps controls playback, not playlist, and it also reads data from playback
20:22:05 Join Mouser_X3 [0] (
20:22:07jhMikeSI changed it around a little :)
20:23:57jhMikeSSo in that case something else could use playlist to control playback using playlists but no explicit engineering...just a side effect. All levels individually useable by anything just like recording.
20:24:09*amiconn still doesn't see the point
20:25:50amiconnSurely pause/unpause, stop, seek forward & backward are all playback functions, not playlist
20:25:58amiconnSkip is debatable...
20:26:41 Part fasmaie
20:26:42 Quit RedBreva ("Time for Tubby ByeBye")
20:32:45jhMikeSwell, even if a few are passed through directly or as inlined passthrough functions I just decided on a clean break in concept. It seems mpeg.c has a cleaner break as well since some calls are reversed in playback.c to playlist. See the #if CONFIG_CODEC != SWCODEC stuff in playlist.c
20:33:52 Join pieroxmysox [0] (i=pieroxmy@
20:34:13pieroxmysoxoh dang
20:34:15jhMikeSAnd if you want audio event, you just register for them. Ignore the ones you don't care about. I don't want a separate callback for every possible event. It's not speed critical.
20:35:56 Part pieroxmysox
20:37:57 Join XavierGr [0] (
20:39:49 Join toer [0] (
20:41:09 Join EnzoTen [0] (
20:41:45jhMikeSFor resume info, playlist can handle it instead of playback doing it by calling back the playlist. I wonder if much calling back will even be needed. It could be done by pushing the tracks to playback instead. Playback can still serve as a cache for the buffered information. I'm sure I'll do a hundred revisions before I'm happy with it. :)
20:46:32 Quit barrywardell (Read error: 104 (Connection reset by peer))
20:46:51 Join barrywardell [0] (n=barrywar@
20:47:30 Join Digamma [0] (
20:48:00DigammaIs there even the slightest chance that Rockbox development will continue on the Sansa e200 port without the desired documentation...?
20:48:19 Quit bluey- ("Leaving")
20:48:36 Quit barrywardell (Read error: 104 (Connection reset by peer))
20:49:32 Join petur [0] (n=petur@rockbox/developer/petur)
20:49:55PaulPositionDigamma - Well all that's been done has been done without documentation, so I don't see why it would stop dead unless devs get bored. The e200 being such a nice little device.
20:50:10PaulPositionDigamma - why?
20:50:30DigammaIs is; I just as well assume that you might not continue without the proper information
20:50:37 Quit WilliamC (Read error: 110 (Connection timed out))
20:50:46jhMikeSdevelopment stopped on it? noone told me :)
20:50:57PaulPositionDigamma - It comes in waves.
20:51:44PaulPositionDigamma - Months ago, someone decrypted the .mi4 format. Not much happened from it on the sansa, but lots happened on 'same chipset' iRivers H10 and iPods..
20:52:21 Part lowlight
20:52:25PaulPositionDigama: Then the H10 guys got E200. They stopped caring for H10 for a time and played and got E200 almost functionning. That's about a month ago.
20:53:22PaulPositionDigamma: No 'talked about' progress in two weeks doesn't mean the port is dead. :)
20:53:38 Join barrywardell [0] (n=barrywar@
20:55:17 Quit EnzoTen ("Chatzilla 0.9.77 [Firefox]")
20:57:31 Join webguest87 [0] (i=5438ae4d@gateway/web/cgi-irc/
20:57:43 Join Jsunu [0] (
20:58:01DigammaI just absolutly love rockbox, I just installed it yesterday
20:58:10DigammaI havent stopped playing with it (well other than coming on here)
20:58:33 Join webguest93 [0] (i=5873b21c@gateway/web/cgi-irc/
20:59:30Juice^i love rockbox too digamma... been playing around with wps themes on my sansa:)
20:59:35 Join tj2master [0] (i=a@
20:59:49DigammaI cannot seem to figure out wps...?
20:59:57DigammaHow do you load them on to the sanasa
21:00:22Juice^you place the needed files in the correct directories...
21:01:15Juice^but currently there is only 1 theme supported
21:01:25Arathis_PaulPosition, barrywardell: don't know if you noticed it already, but instead of skipping playback while running peakmeter, EQ or a plugin (e.g. bubbles) the plugin slows down. is this intended?
21:01:25Digammaumm, one other question; when you are upgrading to the newest CVS version; to you overwrite the rockbox.e200 file?
21:01:33DigammaI know that oragne black theme
21:01:35 Nick Arathis_ is now known as Arathis (
21:01:54DigammaI thought you were talking about the GOOD looking themes...
21:02:16Juice^ah nope... have just been editing that black orange theme, to a newer one i like better
21:02:28Juice^but no special graphics yet
21:02:43DigammaHow do you change the backdrop...?
21:02:55DigammaThere is the folder in .rockbox, but I havent a clue
21:03:06DigammaI have loaded a bunch of Jpegs into it, And I cannot find it in the settings
21:04:10 Join GodEater [0] (
21:05:32***Saving seen data "./dancer.seen"
21:06:48Juice^you need to load it via a .cfg file in \.rockbox\themes
21:06:58Juice^backdrop: /.rockbox/backdrops/ipntwd00.bmp
21:07:01Juice^like that
21:07:53 Join EnzoTen [0] (
21:08:33 Quit webguest87 ("CGI:IRC (EOF)")
21:08:52PaulPositionSorrt, Arathis...
21:09:01DigammaThanks Juice^
21:10:22Digammacan the backdrop be a jpg...?
21:10:25 Quit barrywardell ()
21:10:28PaulPositionArathis: Yes, there was a commit in cvs some weeks ago that made the scheduler give priority to the audio playback over most any other bits of code that wants to run. So the UI slows down instead of the audi skipping.
21:10:40peturDigamma: BMP
21:10:42 Join decayedcell [0] (i=3ba7bfb3@gateway/web/cgi-irc/
21:10:46Juice^Digamma: has to be 8bit bmp i think
21:11:13petur24bit should work too iirc
21:11:19DigammaAny other particulars..?
21:11:19Juice^ok kool
21:11:28 Join barrywardell [0] (n=barrywar@
21:11:29peturthe right size ;)
21:11:38Digamma198 by....
21:11:41jhMikeSAnyone that has playback engine ideas on the technical level go ahead and put up a page to post them or start a dicussion. I could use a few and things have been helpful today. I just gotta take a nap and am too tired... :P
21:12:12pixelmabut it must be in the same resolution as your display (176x220 IIRC)
21:12:48DigammaI didn't know it was that high a resolution
21:13:39pixelmaI mean dimensions... 176 pixels wide - 220 pixels high
21:13:44DigammaI know
21:15:04 Quit decayedcell (Client Quit)
21:15:09 Join decayedcell [0] (i=3ba7bfb3@gateway/web/cgi-irc/
21:16:55Juice^Digamma: if you're not already doing it, i'd recommend you to first test the themes/wps on the simulator before you transfer them to the sansa
21:17:16DigammaWill do
21:17:34 Part EnzoTen
21:18:09PaulPositionWhat's the point in having a WPS if sansa doesn't yet play anything? (Or did I miss a "Gentlemen,..." post?)
21:19:49Juice^Well, i dont know much coding, so why not make some WPS ready for the sansa in the time before it gets sound working
21:21:39PaulPositionThat I understand.. I'm considering converting some for my friend's e200, for when rockbox is in a semi-ready state on that target.. :)
21:22:10Juice^nice then i hope you will share them with us all aswell =)
21:22:38PaulPositionOf course, I'll need someone other than a sim to tell me if they work. :p
21:23:15PaulPosition(might be pi-julius style wps that would require special builds though..)
21:25:44DigammaWoot, juice it worked, in the sim at least
21:26:19Juice^good good
21:26:25Digammawhat is even cooler, is I can play with the settings FROM my sansa; I dont need the computer; I can use the text editor!
21:26:52Juice^and you can change fore and back colors
21:26:56amiconnDigamma: The backdrop bmp can be any standard uncompressed bmp, i.e. with a depth of 1, 4, 8, 16, 24 or 32 bit
21:27:14DigammaMy image prog has a default of 24 which works
21:28:12 Join Mouser_X [0] (
21:28:31 Join robin0800 [0] (i=5003400a@gateway/web/cgi-irc/
21:29:07Juice^photoshop <3
21:29:56DigammaGimp <3
21:30:05DigammaI am a linux freak
21:31:16DigammaI am converting all images in my pics to 220*176*24 bmp's
21:31:33 Quit Criamos ("( :: NoNameScript 4.03 :: )")
21:31:47PaulPositionDigamma: wait...
21:31:58Digammawhat is it?
21:32:19PaulPositionDigamma: The images you want as backdrop need to be bmps, but there's the jpeg viewer on rockbox that read jpegs.. For slideshows and such.
21:32:35PaulPositiondigamma: And, weirdly enough, it doesn't read bitmaps.
21:32:37DigammaI know that, and...?
21:32:43DigammaYeah; that is odd I noticed
21:32:55DigammaProb because of the whole microsoft owns bmp crap...?
21:32:59 Part jdong ("Ex-Chat")
21:33:10PaulPositionDigamma: Nope. No such crap.
21:33:23PaulPositionDigamma: Just that there's no nice .bmp viewer plugin.
21:33:44Juice^there should atleast be general viewing support of bmps... strange
21:33:47Digammaodd; none in the plugin repostitory either...?
21:34:26PaulPositionDigamma: There's a 'bmp viewer' patch, but its old and totally irelevant to the newer bitmap code.
21:34:32PaulPositiondoesn't compile.
21:35:28DigammaWell the sansa port is virtually unusable; so I wouldn't hold my breath just yet (I dont mean menus, I mean audio)
21:36:26 Join Llorean [0] (
21:36:58Juice^i play chess on it sometimes.... hehe. i just want soooound so much :)
21:37:03 Quit robin0800 ("CGI:IRC")
21:37:14 Join robin0800 [0] (i=5003400a@gateway/web/cgi-irc/
21:37:25DigammaI have been playing doom; I would love sound as well
21:37:51 Join WilliamC_ [0] (
21:37:53 Nick WilliamC_ is now known as WilliamC (
21:38:18 Quit MarcoPolo (Remote closed the connection)
21:38:19PaulPositionHmm.. "cvs up -C" to update everything that needs updated and clear local modifications to the source?
21:40:16decayedcelland that WPS can't display jpg
21:40:22 Join bluebrother [0] (i=79RjeMwA@rockbox/staff/bluebrother)
21:40:35decayedcellwhich is why it has to be in bmp :P
21:41:17DigammaDoes anyone here know of a good utility which will convert web pages to jpeg's...?
21:41:23 Join __Margot_ [0] (
21:42:07barrywardellPaulPosition: that clears modifications
21:42:17GodEaterDigamma, Alt-Printscreen, then paste into MS Paint ?
21:42:55pixelmafor a linux guy? ;)
21:43:18GodEatergimp's screen capture ? :)
21:43:55bluebrotherask freshmeat?
21:43:59 Part Llorean
21:44:03 Quit Mouser_X3 (Read error: 110 (Connection timed out))
21:45:43 Quit decayedcell ("CGI:IRC (Ping timeout)")
21:47:32 Part robin0800
21:50:07DigammaIf you are to fix a typographical error at the beginning of a line in the text editor, do you HAVE to delete the whole line to do it...?
21:50:25Digammanvm, found it
21:57:37 Join My_Sic [0] (
21:57:38 Quit _Margot_ (Read error: 110 (Connection timed out))
21:57:43 Nick __Margot_ is now known as _Margot_ (
22:03:26 Join tucoz [0] (n=martin@rockbox/staff/tucoz)
22:03:37tucozhi, and happy new year
22:04:18tucozhi pixelma, do you still have that file?
22:04:21 Join strabes [0] (
22:04:47bluebrotherhi tucoz
22:04:48 Join bluey- [0] (
22:05:05bluebrotherback to Rockbox? ;-))
22:05:08pixelmaHappy New Yoear to you too, tucoz. I suppose you are speaking about the sansa.svg? :)
22:05:47bluebrotherI want to add CREDITS-MANUAL soonish. Any changes to what we already got?
22:05:48tucozhehe. I hope so. My lazy periods tend to last for a while lately
22:07:07tucoziirc, that list looked fine
22:07:20 Join Shawn_K [0] (
22:07:40*tucoz looks at a nice header in the preamble :)
22:07:58tucozlooks great
22:08:22Shawn_Khey all
22:08:56Juice^whats a good place to upload rar files? (to share a test theme for the sansa)
22:09:00Shawn_Kif anyone understands the way threading works w/ CO-P, could anyone help explain that to me?
22:09:12Juice^is megaupload ok=
22:09:13 Quit GodEater (" HydraIRC -> <- The professional IRC Client")
22:09:28Shawn_KI wish to slowly move a few threads of code to CO-P for the iPod
22:09:34Shawn_Kas patches
22:09:40 Join idnar [0] (i=mithrand@unaffiliated/idnar)
22:10:51bluebrotherrar. *urgh*
22:11:18Juice^Digamma: if you'd like to see the simple wps i played with, its here:
22:11:27PaulPositionShawn - Did you first have a look at that patch?
22:11:35DigammaJuice^, I got it working
22:12:32Juice^you're looking at it now?
22:12:39 Quit Thundercloud (Remote closed the connection)
22:12:42Juice^you're fast.
22:13:45bluebrothershould I do it this way, like the existing CREDITS?
22:14:12tucozI think so
22:14:37Shawn_KPaulPosition: yes, I did...I installed it, I'm looking at it and starting to understand
22:14:49Shawn_KI was hoping someone could get me move faster
22:15:26Shawn_Kanyways, is the CO-P slower than the main processor?
22:15:40Shawn_KI noticed the sleep is much larger for it
22:16:06Shawn_K****I don't fully understand the function of sleep() for running threads either
22:16:49PaulPositionShawn_K: I think you need to have specific questions like that last one.. Look at people posting in that flyspray entry and if they're here on IRC, include their name in your question in case they use notification (colour or beep or whatever) in IRC..
22:17:23PaulPositionShawn_K: Wish I knew anything about programming. Happens I only know about asking. ;)
22:17:34Shawn_Klol, heh...
22:17:58Shawn_KI need to look at the actual threads and maybe it will clear it up for me
22:18:26Shawn_KI just started coding on rockbox last night
22:18:50Shawn_Klol, so I made some optimizations...but also made one bug in the process which I am fixing atm too
22:19:24Shawn_KI messed up the waiting until the LCD is ready to be drawn too...if you have more than one function trying to draw, it freezes
22:19:41Shawn_Kso it runs smooth if you don't fast forward tracks too fast
22:19:46Shawn_K*skip tracks
22:20:23Juice^Shawn_K: what player do you have?
22:20:36Juice^ah ipod :x
22:21:36 Join Ribs [0] (n=ribs@
22:22:20Shawn_Khow did you figure I have iPod?
22:23:41Juice^i read up what you where writing about
22:23:56Juice^22:09:32 ‹Shawn_K› I wish to slowly move a few threads of code to CO-P for the iPod
22:28:33Shawn_Koh, lol
22:29:38 Join BobJonkma1 [0] (
22:30:43 Quit strabes ()
22:31:09 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:33:26Bagdergotthardt is now a new cvs committer
22:34:01 Quit spiorf (Remote closed the connection)
22:34:57 Join lavi2 [0] (
22:35:40 Nick Everybody is now known as Everybody|away (
22:37:36 Quit Digamma ("Leaving")
22:38:48tucozbluebrother, you can add Piotr from my latest commit to the manual credits as well, before you commit it.
22:39:46 Quit ashes ("leaving")
22:40:17 Part tj2master
22:40:24 Join kaaloo [0] (
22:40:39scorchetucoz: is there a reason why there are a few options missing in some manuals like car adapter mode from ipod/gigabeat (if it is because they dont support them yet, perhaps we should disable them in the code?)
22:41:00bluebrother"submenu"? Weird name for a button ...
22:41:10tucozi know
22:41:24tucozi only used what was in the e200 manual
22:44:03pixelmahmm... I think that center thing or "glowing" part is a scroll wheel...
22:44:40scorchewell, it is a wheel that scrolls
22:45:50 Quit maxkelley ("leaving")
22:47:11pixelmaI mean in the Ipod manuals it is also labelled "scroll ffwd/back" - so should that be done for the Sansa grapic too?
22:47:30scorchei would imagine it should be
22:47:31 Quit BobJonkman (Read error: 110 (Connection timed out))
22:47:39 Join BobJonkman [0] (
22:48:05 Quit amiconn (Read error: 110 (Connection timed out))
22:48:05 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:48:21Juice^whats up on the sansa updating?
22:48:31pixelmaso... the actual wheel is the backlit part?
22:48:56scorcheyes..that little ring
22:49:00pixelmathanks :)
22:49:17Juice^the light is activated as soon as you rotate the wheel
22:49:38scorchein OF at least
22:50:40 Join strabes [0] (
22:50:42barrywardelland in rockbox :)
22:51:01*scorche points to the at least part ;)
22:52:12scorchealways write your responses with room for error...
22:53:03 Quit barrywardell ()
22:53:36 Join Strath [0] (
22:59:06 Join tucoz_ [0] (
22:59:08 Quit bluey- ("Leaving")
23:00:47 Quit tucoz (Read error: 110 (Connection timed out))
23:03:30tucoz_we can rename the keys another day. at least the sansa images are available now.
23:03:38 Join webguest25 [0] (i=46748840@gateway/web/cgi-irc/
23:04:10 Part tucoz_ ("good night")
23:04:40webguest25Can anyone give me a walkthrough on the rockbox video from installing the plugin to playing a video?
23:05:10 Quit BobJonkma1 (Read error: 110 (Connection timed out))
23:05:38***Saving seen data "./dancer.seen"
23:05:46pixelmaargh... tucoz left... I prepared a graphic with the arrows for it... same link as before (I hope he reads the logs)
23:06:17scorchewebguest25: see the wiki page PluginMpegplayer....and the plugin is already "installed"
23:06:21scorchepixelma: he does
23:06:54webguest25i dont see it in the blugins on my player
23:07:04Juice^pixelma: what graphics is this?
23:07:13bluebrotherwebguest25, what player?
23:07:20pixelmayes I thought so too - I only wanted to explain why I still "talk" to him
23:07:38bluebrothergreyscale? color?
23:07:47bluebrothermpegplayer only works on color targets
23:07:52webguest25yes i know
23:07:55linuxstbIt's not in the plugin list.
23:08:00dionoeayou should use appl firmware for video on the ipod video :)
23:08:00webguest25Where is it
23:08:01pixelmaJuice^: it's for the Sansa manual, to explain the controls
23:08:08linuxstbAnd it performs very badly on the ipod video...
23:08:15Juice^pixelma: ahhh ok. you're working on a manual. cool
23:08:22linuxstb(relative to the Apple firmware).
23:08:24dionoeaapple without the e i guess :)
23:09:12webguest25but how do i play a video on rockbox
23:09:22My_Sichello everyone
23:09:28scorchewebguest25: read the page i linked you
23:09:34scorchewell, told you to go to
23:09:37webguest25can u link it again
23:09:42 Part lavi2
23:09:46bluebrotherscroll back? ;-)
23:10:03webguest25u didnt give a link
23:10:07scorche<scorche> webguest25: see the wiki page PluginMpegplayer
23:10:19webguest25i already did its just confusing
23:10:24pixelmaJuice^: we try to provide a manual for all players rockbox supports - the Sansa on is in preparation, too
23:10:26bluebrothergo to the wiki and type that as page name
23:11:41 Quit webguest25 ("CGI:IRC (EOF)")
23:12:30 Quit Ribs ("eh eh ehhhh!")
23:12:36 Join Llorean [0] (n=Llorean@rockbox/administrator/Llorean)
23:13:40 Nick Everybody|away is now known as Everybody (
23:14:25 Quit juxtap (Read error: 145 (Connection timed out))
23:14:27 Quit kaaloo (Read error: 113 (No route to host))
23:15:13 Quit WilliamC (Read error: 131 (Connection reset by peer))
23:15:30 Join WilliamC_ [0] (
23:15:30 Nick WilliamC_ is now known as WilliamC (
23:15:37 Join lachlan__ [0] (i=3a6ce792@gateway/web/cgi-irc/
23:18:26 Quit lachlan__ (Client Quit)
23:19:51 Quit debauched_sloth ("Leaving.")
23:20:44 Join x1jmp [0] (
23:21:08 Join dan_a [0] (n=dan-mirc@
23:21:45dan_aBagder: I've got home to find a package from Sweden :D
23:21:49linuxstbThis doesn't sound good (from the Changelog to the Archos 104 firmware...) - "- a hourglass is displayed at the transition between two songs,"
23:22:07Bagderlinuxstb: hahaha
23:22:25LloreanSo... gap++?
23:22:30linuxstbSo I guess it isn't gaples...
23:24:06 Quit petur ("sssssssssss---------PLOP!")
23:25:10 Join Kijutsu [0] (
23:25:23KijutsuJust a quick question, do you have plans to support iPod Gen 5.5?
23:25:37preglowwe already do, to some extent
23:25:41preglowwe just don't support the 80 gig version yet
23:26:03KijutsuIs it safe to assume the 80 gig will be supported eventually?
23:26:14Bagderwe also plan to conquer the world and make all mp3 players come with a kitchen sink built-in
23:26:15preglowi can't imagine why it wouldn't
23:26:46KijutsuBagder, the kitchen sink would make it not-so-portable anymore. :P
23:26:49LloreanThe 80gb iPod 5.5G is probably the only port it's safe to assume will happen.
23:27:15Kijutsupreglow, okee. Just curious, I was poking around the website and saw the 80 gig wasn't supported at all right now, but nothing as far as possible plans for it.
23:27:15LloreanWell, and the rest of the Sansa port, so long as the black magic doesn't give out.
23:28:02LloreanKijutsu: "Plans" aren't really made. Either things happen or they don't, but work has happened on the 80gb but the key problem just hasn't been solved yet.
23:28:15dan_aThe Sansa might end up doing everything except playing sounds, though...
23:28:30Kijutsuthanks for the info, folks.
23:28:32 Part Kijutsu ("Leaving")
23:28:34linuxstbCan anyone tell what's inside the Archos 104 from the firmware?:
23:28:41preglowdan_a: i thought sound wasn't too far off?
23:28:42*linuxstb back later
23:29:10LloreanSo far Mr H has restricted himself to PP5024-related stuff, right?
23:29:20BagderPP at least
23:29:29dan_apreglow: It's one datasheet away. And nobody's managed to get a copy which can be legally used to write a driver
23:29:54dan_aLlorean: He's done some of the GPIO driven interfaces (eg the radio) too
23:30:02 Quit Mouser_X (Read error: 110 (Connection timed out))
23:30:11dionoeayou have the radio but no sound ? :)
23:30:43preglowdan_a: right, right, remember now
23:30:51 Quit Kasperle (Read error: 131 (Connection reset by peer))
23:30:58dan_adionoea: No driver for the radio yet. But it's a bit closer now the developer writing the driver has a device with a radio!
23:31:24dionoeaheh :)
23:31:36tsuyoshidan_a: you can't reverse engineer the sansa audio?
23:32:07Lloreantsuyoshi: There's really not a terrible lot of reverse engineering going on.
23:32:33 Join phrozen77 [0] (
23:33:24dan_atsuyoshi: The guy who is doing the reverse engineering worked out which chip is used, which is a lot quicker and more accurate than writing a document describing how he thinks we should talk to unknown hardware
23:34:40dionoeaand the chip manufacturer's didn't reply to the email you sent them i guess ?
23:35:52 Part BobJonkman
23:36:00BagderMrH got the data sheet
23:36:11BagderI got as far as a phone call from them
23:36:20Bagderno one else seems to have got a response
23:36:25tsuyoshimrh got the audio chip data sheet?
23:36:37tsuyoshiwithout an nda?
23:37:04*dionoea guesses that Bagder and MrH are the same person ... :D
23:37:07Bagderyes, although the doc is marked as "secret" or something
23:38:34Bagderlinuxstb: the UPDATE_Archos104.AOS file seems to use the same format as the gmini402 update image so the archopen guys might have some clues
23:40:25Juice^too bad the data sheet cant be shared on rockbox's website. i guess it would make stuff easier for the sansa port...
23:41:29Bagdertoo bad austriamicrosystems are this uptight
23:41:50tsuyoshiso.. are we waiting for the person with the data sheet to write the audio driver or what
23:41:53dan_aWhen I get some spare time during the working day, I'll give their UK sales office a ring and see if I can get anywhere through talking to someone
23:42:03 Quit PaulPosition ()
23:42:23Bagderdan_a: I believe the sales offices forward info to the main place to do the actual decision
23:42:24tsuyoshiI was going to try to figure it out by disassembling the original firmware
23:42:45Bagderat least the Swede who called me gave me that impression
23:42:46tsuyoshiI'm wondering if I should still go ahead and do that
23:42:58dan_atsuyoshi: No - MrH (who has the datasheet) wishes to be anonymous, and Rockbox doesn't accept anonymous code.
23:43:24tsuyoshiok... I'll go ahead then
23:43:39dan_aIt would be a valuable contribution :D
23:43:52bluebrotherwould it be possible to use information from a "secret" data sheet in gpl code anyway?
23:43:55tsuyoshiwell yeah.. the original firmware sucks
23:44:10tsuyoshiusing it is driving me grazy
23:44:21Bagderbluebrother: not if we would be prohibited to reveal details, no
23:44:41bluebrotherthought that.
23:44:50dan_abluebrother: I understand that some Linux drivers have been written from NDA data sheets, with the explicit agreement of the manufacturers
23:45:21bluebrotherbut they need the explicit agreement of the manufacturer, don't they?
23:45:53bluebrotherI read about some issue regarding a ati graphics card some while ago −− no response from ati, so no support for linux
23:46:25bluebrother(and, as far as the dev said, it was only trivial stuff that allowed 2d only without revealing anything special)
23:48:40tsuyoshias if their competitors can't just disassemble the driver anyway.. jesus
23:48:55bluebrotherworking with publicly available data sheets is much more fun :(
23:49:29bluebrotherbut is the work needed for disassembling worth it for competitors?
23:49:39bluebrotherI guess usually not
23:50:14Bagderbut in cases like this, companies just must lose business by being so secretive
23:51:17preglowtsuyoshi: disassembling something that big is really huge undertaking
23:51:23preglowtalking 3d graphics drivers
23:51:25LloreanI'm sure they feel it's justified somehow
23:51:42preglowbig and complex
23:51:50bluebrothermoney justifies mostly everything for the selling guys ...
23:52:02Shawn_Klol, disassembling isn't just looking at some C++ code...that's pretty much looking at a mix of assembly and machine code >.<
23:52:07preglowmost people in that league like to play it safe anyway
23:52:12preglowwhich involves doing what everyone else is doing
23:52:14preglowwhich means nda
23:52:23preglowShawn_K: assembly is machine code...
23:52:36Shawn_Kno, assembly is higher level
23:52:45preglowShawn_K: well, all you get is assembly anyway
23:52:54dan_aShawn_K: Did you figure out how to start a thread on the COP?
23:52:55bluebrotherso what's the difference between assembly and machine code?
23:52:55preglowlooking at machine code is for Special People
23:53:05preglowalso, it's pointless
23:53:16LloreanAssembly is just human readable machine code.
23:53:18Shawn_Kdan_a: I looked at it, but I need some help...It's real tricky
23:53:20LloreanThe level isn't really any different.
23:53:28 Quit Arathis ("[rl_bot quit]")
23:53:30LloreanIt's just the expression.
23:53:31bluebrotherassembly == mnemonics for machine code
23:53:32Shawn_KLlorean: pretty much
23:53:48Bagderbluebrother: assembly is usually/sometimes built with some "pseudo-instructions" that might not be native machine code
23:53:48bluebrotherunless someone tells me better ;-)
23:54:02preglowbluebrother: aye
23:54:09Shawn_KBadger: EXACTLY!
23:54:25bluebrotherthat sounds like a macro assembler ...
23:54:44Shawn_Kmachine code can be simplified into assembly, but it's not 100% the same
23:54:59Bagderit depends on the arch/cpu
23:55:49Shawn_Kanyways, as dan_a mentioned, I want to move some threads on rockbox over to CO-P...but I'm having difficulty understanding the threading system in full
23:56:02 Nick Everybody is now known as Everybody|away (
23:56:04Shawn_Kanyone who can help me out?
23:56:12 Join Arathis [0] (
23:56:30dan_aShawn_K: You use create_thread_on_core(core, function, stack, stack_size, name) where core is either COP or CPU and the other arguments are the same as create_thread()
23:56:54linuxstbShawn_K: And have fun with the caches...
23:56:59Shawn_Kstack = ???? (assembly code?)
23:57:44Shawn_Kyeah, I understand stacks when talking about low-level code
23:57:45dan_aShawn_K: Find an example of create_thread() in the code :)
23:57:49Shawn_Kbut not C++...
23:58:05linuxstbWho's talking C++?
23:58:10*bluebrother never saw c++ in Rockbox
23:58:40Shawn_Kok, C
23:58:51Shawn_Ki confuse my programming languages

Previous day | Next day