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 2005-09-25

00:04:01 Join NeoRa_fu [0] (
00:04:15 Part NeoRa_fu
00:09:39 Join paugh [0] (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
00:14:34***Saving seen data "./dancer.seen"
00:20:23 Quit Maxime` (Read error: 104 (Connection reset by peer))
00:21:54 Join Maxime` [0] (
00:36:52 Join XavierGr [0] (
01:07:30 Quit Moos ("Glory to Rockbox")
01:26:06 Quit _DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
01:37:15 Quit XavierGr ()
02:13:46 Join peter200lx [0] (
02:14:36***Saving seen data "./dancer.seen"
02:18:00 Quit peter200lx (Client Quit)
02:21:03 Join Peter200lx [0] (
02:23:01 Quit Peter200lx (Client Quit)
02:23:30 Join Peter200lx [0] (
02:24:22 Quit dpassen1 ()
02:24:43Peter200lxI Downloaded 2.5 yesterday and installed it
02:25:05Peter200lxBut all the plugins say [Incompadable version]
02:25:15Peter200lxSo I'm back to using 2.4
02:25:36Peter200lxI have a Rockbox Recorder 20GB
02:28:59Peter200lxWould the latest build solve this?
02:29:29 Join webguest56 [0] (
02:29:46webguest56I think there is something wrong with the installer
02:29:59webguest56You need to install using the zip instead
02:30:15webguest56Did you use the win-installer?
02:30:41 Quit Peter200lx ("CGI:IRC")
02:30:48 Join Peter200lx [0] (
02:31:04Peter200lxYes I used the windows installer
02:31:04 Join ashridah [0] (
02:31:10Peter200lxI'll try the Zip
02:31:14webguest56Ok, then try with the zip instead
02:31:55webguest56or maybe the exe is updated now. I think so
02:32:00webguest56It works?
02:32:29Peter200lxOn Dialup
02:32:37Peter200lxstill downloading
02:34:54Peter200lxRight now I have Rockbox 2.4 in the memory
02:35:29Peter200lxWith 2.5 It's still ok to just run the rockbox.ucl file?
02:35:48webguest56I have no idea :)
02:35:54webguest56I have rockbox for iriver
02:36:06webguest56But, I read somewhere in the forums about this
02:36:32webguest56what kind of player do you have?
02:36:55Peter200lxRockbox Recorder 20GB
02:37:20Peter200lxI use it for Recording our church Sermons
02:37:21webguest56look there, I think that is the way to do it
02:39:36webguest56according to this post in the forums
02:39:57webguest56The wininstaller seems to be updated.
02:40:18webguest56That would be the last post in that thread
02:42:20Peter200lxWorking great
02:43:16webguest56good night
02:43:19Peter200lxhehe Now I can toy around with the latest version
02:43:27 Part webguest56
02:43:42 Quit Peter200lx ("CGI:IRC")
02:51:11 Quit t0mas (" HydraIRC -> <- IRC with a difference")
02:52:16 Quit Sucka ("a bird in the bush is worth two in your house")
03:30:21 Quit ender` (Read error: 113 (No route to host))
04:03:31 Quit JoeBorn (Nick collision from services.)
04:04:56 Join jborn_ [0] (
04:05:20 Join QT_ [0] (i=as@madwifi/users/area51)
04:14:37***Saving seen data "./dancer.seen"
04:17:13 Quit QT (Read error: 110 (Connection timed out))
04:45:13 Quit cYmen ("zZz")
05:35:53 Nick paugh is now known as AliasCoffee (n=kickback@2001:5c0:8fff:ffff:8000:0:3e03:6822)
06:10:22 Join Coldtoast [0] (
06:10:35ColdtoastI found a prob with rockbox
06:11:30Coldtoastif you select a file Rockbox doesn't support and you get an illegal instruction and Rockbox crashes, if you have Autoplay enabled, when it reboots, it tries to play the file
06:11:37Coldtoastyou get stuck in an endless loop
06:12:21Coldtoastwell, if you select any file that makesRockbox crash, actually
06:14:41***Saving seen data "./dancer.seen"
06:22:33Coldtoastwhat's weird?
06:32:27 Quit webguest80 ("CGI:IRC (EOF)")
07:47:27 Quit Coldtoast ("Peace and Protection 4.22")
08:14:45***Saving seen data "./dancer.seen"
08:53:29 Quit ashridah (Read error: 110 (Connection timed out))
09:00:34 Quit Vladoman (Read error: 110 (Connection timed out))
09:03:14 Join Vladoman [0] (
09:44:14 Join ender` [0] (i=ychat@
09:46:03 Join ghode|afk [0] (
09:49:46 Quit Maxime` ()
09:52:25 Join Maxime [0] (
09:55:36 Join [AD]Ska [0] (
09:55:57 Nick Maxime is now known as Maxime` (
09:56:28 Part [AD]Ska
10:14:48***Saving seen data "./dancer.seen"
10:50:19 Join ashridah [0] (
11:14:55 Join Lear [0] (
11:30:00 Join webguest10 [0] (
11:50:08 Join DangerousDan [0] (
11:52:15 Quit webguest10 ("CGI:IRC")
12:06:27 Quit Zagor ("Client exiting")
12:06:35 Join Zagor [0] (i=foobar@pdpc/supporter/sustaining/Zagor)
12:14:51***Saving seen data "./dancer.seen"
12:45:20 Nick ashridah is now known as Lost-ash (
12:57:45 Join cYmen [0] (
13:09:01 Join fogcat [0] (
13:11:59fogcatHello all - it was suggested by Nobby on the forum that I draw attention to a bug
13:12:06fogcatthe details are at
13:12:06fogcatOccasional I get distortion and a staic like burst on playback
13:12:06fogcatIt seems to be when I am using a config which includes speech (spoken menus and dir/track names
13:12:06DBUGEnqueued KICK fogcat
13:12:06fogcatAt first I thought it was only when switching configs but has now happened after "cold boot"
13:12:06fogcatSwitching off and on again clears it.
13:12:29fogcatI'll leave it here for when some one reads the log - ta
13:12:33 Part fogcat
13:29:13 Quit Lear ("Chatzilla [Firefox 1.4/undefined]")
13:36:06 Quit AliasCoffee ("Leaving")
13:38:44 Join rasher [0] (
13:47:44 Join webguest18 [0] (
13:49:53 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
14:13:14 Nick Lost-ash is now known as ashridah (
14:14:54***Saving seen data "./dancer.seen"
14:29:36 Join ]RowaN[ [0] (
14:29:52]RowaN[is it possible to have battery indicator on the USB screen?
14:32:04rasherOf you have the statusbar on, it's already there
14:32:21]RowaN[oh i see
14:38:00]RowaN[im sure someone has already mentioned that it would be cool to be able to disable the statusbar on WPS only =]
14:43:12linuxstb]RowaN[: I think the suggested solution is to add a WPS tag to disable the status bar.
14:43:48rasherthere's a patch to do just that
14:45:42]RowaN[sounds good
14:52:47HClwhy isn't it committed?
15:15:42rasherlack of interest
15:18:30rasher is the patch
15:28:42CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
15:28:42*linuxstb thinks he has just got lucky with the new ipod LCD...
15:34:23linuxstbI change the "lcd_base" value in the driver for the "old" colour LCD to a new (similar) value which I saw being referenced in the new firmware. Now the LCD goes from doing nothing to displaying nice coloured squares (what my test code should be doing).
15:34:46rasherthat's pretty great
15:34:51Maxime`plan to port rockbox on ipod now? o_O lol
15:35:11linuxstbMaxime`: I bought an iPod a couple of weeks ago for that very purpose.
15:35:21linuxstbBut I don't know how far I'll be able to go by myself.
15:35:31rasherlinuxstb: I'm guessing the ipodlinux guys would be "interested"
15:35:37Maxime`a way to "improve" ipod :p great :p
15:36:19linuxstbrasher: If I can get the LCD working, then yes. Hopefully it's not a false alarm - I need to continue playing.
15:38:16rasherWell, how could it be a false alarm if it's actually drawing the right stuff?
15:38:51linuxstbIt's not quite drawing the right stuff, but it's drawing something - which is more than it has done for the past week.
15:42:30rasherAh, okay
15:43:03rasher"displaying nice coloured squares (what my test code should be doing)" just sounded like complete success to me
15:46:16linuxstbI think I know what's going on now, and it's not what I said before.
15:47:31linuxstbThere are commands to tell the LCD which area you are about to update. Those commands don't work the same with the new LCD. But the format and the destination of the data you send to the LCD seems to be the same.
15:48:40linuxstbBut just ignore my ramblings for now....
15:49:41 Join Slyck [0] (
16:04:53linuxstbI now seem to be able to reliably draw in a 78x100 pixels rectangle in the middle of the iPod's LCD - this is obviously the "update region" set by the Apple firmware when displaying the apple logo on the middle of the screen. So the only problem with the new LCDs seem to be with setting the update region.
16:08:23rasherlinuxstb: that doesn't sound too bad
16:10:00linuxstbYes - I'm happy that I can finally make the LCD do something.
16:12:31 Quit Strath (Read error: 104 (Connection reset by peer))
16:13:17 Join Strath [0] (
16:14:57***Saving seen data "./dancer.seen"
16:25:08 Quit linuxstb ("Leaving")
16:31:31 Join linuxstb [0] (
16:37:20 Join Coldtoast [0] (
16:37:29Coldtoasthey all
16:39:35ColdtoastI'll mention an issue I ran into today. came in before but ppl were afk
16:40:00ColdtoastI went to load something which caused Rockbox to crash with an illegal instruction
16:40:41HClon what player?
16:41:14Coldtoastand had to reset Rockbox. Problem is I had Resume Play on Startup enabled so every time Rockbox booted, it'd try and play the track again
16:41:16Coldtoastendless loop
16:41:41 Join webguest66 [0] (
16:42:16 Quit webguest66 (Client Quit)
16:43:50Coldtoastif you're not near a computer, you have no choice but to boot the iriver firmware til you get access to a machine where you can delete the offending track
16:46:16Coldtoastso it's not a bug, it's just a situation that (possibly) nobody's run into so hasn't been considered
16:46:29rasherHCl: could you upload makedbj.bat and SongDB.jar to the wiki?
16:46:36ashridahColdtoast: uh, the iriver firmware can delete files thesedays, you remember
16:48:05ashridah(not that this fixes the bug in rockbox, natch)
16:48:53Coldtoastcan it?
16:49:03ashridahyeah. don't remember exactly how
16:49:09Coldtoastbeen so long since I used it
16:49:15ashridahbut they added the deletion of files in a firmware revision ages back
16:49:48Coldtoastwhen I ran into the prob, I'd just finished putting some stuff on my player so was at my machine
16:51:46Coldtoastsomething else that's sort of a bit of a bother to me personally is when I power my player on and it DOES resume playing, if I then go into the file tree, it defaults to the root of the player, not the dir the track playing is in
16:53:03Coldtoastwhich is more my personal preference. Probably just a personal thing tho
16:53:06rasherTurn oon "follow playlist"
16:53:25Coldtoastoh? didn't even know there was tha option. thanks
16:53:25rasherThink it's under playback
16:53:34rasherthat should do just that
16:53:42ColdtoastI'll try it
16:54:41Coldtoastit's in File View
16:56:49Coldtoastcool. that did it
16:56:52Coldtoastthanks rasher
16:57:23rasherNo problem
16:58:17Coldtoastthere are still certain things I have to use the iriver fw for, unfortunately. I've not bothered listening to some tracks for ages so I don't have to use the iriver fw
16:58:43Coldtoastjust some music that really needs SRS to thicken the sound up cos of my earphones
17:09:20 Quit ashridah (Read error: 110 (Connection timed out))
17:11:13 Quit Slyck ()
17:15:45 Quit Strath (Read error: 104 (Connection reset by peer))
17:16:58ZagorColdtoast: get new earphones :-)
17:17:52 Join DangerousDan [0] (
17:18:52Coldtoastthat or just keep suig the iriver fw for some things
17:28:22 Join bagawk [0] (i=1000@unaffiliated/bagawk)
17:41:04 Quit bagawk ("Leaving")
17:45:12 Join Lear [0] (
17:48:27 Quit rasher (Remote closed the connection)
18:14:59***Saving seen data "./dancer.seen"
18:19:23 Join Sucka [0] (
18:19:24 Quit ghode|afk (Read error: 104 (Connection reset by peer))
18:20:45 Join ghode|afk [0] (
18:20:51 Join webguest26 [0] (
18:20:56 Quit webguest26 (Client Quit)
18:22:41 Join help [0] (
18:23:53help(maybe i should keep that name :/ )
18:26:25 Quit help (Client Quit)
18:27:14 Join RoadRunner [0] (
18:28:29 Quit RoadRunner (Client Quit)
18:28:38 Join lamedman [0] (
18:28:56 Quit lamedman (Client Quit)
18:29:10 Join lamed [0] (
18:31:20lamedhelp, please. i'm only trying to build one plugin file without compiling the whole thing. i've went through the makefile but i can't understand what should i command to make it compile?
18:34:00 Join tvelocity [0] (
18:36:16 Quit gromit` (Remote closed the connection)
18:38:42 Quit ]RowaN[ (Read error: 110 (Connection timed out))
18:39:29 Join Mxm`Pas`Bien [0] (
18:47:45 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
18:56:22 Quit Maxime` (Read error: 110 (Connection timed out))
19:03:32 Nick Mxm`Pas`Bien is now known as Maxime` (
19:05:06 Join amiconn [0] (
19:08:04 Quit lamed ("CGI:IRC (EOF)")
19:21:17 Join gromit` [0] (
19:22:03 Join dpassen1 [0] (
19:35:11 Join Mxm`Pas`Bien [0] (
19:35:15 Quit Maxime` (Read error: 104 (Connection reset by peer))
19:42:56linuxstblamed: I don't think there is a simple command you can type to just compile a single plugin - it's far easier to just compile the whole of Rockbox.
19:51:48amiconnlinuxstb: Seen the sudoku bug report (sort of) in the forum?
19:53:45linuxstbLiterally just read it now.
19:53:57linuxstbI was aware of the problem trying to solve empty grids.
19:56:12linuxstbI'm not sure if the patch is useful - it's more of a workaround than a solution.
20:03:11amiconnAn edit mode might be useful after all
20:04:00linuxstbIf by "edit mode", you mean a way to manually enter a game coped from (e.g.) a newspaper, then I agree. It's something I want to add myself but haven't got around to.
20:04:16amiconnyes, that's what I mean
20:04:19amiconnI'm thinking about a way to make the difference between fixed and user-selected numbers visible on archos
20:04:52linuxstbCould the greyscale lib be used? Or is that a bad idea?
20:04:53amiconnI could use the grayscale lib for that, but it would rule out the archos sim
20:05:43linuxstbIs it a problem if it doesn't work on the sim?
20:08:02amiconnNot for the user, but maybe for a developer who doesn't have an archos
20:08:36amiconnThat's one of those things that are waiting to get implemented - grayscale lib support for the sims
20:15:02***Saving seen data "./dancer.seen"
20:18:36 Quit tvelocity (
20:18:36 Quit linuxstb (
20:18:36 Quit Zagor (
20:18:36 Quit HCl (
20:18:36 Quit lImbus (
20:18:36 Quit flurble (
20:18:36 Quit atubbs_ (
20:18:36 Quit ze (
20:18:36 Quit void (
20:18:36 Quit dpassen1 (
20:18:36 Quit amiconn (
20:18:36 Quit t0mas (
20:18:36 Quit ghode|afk (
20:18:36 Quit Coldtoast (
20:18:36 Quit webguest18 (
20:18:36 Quit jborn_ (
20:18:36 Quit pike (
20:18:36 Quit Seed (
20:18:36 Quit pabs (
20:18:36 Quit Lear (
20:18:36 Quit DangerousDan (
20:18:36 Quit elinenbe (
20:18:36 Quit Nilisco (
20:18:36 Quit merbanan (
20:18:36 Quit crash__ (
20:18:36 Quit cYmen (
20:18:36 Quit bluebrother^ (
20:18:36 Quit stamppot (
20:18:36 Quit dwihno (
20:22:28NJoindpassen1 [0] (
20:22:28NJoinamiconn [0] (
20:22:28NJoint0mas [0] (n=Tomas@unaffiliated/t0mas)
20:22:28NJointvelocity [0] (
20:22:28NJoinghode|afk [0] (
20:22:28NJoinLear [0] (
20:22:28NJoinDangerousDan [0] (
20:22:28NJoinColdtoast [0] (
20:22:28NJoinlinuxstb [0] (
20:22:28NJoinwebguest18 [0] (
20:22:28NJoincYmen [0] (
20:22:28NJoinZagor [0] (i=foobar@pdpc/supporter/sustaining/Zagor)
20:22:28NJoinjborn_ [0] (
20:22:28NJoinpike [0] (
20:22:28NJoinHCl [0] (
20:22:28NJoinlImbus [0] (
20:22:28NJoinflurble [0] (
20:22:28NJoinbluebrother^ [0] (
20:22:28NJoinSeed [0] (
20:22:28NJoinatubbs_ [0] (
20:22:28NJoinze [0] (
20:22:28NJoinvoid [0] (
20:22:28NJoinelinenbe [0] (
20:22:28NJoinNilisco [0] (
20:22:28NJoinmerbanan [0] (
20:22:28NJoinstamppot [0] (
20:22:28NJoincrash__ [0] (
20:22:28NJoinpabs [0] (
20:22:28NJoindwihno [0] (n=dw@
20:22:57Vladoman /msg NickServ IDENTIFY hgpj41vp
20:25:30 Join Domonoky [0] (
20:26:34DomonokyVladoman, i think you should change your password for irc identify now .. :-)
20:26:44Vladomanyep :-)
20:29:27Domonokychange it with /msg Nickserv set password <newpass>
20:29:55 Quit Lear ("Chatzilla [Firefox 1.4/undefined]")
20:30:28Vladomandone, thx
20:30:57Coldtoasttee hee
20:31:16 Join solexx_ [0] (
20:36:25 Nick Mxm`Pas`Bien is now known as Maxime` (
20:47:05 Quit solexx (Read error: 110 (Connection timed out))
20:48:51 Join muesli- [0] (
21:00:48linuxstbAnyone have any ideas about what to do with the ReleaseTodo Wiki page?
21:00:51linuxstbChange the title to "things to do before 2.6" ?
21:24:04 Join preglow [0] (
21:25:46 Join xen` [0] (n=xen@
21:27:13 Join Mxm`Pas`Bien [0] (
21:27:38preglowahh, cvs activity
21:28:09linuxstbIt's been a long time...
21:29:55 Nick TiMiD[FD] is now known as TiMiD (n=TiMiD[
21:30:07TiMiDany devs here ?
21:36:46TiMiDI would like to talk about remote support implementation
21:36:51TiMiDif someone can
21:41:31 Quit preglow ("CGI:IRC (EOF)")
21:41:57 Join preglow [0] (
21:42:30preglowwhat about remote support?
21:43:06 Quit Maxime` (Read error: 110 (Connection timed out))
21:46:33 Quit gromit` (Remote closed the connection)
21:51:32 Join gromit` [0] (n=gromit`
21:54:26TiMiDI would like to implement remote support in rockbox
21:55:06TiMiDbut I wanted to discuss the way to do it since there are some possibilites
21:55:52TiMiDfor example a single thread or multi-threads
21:56:05Zagorsingle, definitely
21:56:20TiMiD(this was proposed on the mailing list)
21:59:29TiMiDso if it's single thread, the patch xavier made was adapted (some bugs apart)
22:00:38Zagorwell the problem is of course the different screen sizes. so you can't show exactly the same thing.
22:01:06Zagorafaik xaviers patch was on the graphics level. it needs to be handled higher up, on the application level.
22:01:07Coldtoastthere's no need to anyway
22:01:56ZagorColdtoast: ?
22:02:11Coldtoastwhy would you need to show the same amount of info?
22:02:16TiMiDmultithread I think
22:02:16Coldtoastthe iriver fw doesn't
22:02:27ZagorColdtoast: you don't want to, and you can't
22:02:43ZagorTiMiD: why? multithread will give you mountains of problems and very little gain.
22:03:02Coldtoastwill we eventually be able to write .wps for the remote?
22:03:28Coldtoastis that the eventual aim?
22:03:51Zagorif you wanted to do a very quick solution, slightly better than xavier but still not all the way, then toggle the screens. if remote is plugged, only show on remote. that means the application only has to handle one screen size and everything becomes very simple.
22:03:58Zagorbut it's not The Right Way, of course
22:04:07ZagorColdtoast: yes
22:04:32TiMiDnope I want the two screens activated :)
22:04:54TiMiDmultithread would be a little more heavy
22:05:02TiMiDbut on iriver there is sufficient memory
22:05:21Zagormemory is not the issue. code complexity is.
22:05:24TiMiD(I don't think the tree_context can become that huge
22:05:32Zagorwhat will you gain by running to threads?
22:06:03TiMiDfor me it would be very convenient
22:06:31TiMiDto be able to control my music on the remote while doing other things on the main display
22:06:56ZagorTiMiD: yes, user-wise it might be a good idea. but I think you should look at bit at the code too... :-)
22:07:37TiMiDsome globals variables apart are there any other problems ?
22:08:12 Quit preglow ("CGI:IRC (Ping timeout)")
22:08:58TiMiD(I dont know how multi thread is handled on rockbox)
22:10:19TiMiDin my opinion, doing it single thread would be similar than what xavier did
22:12:03TiMiDjust to handle a different wps on the remote which can mean code duplication and handling cutting the things to display on main screen (since you have to display them on the main screen, no need to keep 2 contexts)
22:12:36TiMiDfor the tree and menus
22:15:06***Saving seen data "./dancer.seen"
22:15:40 Join polytan [0] (
22:16:21TiMiDhi polytan
22:16:21polytanrockbox can't have an ogg suport for archos mp3 readers ?
22:16:27polytanhi TiMiD
22:16:58ZagorTiMiD: the entire GUI code is written to run in one thread. variables are global. memory is shared between screens etc.
22:17:07TiMiDit is explained on the wiki (and the answer is no :-( )
22:17:08Zagorpolytan: no
22:17:23 Join rcohen [0] (
22:17:34ZagorTiMiD: of course it CAN be done. but should it?
22:17:34polytanit's an hardware problem i think, isn't it ?
22:17:39Zagorpolytan: yes
22:17:55TiMiDmemory is shared between screens ?
22:18:29TiMiDgraphic buffer oO ??
22:21:43rcohenI'm having an issue with fwpatcher, when i run it against the "H120 and H140 1.65US" firmware linked from the IriverBoot page in the wiki it says "unrecognized firmware"
22:22:05rcohenfwpatcher.exe is from that page as well
22:23:01TiMiDrcohen: try this one :
22:23:04 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
22:23:15TiMiDit's a little old, but it worked for me
22:24:42rcohennope, same deal
22:25:16TiMiDI upload you my patched formware (if I can find it)-___-
22:26:43rcohenaha, if i use the 163 firmware it works
22:26:49TiMiDit should be there (when the upload @15k/sec will be finished)
22:27:21rcohenthe wiki really shouldn't be giving instructions which don't work, though
22:28:19rcohenit links to firmware which fwpatcher.exe refuses to work on
22:29:00TiMiDZagor: ok I understood : wps, file tree etc is running in one single program -______-
22:29:44TiMiDI didn't understood that
22:29:47ColdtoastI just downloaded the fwpatcher and 1.65US and ran it
22:29:50Coldtoastand it works perfectly
22:29:52TiMiDso it's definitively single thread :)
22:29:54Zagoraha, that explains it :-)
22:30:00Zagoryes it is
22:30:00Coldtoastso don't blame the wiki
22:30:14Coldtoastthere's a prob somewhere on your end
22:30:25Coldtoastyou ARE extracting the firmware from teh zip?
22:30:53Coldtoastnot, like, renaming the ip to .hex or something odd?
22:31:01rcohenit might be an issue with wine
22:31:17Coldtoastsorry. I'm runnign native XP
22:31:34Coldtoastdidn't realise you were using WINE
22:31:41Coldtoasthere you go tho. I juat patched it if you want it
22:32:07rcoheni'm behind NAT, file transfers don't work so well
22:32:35 Join davidd [0] (i=davidd@
22:33:04rcohenthanks, though
22:33:16Coldtoastit'll be a WINE issue, anyway
22:33:45Coldtoastyou'd have access to a Windows machine somewhere tho shurely?
22:34:31linuxstbrcohen: It's not very difficult to patch the firmware "the hard way" from the command-line if you're using Linux.
22:34:37rcohennot unless i commandeered one of my roommates' machines
22:34:42rcohenall of mine run linux
22:34:44 Quit rcohen (Remote closed the connection)
22:35:45 Join rcohen [0] (
22:35:53rcohensorry, gaim crashed on me
22:36:35rcoheni went through the manual process of patching firmware on linux, but the bootloader.bin in the wiki is version 2 and it looks like v5 is what fwpatcher uses
22:37:04rcohenplus none of the md5sums matched up and i'm paranoid
22:38:05linuxstbIt looks like bootloader.bin needs updating on the wiki page. I don't think Linus is around at the moment though.
22:38:59rcohenplus i'm too lazy to get a cross-compiler installed
22:41:05linuxstbI think I've got the v5 bootloader.bin I downloaded from the Wiki - let me check it creates a firmware matching the published MD5 checksums
22:43:35linuxstbYes - I used it to patch a 1.65EU firmware, and get the same checksum as on the wiki. You can grab it from here:
22:44:03linuxstbThat's the v5 bootloader I grabbed from the Wiki on 9th September.
22:44:29 Join webguest53 [0] (
22:44:42webguest53looks like some wiki spam on IriverPort
22:45:23webguest53I do not think AffordableBedroomFurniture has a lot to do with rockbox ;)
22:46:03HCl :p
22:46:21muesli-think ikea
22:47:05webguest53IDA Pro is a very powerful disassembler which Affordable Bedroom Furniturealso supports ... ;(
22:47:48 Part webguest53
22:53:06rcohenlinuxstb: my md5sum isn't matching what's on the wiki: 165US + bootloader.bin you gave me
22:53:26lImbusHCl, yre you there ?
22:53:28rcohenso, just how dangerous would it be for me to try and use this?
22:53:56lImbusHCl: I am getting some perl-crashes with with my tagged files
22:54:00Coldtoastif you killed your player, you'd be the first one to do it
22:54:11rcohensome consolation that would be :)
22:54:16linuxstbrcohen: What's the md5sum of bootloader.bin ?
22:54:22rcohenhow does the patching work?
22:54:46linuxstbThe file I uploaded for you was: 1e45c54383f8461b0e5724d03fc8cc58 bootloader-h120.bin
22:55:04HCllImbus: perl crashes?
22:55:07linuxstbrcohen: What if you try the 1.65 EU firmware?
22:55:14HCllike what? o-o
22:55:17rcohengimme a minute to download it
22:55:40lImbusHCl: looks like files with german umlauts within title tags are not well seens. do you want such a file ?
22:55:50lImbusit happens while scanning the file
22:56:09HClwell, can you paste the crash in privmsg?
22:56:28lImbuswell, it's a crash on windows, nothing special to see
22:56:32lImbusperl.exe crashes
22:56:46lImbusbut I can send you the mp3file per mail
22:56:48HClask bagder, i'd say.
22:56:54lImbusthere is a nother crash at the end of the procedure, which I could not yet track down to a single file
22:56:56HCli'm mostly responsible for the java tool
22:57:05HClall i did was alter the perl tool to work with the new database format
22:57:07lImbusaha. yes, I remember
22:57:38rcohenah, there we go
22:57:53rcoheni've got too many zip files around here with various firmware versions
22:58:04 Join webguest61 [0] (
22:58:05rcohenthat first one much have been the wrong one
22:58:31HCllImbus: try the java tool instead?
22:58:55lImbusyes, that's an idea
22:58:59lImbusthe wiki page is complete ?
22:59:04lImbusgonna try that then
22:59:18amiconnZagor: While I'm not that interested in getting the remote working for myself, I do think that a multithreaded approach would be way better
22:59:28amiconn...and still not too complex to handle
22:59:51 Quit muesli- ("ich will Kühe!!!")
23:02:26amiconnOf course there are situations where the gui threads would need to lock things and can't work independently
23:04:17amiconnTwo situations come to mind: (1) playback and recording are mutually exclusive. (2) Browsing different directories is also not possible if we don't want to duplicate the buffer
23:06:27amiconnTwo gui threads would e.g. allow to use the menu or run a plugin or browse for new music on one display, while the other continues showing the wps and allowing playback control
23:07:04amiconnDoing this kind of stuff would be about the only reason to actually use the remote for me
23:07:39ZagorI get a headache just thinking about it ;)
23:08:02Zagormaybe I'm pessimistic, but I think we'll run into quite a few areas of trouble.
23:08:14amiconnI don't think it would be overly complicated
23:09:23amiconnI don't know all areas that well, but from what I've seen the gui for a task is somewhat independent from the engine doing the actual work
23:10:00amiconnFor example, the recording screen just sends messages and reads back status. The mpeg thread does the dirty work
23:10:59 Quit webguest61 ("CGI:IRC (EOF)")
23:11:01Zagoryes. the browser is the biggest worry, i think
23:11:56Zagorbut even the mpeg thread needs reworking, to know which thread to send messages to. and what if both threads need it?
23:12:21amiconnThe mpeg thread doesn't send messages to other threads, it only receives them
23:12:36amiconnShouldn't matter from which thread they are sent
23:12:57amiconnStatus reading is done by calling functions from the recording screen
23:13:17Zagorah, right. i've been away too long. :-)
23:13:22amiconnI think the browser problem can be solved. There are 2 situations:
23:13:49amiconn(Starting with the browser being displayed on one screen, no matter which one)
23:14:17 Part rcohen
23:14:51amiconn(1) You try enter the main browser on the second screen. In that case it should simply show the same directory as the first (not necessarily the same position in the list)
23:15:53amiconn(2) You try to enter a sub-browser from the menu (fonts or such). In that case it should refuse to enter the sub-browser if entering it would imply to change the dir
23:16:32TiMiDwhy should current dir be the same for the two threads ?
23:16:49ZagorTiMiD: otherwise we need do double the dir buffer
23:17:55TiMiDwould that be a problem ? I mean the dir buffer doesn't seems to be that huge (aprt the case when you have 9999 files in the same dir, but ...)
23:17:57amiconnI would like to avoid that, because it would have a negative effect when not using the remote
23:18:17amiconn(1) If we reserve the buffer statically, it would be wasted
23:18:19TiMiDwhen not using the remote you could simply kill the thread
23:18:36amiconn(2) Dynamic allocation would really complicate things
23:19:06amiconnTiMiD: The thread isn't the problem, the allocated memory is
23:19:10ZagorI still think it would be much simpler to just draw each screen twice
23:19:26amiconnAnd btw, I wouldn't kill the thread, I would just let it idle
23:19:28 Quit polytan ("Leaving")
23:20:17Zagorwhat you guys consider a feature (different things on remote and main), I think a lot of other people will find annoying.
23:20:21amiconnZagor: Maybe simpler, yes, but way less useful (imho)
23:20:29TiMiDhow much big is the dir buffer (approximately) ?
23:20:29Zagorand I personally find it pointless
23:20:49TiMiDin normal file structures conditions
23:20:54Zagoramiconn: what's so much better browsing AND showing wps at the same time?
23:21:19amiconnI don't see the point of the remote lcd when it just mirrors the main lcd content
23:21:25linuxstbI agree with Zagor - I think users will be using the main device OR the remote. Not both.
23:21:36amiconnThe I don't need a remote, or one with just buttons
23:21:39Zagorthe point of the remote is that you don't see the main because it's in your bag/pocket
23:22:02HCli agree with amiconn.
23:22:11TiMiDthe remote could be used when another app is taking the main screen
23:22:12HClits highly preferable to have them independant.
23:22:20 Quit davidd ()
23:22:20TiMiDlike game or text viewer
23:22:23ZagorHCl: why?
23:22:42HClbecause it allows you to do two things at once
23:22:46Zagorsuch as?
23:22:47HClwatch the wps, and browse for a new file.
23:22:50HClplay a game
23:22:50linuxstbI would prefer that we were about to "task-switch" on the main device to go between different screens in a more natural way.
23:22:51HCland watch the wps
23:23:28Zagoryou can't watch the wps and browse at the same time. you only have one pair of eyes. swapping between the displays is just like swapping between the screens today.
23:23:33amiconnZagor: e.g playing music (and being able to adjust volume/ skip certain tracks etc) and running a game plugin
23:23:41 Quit ze (Read error: 110 (Connection timed out))
23:23:46HCli disagree
23:23:54HClyou can switch focus much faster
23:23:56HClthan you can switch modes
23:24:13Zagoreh? it's instantaneous
23:24:23amiconnYou can't do that atm, because for adjusting anything playback-related, you have to leave the plugin
23:24:29amiconn...losing status etc
23:25:01Zagoramiconn: that's the first valid reason i've heard yet
23:26:01amiconnOf course, being able to cycle through screens would solve this problem, also for units without an lcd remote
23:26:34amiconn...but imho cycling through screens calls for even more problems than having 2 gui threads
23:26:53 Join rasher [0] (
23:27:07linuxstbBut it would be very nice to be able to do it.
23:29:47amiconnHaving 2 gui threads would simply run the plugin in the main gui thread, nothing would change for the plugin compared to now
23:30:03amiconn(except the few plugins that use the remote now)
23:30:29ZagorI don't think it's worth it. I think this is yet another "wouldn't it be cool if..." things.
23:30:59amiconnScreen cycling would open a whole can of problems, dealing with which gui part has focus etc
23:31:08Zagorwe'd work for weeks getting it right, then five people would use it twice each.
23:31:46Zagormost people simply don't use the player like that. they use either the remote OR the main. not both.
23:32:48webguest18well it makes no sense to use both atm, doesn't it?
23:33:35amiconnHmm. I don't use the remote at all atm. For simple tasks I don't need to look at the display, and for things I do need a display it doesn't make a difference whether I grab the remote or the main unit
23:33:41Zagorthat's not my point. the point is I think people only WANT to use one screen. if you have the main in your hand, you don't want to fiddle with the remote.
23:34:15Zagoramiconn: I interpret that as supporting my point :-) I'm better off not carrying around an extra piece of hardware
23:35:01amiconnIf I could actually do more with the remote plugged, I would probably actually use it
23:35:11webguest18amiconn: in some regions of the world taking your whole player out of your pocket might be risky whilst a quick peek at the remote can go totally unnoticed
23:35:30webguest18like here were i live :/
23:35:37TiMiDwebguest18: me too :/
23:35:49amiconnI'm not using the iriver within a pocket. I doubt that it does fit a pocket, it's too bulky
23:36:01dpassen1fits in all my pockets
23:36:04TiMiDI put it in my bag
23:36:11webguest18it's not the point.
23:37:21webguest18the point is, you're on shuffle and all of a sudden - OMG how cool dat tune is wtf is this i wanna know now! - you take out the player and bam, you have no player and a broken nose, you take a look at the remote smartly hidden under your jacket and voila ;)
23:37:25webguest18ghetto tactics
23:38:13linuxstbwebguest18: No-one is arguing that the remote isn't useful, or that Rockbox shouldn't support it at all.
23:38:20amiconnHmm. Can't say much about this kind of problems...
23:38:25 Join Strath [0] (
23:38:46linuxstbI think the only debate is if the remote and main display should display the same screen or not.
23:39:23webguest18imho - head to head gaming on main unit vs remote would be the ultimate in geeky coolness
23:39:47linuxstbA plugin could easily do that today (I think).
23:39:58amiconnThat would be an additional gimmick, but has nothing to do with how the core handles the remote
23:40:03Zagorin fact it would have a problem doing it with two threads
23:40:18webguest18or changing the wps on the remote and watching the immediate change on the main unit ;o
23:40:19amiconnZagor: I don't think so
23:40:43amiconnwebguest18: A bit unfair though, depending on the game
23:40:52webguest18amiconn :)
23:40:53Zagoramiconn: either one thread takes over both displays or they have to communicate. either way is unclean.
23:41:23amiconnThe remote lcd is clearly smaller, but otoh it's definitely faster
23:42:01amiconnZagor: Usually each thread would handle one display. A plugin could request control over the remote lcd from the remote gui thread
23:42:17amiconn(remote lcd + buttons)
23:43:15 Join ze [0] (
23:43:17TiMiDkevin@arrakis /mnt/panoramix/download/iriver/rockbox/apps $ grep -R extern *|grep -v '('|grep -v README|wc
23:43:21TiMiD 225 1056 15104
23:43:27TiMiDthat the main problem :(
23:44:32amiconnOnly slightly related: I'm curious whether anyone actually tried the 3-player mode of wormlet.rock on archos recorder ;)
23:44:33TiMiDdoing it multithread would proobably imply rewriting a lot of code
23:45:22Zagor<Zagor> we'd work for weeks getting it right, then five people would use it twice each.
23:45:57rasheramiconn: that sounds .. cramped
23:45:59amiconnDoing content mirroring in a clean way will also imply rewriting a lot of code
23:46:35amiconnrasher: Two players using the buttons on the main unit, the third using the remote. All looking on the main display of course ;)
23:46:39Zagoramiconn: not nearly as much
23:46:56amiconnI doub't that the difference would be that big
23:46:57Zagorwe already support dynamic sizing in most screens (since we have variable size fonts)
23:47:53amiconnYes, but we don't have a way to handle a different number of lines between two displays yet
23:48:11amiconnHow would scrolling up/down be handled?
23:48:31ZagorI thought about that yesterday. it's actually quite simple.
23:48:44amiconnUsing one thread adds the problem that there are two keypads that can fire events
23:49:11amiconn...which require different actions in the code
23:49:12Zagorsimply adjust the dirstart (in case of tree.c) if it is outside the screen.
23:49:40Zagorthis means we don't have to keep track of dual positions/lines
23:49:55amiconnWe do, definitely
23:50:43amiconnThe position within the list would be the same, but the start of the list would not
23:50:57Zagorare you reading my lines?
23:51:12Zagor<Zagor> simply adjust the dirstart (in case of tree.c) if it is outside the screen.
23:51:15rasherZagor: I just uploaded bootloader-h120.bin and bootloader-h100.bin to the wiki - could you md5sum them for me?
23:51:25Zagorrasher: ok
23:51:46rasherI already checked that they produce valid patched firmwares
23:51:49amiconnyes, and that needs s to be done for both screens independently
23:51:54rasherJust need to check that the uploads went fine.
23:52:41Zagoramiconn: dynamically, by the draw routine. stateless.
23:52:59amiconnI'm still not convinced that mirroring is the way to go
23:53:59Zagor05d368b62f484904d4afac2064b0f4f2 bootloader-h100-20050722.bin
23:53:59Zagor1e45c54383f8461b0e5724d03fc8cc58 bootloader-h120-20050722.bin
23:54:00amiconnZagor: It's not stateless, you'd need to store dirstart per screen.
23:54:10rasherZagor: thanks, matched.
23:54:11amiconnOtherwise scrolling would look really odd
23:54:31amiconnMirroring just looks like an incomplete, cheap solution to me
23:55:00Zagormultithreading looks like overworked unnecessary bloat to me :-)
23:55:07linuxstbrasher: And as a double-check, that h120 checksum is the same as my local copy of the v5 bootloader as well.
23:55:40rasherlinuxstb: alright, I think it's safe to say it's the right file
23:56:02linuxstbYou can't be too safe with those bootloaders.
23:56:06rasherCertainly not
23:56:15Zagoramiconn: why would scrolling look strange?
23:56:46rasherwow, #ipl is like kindergarten
23:57:15amiconnYou need to keep dirstart the same _per screen_ until the cursor leaves at the top or bottom, _then_ adjust
23:57:28Zagoramiconn: why?
23:57:31amiconnYOu can't do this without saving the status per display
23:57:40linuxstbrasher: You don't have to tell me that. Buit I did have a useful conversation with leachbj earlier.
23:58:06Zagorthe cursor is what's important. dirstart should be adjusted so the cursor is visible
23:58:08amiconnZagor: Well, for the same reason as we do it now (for one display)
23:58:22Zagorthis can be done dynamically, without storing
23:58:27rasherlinuxstb: Yeah, I don't doubt the real people are also there, but I guess running linux on ipod just attracts the wrong crowd
23:58:29amiconnNo, it can't
23:58:37Zagoryes it can, no it cant :-)
23:58:58linuxstbrasher: I'm not sure who the crowd are - they don't talk about Linux or the iPod...

Previous day | Next day