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 2005-09-10

00:03:52 Quit pixelim ("Leaving")
00:06:40***Saving seen data "./dancer.seen"
00:10:20 Nick paugh is now known as AliasCoffee (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
00:10:56amiconnBagder: Nice :)
00:14:10 Quit ender` (Read error: 113 (No route to host))
00:15:35flurbleso if ~/rockbox$ md5sum new.hex
00:15:43flurbledoesn't look like any of the ones on the wiki page
00:15:48flurblewould it be a bad idea to do anything with it?
00:16:02bluebrother^trash it and try again :)
00:16:44flurblehm, and if it comes out the same again?
00:18:51 Join rasher [0] (
00:19:02rasherflurble: which bootloader are you patching with?
00:19:16rasherAnd which firmware are you patching?
00:21:48flurblethat one's better
00:22:16flurblewith a different (presumably the right one this time) bootloader
00:22:38bluebrother^I just tried it with the files from the wiki without any problems.
00:23:34 Join EEzde [0] (
00:23:43bluebrother^h120, 1.65eu.
00:24:03 Quit Moos (" HydraIRC -> <- State of the art IRC")
00:24:06linuxstb__I've also just tried it with the H120/H140 1.65-EU firmware (alt. download site) and it worked perfectly. This is doing it the "hard way" using the command-line tools under Linux.
00:24:35EEzdetach auch
00:24:52linuxstb__I'm still using v1 of the bootloader, so it's about time I upgraded.
00:25:21rasherIsn't that unsafe?
00:25:33rasher(no usb mode)
00:25:39rasherOr was that even earlier
00:25:57 Join pike| [0] (
00:26:47linuxstb__Yes, I don't think there's a USB mode. But I've had no problems at all with it.
00:27:04 Join webguest02 [0] (
00:27:25rasherA broken filesystem would've left you with a non-booting system
00:27:37amiconnbluebrother^: Want to try something for the volume change problem?
00:27:52linuxstb__As I said, it's about time I upgraded :). I'm doing it now.
00:29:36amiconnbluebrother^: In firmware/drivers/uda1380.c, locate line 228 (within uda1380_enable_recording())
00:29:57amiconnChange the MIX_MODE(3) into MIX_MODE(1). Imho it should help...
00:30:09rasherI'm thinking I may have some strings that are too long in dansk.lang. I find it hard to tell which strings have a size-limit
00:30:40amiconnrasher: Yes. Guess why I'm bothering Bagder about localisation v2 all the time? ;)
00:31:15amiconnThe comments should clearly state where these strings are used, and what the length limit is, per platform if necessary
00:31:16flurblewoot! i installed it! how exciting!
00:31:33rasherI agree
00:31:44rasherThough that could've been done with the current system as well
00:31:55amiconnYes, but really cumbersome
00:32:01rasherNot as elegantly
00:32:34rasherLooking forward to v2
00:32:36 Join edx [0] (
00:32:50bluebrother^amiconn: didn't work for me :(
00:32:52rasher(and the v1->v2 script that makes notes of where to pay attention)
00:33:29amiconnbluebrother^: Mhm, I just tried it here. Same result. Strange.
00:34:11amiconnrasher: I'm afraid we have to adjust a number of strings manually
00:34:26amiconnLike, the current system sometimes has extra strings for the player
00:35:13amiconnThese should be merged (ID-wise) with the strings for the other platforms, since it is possible to specify per-platform strings for an ID with v2
00:35:31rasherOf course there'll be a lot of manual editing
00:36:10amiconnThere are also some strings that should be merged in general. Sometimes there are 2 strings for one message, just because back in the old rockbox days there was no splash() function
00:36:28rasherI noticed a few of those
00:36:28amiconn...and the message should be displayed in 2 lines
00:36:55rasherLooks absolutely revolting on iriver
00:37:02amiconnI encountered one more of these just a short time ago
00:37:07rasherBookmarking code has some iirc
00:37:33amiconn(and did a temporary fix by joining them in one splash())
00:37:49amiconnsplash(HZ, true, "%s %s", string1, string2);
00:37:59amiconnNot exactly elegant...
00:38:57rasherWorks for the user
00:39:19amiconnBagder: What do you think, should we do an incremental cleanup of all IDs *after* commiting the v2 system?
00:40:02amiconnIt will break intermediate language files, but we will probably end up with rather clean files
00:40:24amiconnOf course we should "freeze" the IDs again at some point
00:40:43rasherMy idea is that after v2 is introduced, only english should be available until it's cleaned up
00:42:00 Quit pike (Read error: 110 (Connection timed out))
00:42:13amiconnI think the work on other languages can be done in parallel - since the IDs are "linked" to english there's no big risk of messing up
00:43:12amiconnThe language files have a version number. This should be bumped 2 times, once when committing v2 and again when the cleanup is done
00:44:28 Join TCK [0] (
00:47:03 Join stripwax [0] (
00:59:38 Join bagawk [0] (i=1000@unaffiliated/bagawk)
01:05:00stripwaxis svante online ? ;-)
01:05:46rasherSeems not
01:05:57*stripwax sighs
01:06:53rasherEither that, or he stopped responding. I'm not holding my breath on that one though.
01:07:31 Quit DangerousDan (Read error: 104 (Connection reset by peer))
01:07:38stripwaxI think I've mashed my reset button on my iriver too much, and now it won't actually turn on *unless* I've got some kind of metal doodad in the reset hole ;-( any ideas?
01:09:18stripwaxshould i just take it apart and assess the damage?
01:09:25rasherOpen it up and fiddle with the reset button
01:09:43rasherI seem to recall other people being successful in doing so
01:10:27stripwaxokie doke. i'm suspicious that reset button might actually have come off .. guess I'll find out in the morning, way too drunk right now to try something like that
01:11:27rasherMy guess is that it's the little metal-thingy
01:11:52stripwaxthose thingys. they always break, darn.
01:14:08rasherI'm not sure why I called it a metal thingy.. don't know what it's made of
01:14:13rasherThe actual button
01:15:18stripwaxah- yes, that's probably it. i guess I could solder on a new one if necessary
01:16:38 Join gromit` [0] (n=gromit`
01:16:52rasherno idea what it looks like on the inside
01:17:03rasherForgot to check it when I changed the battery
01:17:52stripwaxyeah, maybe I'll upgrade the batt and hd at the same time.. anyone found those mk4007gal's in stores yet?
01:20:20rasherNo idea
01:23:13*rasher stares disillusioned at the ReleaseTodo
01:25:37flurblemy iRiver with their firmware always refused to play anything I'd recorded off a tape originally, and it's playing fine now, is that something you can put on your feature list? Never worked out why it wouldn't play though so I don't know what the feature is...
01:26:09 Quit matsl (Remote closed the connection)
01:26:10rasherHeh, what kind of file is it?
01:26:18rasherHow was it recorded?
01:27:03flurblethe obvious difference between them and most of my files was that they were recorded at 8kHz instead of 44.1, I guess
01:27:19flurbleoriginally as oggs, but i tried converting to mp3s to try and convince them to play
01:27:24 Quit EEzde (Read error: 113 (No route to host))
01:27:26linuxstb__Did you flash your iriver with the same firmware version that was on there originally?
01:28:00rasherWait, is iriver now able to play them?
01:28:33linuxstb__Did you mean to say 8KHz or 48KHz?
01:28:41flurble8000 Hz
01:29:08stripwaxcool, so stock iriver firmware doesn't support 8khz oggs and we do?
01:29:09linuxstb__OK. It's just that I remember reading about an iriver update that meant it could play 48KHz Oggs. Previously it couldn't.
01:29:39flurblewell, the doc for the iriver said it had variable support for oggs, but could play all kinds of mp3, hence my trying converting them
01:36:29 Quit San (Read error: 110 (Connection timed out))
01:38:14 Part stripwax
01:45:22 Quit pabs (Remote closed the connection)
01:45:55 Join pabs [0] (
01:49:28 Join ncp [0] (
01:50:10ncphi, does rockboy for (iriver port) play music as well?
01:50:24ncp(the game's music, that is)
01:50:27bagawkncp, Not sure Why not try it yourself?
01:51:10ncpi did and on my player it doesn't and i was wondering if it needed some added files or configs.
01:52:43ncpanyhow. see ya
01:52:45 Part ncp
02:06:44***Saving seen data "./dancer.seen"
02:07:18linuxstb__Here's my proposed patch for radio.c which enables the presets for iriver:
02:07:28 Quit rasher ("CGI:IRC (EOF)")
02:07:52linuxstb__The On button brings up the presets menu, and the "Add Preset" option (labeled "Add") is added to the radio menu.
02:10:10linuxstb__Gotta sleep now, but I would like to commit that to CVS. I'm hoping it can be classed as a bugfix.
02:31:11 Quit cYmen ("zZz")
02:45:35 Join cYmen [0] (
02:45:52 Quit cYmen (Client Quit)
02:46:30 Join cYmen [0] (
02:58:47 Quit cYmen ("zZz")
03:23:19 Quit bagawk ("Leaving")
03:42:37 Join samwichse_ [0] (
03:47:46 Join ashridah [0] (
03:59:00 Quit samwichse (Read error: 110 (Connection timed out))
04:04:40ashridahyou know, that's got to be the first time i've ever seen godwin's law kick in without being deliberately invoked
04:05:51 Join QT [0] (i=as@madwifi/users/area51)
04:06:47***Saving seen data "./dancer.seen"
04:16:15 Nick Sucka is now known as Sucka`zZzZz (
04:18:20 Quit QT_ (Read error: 110 (Connection timed out))
04:47:11 Quit gromit` (Remote closed the connection)
04:50:27 Quit Rick (Read error: 104 (Connection reset by peer))
04:51:37 Join Rick [0] (
05:00:34 Quit noC|andY`fRa (Read error: 104 (Connection reset by peer))
05:42:44 Join JoeBorn [0] (
05:50:15 Quit AliasCoffee (Remote closed the connection)
05:50:52 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
06:05:33 Join jborn_ [0] (i=jborn@
06:05:59 Quit jborn_ (Client Quit)
06:06:49***Saving seen data "./dancer.seen"
06:16:36 Quit JoeBorn (Nick collision from services.)
06:19:28 Join JoeBorn [0] (i=jborn@
06:22:58 Join Z-Saber [0] (
06:23:46Z-SaberAnyone here familiar with ATA errors on the H140?
06:25:05ashridahi'm not, but one of the devs here will almost certainly be if you stick around
06:26:37Z-SaberI'm reformating it now, but it doesn't seem to be doing much. Progress bar is stuck, although it's all the way at the end.
06:28:26Z-SaberOh goodie. Explorer crashed.
06:28:33 Quit thegeek (Read error: 104 (Connection reset by peer))
06:28:47ashridahiriver's firmware has a built in format utility, you know
06:28:53Z-SaberCan't get to it.
06:29:14Z-SaberWhen I boot it into the original firmware, it believe it is plugged into the USB port, but it isn't.
06:29:33 Join thegeek [0] (
06:29:42Z-SaberLesson learned: Don't EVER drop your MP3 player!
06:30:30Z-SaberWell, I honestly don't know what to do right now. I could let it sit, hoping it'll magically be reformated.
06:30:50Z-SaberOr I could restart it and check it then.
06:31:19Z-SaberProblem right now is that I have no "windows," basically. No start bar, no desktop, no nothing, yet I think it is still reformating.
06:32:25 Quit Z-Saber (Read error: 104 (Connection reset by peer))
06:36:07 Join Z-Saber [0] (
06:36:23 Quit Z-Saber (Read error: 104 (Connection reset by peer))
06:36:28 Join aliask [0] (
06:41:14 Join Z-Saber [0] (
06:41:55Z-SaberThe reformat ain't doing crap.
06:42:27Z-SaberI know it's some kind of faulty connection inside the darn thing, and if I had a little direction, I could probably fix it.
06:43:16aliaskYeah, I'm pretty sure that the hard drive has little to do with it.
06:43:47Z-SaberAgreed. And using the built-in Windows reformating program, I can't even reformat it.
06:43:48aliaskI'd guess that the USB port has been hit, and it's registering as plugged in all the time.
06:43:59Z-SaberThat's exactly it.
06:44:13aliaskDoes the actual connection look broke?
06:44:57Z-SaberI honestly don't know exactly how it should look, but if something is not as it should be, it looks like it is slightly lower than it should be.
06:45:10Z-SaberBut that's my opinion and I really don't have a clue.
06:45:30aliaskAre you comfortable opening it up?
06:46:05Z-SaberAbsolutely, considering my warranty is up and I have nothing to lose really. Right now, I have a nice look storage device.
06:46:21Z-SaberNot a DAP.
06:46:33Z-SaberI don't have the right screwdriver to get it open though.
06:46:55Z-SaberAnd I really don't know how to turn it off if it thinks it's connected all the time
06:47:05aliaskHeh. I'm no expert on anything to do with electronics, or USB or anything for that matter, but it sounds like some part of the USB has shorted, so just open it up and take a look for stuff that looks plainly wrong.
06:47:08 Quit JoeBorn ("")
06:47:43Z-SaberI don't know if it did this before, but I do hear a slight rattling noise if I shake it a bit.
06:47:55Z-SaberRattling is never good.
06:48:27aliaskI just picked up my H300 and shook it, it seems to have a slight rattle to it too. Like you say, rattling is never good. :)
06:49:42Z-SaberIn my head, I'm picturing little connectors and wires, and one of those connectors is very slightly bent/disconnected.
06:50:04Z-SaberIt has to be something small, but it is so frustrating not knowing what to do.
06:50:29Z-SaberThis thing was the best investment I've ever made, and now it's just a 40GB USB drive.
06:51:24aliaskNot a good feeling. Anyway, good luck with fixing it, I have to run.
06:51:32Z-SaberAlright, nice talking to you.
06:51:45aliaskYou too.
06:51:49 Quit aliask ("Chatzilla 0.9.67 [Firefox 1.0.6/20050717]")
06:54:19Z-SaberWell, I'm gonna run here in a little bit. Guess I'll be on when I get up.
06:54:28Z-SaberI'll check the logs to see if anyone has any ideas.
06:54:31Z-SaberThanks guy.
06:54:44 Quit Z-Saber ()
08:06:52***Saving seen data "./dancer.seen"
08:09:54 Quit dpassen1 ()
08:15:05 Join JoeBorn [0] (i=jborn@
08:19:59SlasheriHmm, can i use the settings structure to implement a non-volatile variable clean_boot or is there some better place to do that? If the user has not shutdown the player properly (e.g., he has used the reset button), then we would skip some potentially harmful things on next boot (such as automatic resume if enabled)
08:33:36 Join Maxime [0] (
08:46:12 Quit paugh (".")
09:06:02amiconnSlasheri: Huh? Why do you think automatic resume is harmful?
09:06:36Slasheriamiconn: it is if user plays a track that causes the system to hang
09:06:51amiconnIt's still not harmful
09:07:11Slasherithen it's not possible to reset rockbox back to working state without usb connection
09:07:38amiconnRockbox should never hang caused by a track
09:07:46Slasheriso that feature should be added, it's quite useful when traveling and the player hangs and can't boot up
09:08:09Slasheribut sometimes it will
09:08:15amiconnI would not like that added
09:08:36amiconnSOunds very Windows-ish
09:08:38Slasherithat option could be then used for other failsafe methods also
09:08:53amiconnClean shutdown, haha
09:08:54Slasheriyes, but there is a good to be a failsafe startup
09:09:24amiconnGenerally I don't like features that work around flaws in other places. Rather fix the flaws
09:10:17Slasherithat's true also..
09:10:46Slasheribut can we fix all of the flaws and be sure they will never happen?
09:11:11amiconnWe should try
09:11:17Slasheriof course
09:12:08amiconnIf rockbox hangs caused by a track, this is clearly a bug. If a decoder is fed with data it can't understand, it should either return with an error (preferred) or decode garbage, but never crash or hang
09:12:19amiconnSame goes for the metadata parser
09:14:25amiconnI've got the impression that the swcodec metadata parser isn't really robust. Why does a vorbis file with id3 headers (even if it's not a "legal" combination) make it hang?=
09:14:39amiconnI never observed such behaviour on archos
09:15:09Slasherihmm, yes.. but keep in mind that many (and all critical) hadrware designs have also watchdog timers even that they should never crash
09:16:20amiconnI wonder why nobody is working on improving this situation. It's bugfixing time during the feature freeze...
09:26:22amiconnThere's a good test for robustness that isn't even far-fetched compared to actually possible usage situations:
09:27:23amiconnTake an arbitrary audio file, rename it to an arbitrary supported extension and try to play it. Rockbox should never crash or hang.
09:28:11amiconnI tried this on archos when I wanted to know whether the mas (unofficially) supports layer 1. Nothing crashed, it just behaved a bit strange
10:06:55***Saving seen data "./dancer.seen"
10:06:58 Quit ashridah (Read error: 110 (Connection timed out))
10:13:17 Join matsl [0] (
10:32:05 Join amiconn_ [0] (
10:32:53 Quit amiconn (Nick collision from services.)
10:32:54 Nick amiconn_ is now known as amiconn (
10:33:52 Quit matsl (Remote closed the connection)
10:38:53 Join ashridah [0] (
10:49:33 Join DangerousDan [0] (
10:50:20*HCl snickers lightly over the remaining bit of the svante thread he didn't read yet
10:50:29HClfun fun. well, i'm off
10:53:39 Quit tvelocity ("Leaving")
11:04:11 Join ender` [0] (
11:04:34 Quit webguest02 ("CGI:IRC (Ping timeout)")
11:21:35 Join cYmen [0] (
11:27:29 Join linuxstb_ [0] (
11:33:30 Quit linuxstb_ ("Leaving")
11:42:19 Join [-AIR-] [0] (
11:44:19 Join solex_ [0] (
11:44:46 Quit linuxstb__ (Read error: 110 (Connection timed out))
11:48:05 Join linuxstb [0] (
11:48:59t0mas[10:50] <HCl> fun fun. well, i'm off <−− yeah
11:49:23*t0mas doesn't understand what the guys problem is
11:49:30t0masbut the end is funny anyway :P
11:51:29 Quit [-AIR-] ()
11:54:15*amiconn is going to test his improved mpeg playtime calculation code
11:58:58 Join stripwax [0] (
12:01:20 Quit solex (Read error: 110 (Connection timed out))
12:06:59***Saving seen data "./dancer.seen"
12:07:43 Join preglow [0] (
12:09:23preglowamiconn: you havent got any layer12 files that actually cant play?
12:09:53amiconnhi preglow
12:10:58amiconnI have test files for layer 2 (all combinations of sample frequency (V1 and V2) and bitrate). All of them play iirc
12:11:19amiconnFor layer 1 I only have one file (but could produce more)
12:11:21preglowsuddenly remembered i did a couple of iram fixed to layer12.c that i forgot to commit
12:12:04 Join muesli- [0] (i=muesli_t@
12:13:23preglowthink ill just commit them
12:13:34preglowcan measure performance benefits when i have the means sometimes later
12:14:45preglowfeel free to check boost factor gains, if you feel like it
12:18:44amiconnBleh, I don't get this. Resume position isn't exact :(
12:19:02muesli-how big is the gap?
12:19:07amiconnGuess it wasn't exact before as well, but I don't know why...
12:19:19*amiconn tests
12:30:38 Nick Sucka`zZzZz is now known as Sucka (
12:31:32amiconnYes, the problem is there with the old code as well.
12:31:40amiconn(testcase: vbr with toc)
12:31:55amiconnTesting cbr...
12:33:48 Join Temani [0] (
12:33:58TemaniHey there...
12:34:29TemaniCan anyone tell me how to install the rockbox firmware on iHP 120 please?
12:34:51muesli-are u member of misticriver?
12:35:03TemaniNo, what is it?
12:35:27muesli-excellent website
12:36:15TemaniAnyway, don't you know how to do it?
12:36:18muesli-pls try the search..there are loads of topics how to do it
12:36:28TemaniOK, thanks
12:39:15muesli-oh, rbx's-how to was pimped
12:40:03amiconnrtfw ;)
12:40:28muesli-rtf - W ? ;9
12:40:40muesli-ah ok ;)
12:40:56amiconnHmm, cbr resume is exact :/
12:41:06stripwaxrtfw - heh, nice.
12:41:26stripwaxyeah so rasher was right - i've totally mashed the reset button on my iriver :-(
12:42:09stripwaxtaking the ihp apart for the first time, you gotta wonder why they didn't use a THINNER metal for the case .. ?
12:42:41 Quit Temani ("CGI:IRC (EOF)")
12:43:33amiconnThe remainder calculation for vbr with toc looks a bit fishy...
12:46:09stripwaxwow - and the reset button iriver used is the pissiest tiny thing I've ever seen, no wonder it broke apart.
12:46:49muesli-i've the feeling mine is going down the drain soon too
12:47:30stripwaxit's just a tiny self-sprung plastic gasket, about a third of a milimeter thicl!
12:47:44preglowbut ok, im out
12:47:48 Quit preglow ("woop")
12:55:48stripwaxyou know, I wonder if it's possible to shave about 3mm of internal space out of an existing ihp120 and squeeze in a 60GB disk ...
12:56:15muesli-yeah, try it. i am interested too :D
12:57:02stripwaxyeah. the casing is really pretty thick, and there's about a 1mm spacer between hdd and the battery (
12:57:28stripwaxi reckon a bit of dremelling and removing the spacer might just work
12:58:01muesli-i guess the spacer is the only thing that seperates 120 and 140
12:58:25muesli-beside the thicker battery oc
12:58:32muesli-err hdd
12:59:22bluebrother^muesli-: the h140 is a bit thicker.
12:59:36muesli-i know, got one ;)
12:59:54bluebrother^read a thread at misticriver about putting a double platter hdd in a h120
12:59:55muesli-and had an 120 before
13:00:09bluebrother^ah, ok
13:00:10muesli-wow, there's one?
13:00:30bluebrother^yes, but you need to search yourself. I just stumbled upon it once.
13:00:31muesli-somebody could do a double one in a 120??
13:01:07bluebrother^with some kind of physical force.
13:01:17muesli-ah :)))
13:01:22bluebrother^read to me as something-i-don't-want-to-do
13:01:43muesli-let me guess..he needed a squeeze or a hammer?
13:02:47bluebrother^something like this :)
13:03:34muesli-gotta go...
13:04:45 Join ep0ch [0] (n=ep0ch@
13:04:47 Quit muesli- (Read error: 104 (Connection reset by peer))
13:09:35ep0chpreglow: if you read the logs, i've done a quick test with 48 kHz stereo mp2 @ 192kbps, before 26% cpu boost, now 19%. (percentages are approximate) not sure how much of an overhead resampling 48khz->44.1khz has though.
13:11:21linuxstbQuick Archos question - does the Archos charger say "Kings Electric Ltd" on it? I've got a pile of chargers, and I can't remember which one is for my Recorder.
13:11:50linuxstbIt also says "Output: 9VDC 600mA"
13:12:04 Join noC|andY`fRa [0] (
13:12:34amiconnlinuxstb: There are several archos chargers.
13:12:39stripwaxok lads, so I've taken my ihp apart, accidentally turned it on, and it says Panic: ata -11. Presumably because the hard drive is sitting on the table next to it
13:12:45stripwaxso ... how do i now turn this off?
13:13:13amiconnThe older type is Kings Electric Co. Ltd., Model Kings41G-9-600
13:13:48linuxstbYes, that's the one I've got, thanks.
13:14:33stripwaxanyone? is it *impossible* to turn off rockbox if it gets this ata panic?
13:15:14ep0chwhat does reset switch do? and boot into iriver fw?
13:15:38ep0chtake out the battery
13:15:54ep0chuse a hammer...
13:16:16stripwaxread up a bit.. i've opened up the iriver *because* the reset switch is broken :-(
13:16:41stripwaxi thought holding 'play' will reset rockbox. does holding 'stop' not turn it off?
13:16:51ashridahstripwax: hm. either the stop or the play button should have done the trick iirc
13:17:06ashridahyou could just open it up and disconnect the power
13:17:14ashridahit's not like it's going to damage the hard drive :)
13:19:22amiconnHmm, it seems I found why the resume time often isn't exact with vbr files with toc. The archos playback code has the same problem as the iriver playback code - it mixes file and play positions :(
13:27:22linuxstbWhich gcc is best for the SH? The Wiki says 3.3.4, but 3.3.6 is the latest in the 3.3 line.
13:27:32stripwaxashridah dude it's already open.. but yeah, i had to disconnect the battery connector. holding stop or play didn't do anything
13:27:49amiconnUse 3.3.6, the official builds use the same
13:28:00 Quit noC|andY`fRa (Read error: 104 (Connection reset by peer))
13:28:05linuxstbI thought so. Thanks. Should I update the Wiki to recommend 3.3.6?
13:28:39amiconnIt says 3.3.x, use latest of the series...
13:28:48linuxstbYes, I've just seen that.
13:29:19linuxstbBut the example still refers to 3.3.4 - which is the first number I saw when scanning the Wiki page.
13:29:20amiconnWhere did you read about 3.3.4?
13:29:30linuxstbUnder "Unpack the archives"
13:30:12amiconnYes, saw it
13:30:38amiconnThe question is whether this should be changed. The recommended versions are different depending on target...
13:30:59stripwaxSo, the tiny component labeled "ST" behind the power switch is damaged, three of its (five?) legs are no longer attached. What's that component do, and (other than soldering its legs back on) can my ihp be saved?
13:31:17stripwaxsorry, behind the reset switch (not the power switch)
13:32:33linuxstbNo, I don't think that needs changing. But maybe a table at the top of the page listing the recommended binutils/gcc versions for each target would make that information clearer, and harder to miss.
13:33:50linuxstbe.g. Under Download the source code, we could have a table of versions, possibly with direct links.
13:40:49 Join gromit` [0] (n=gromit`
13:53:09stripwaxno-one knows what that little 'ST' component does?
13:55:25ashridahhasn't linus created a bunch of schematics?
13:56:18stripwaxYes−− but it's not at all obvious how one would find this particular component on any of the particular schematics
13:57:19stripwaxI imagine it's part of the power circuitry. But I have no way of telling. And I don't know which of the seven iriver schematics contains the power circuitry - if any
13:58:59stripwaxHmm.. it might not be "ST" actually - could be "1S" upside down ..
14:00:10stripwaxlooks like ST on the rockbox scan too, so oh well.
14:07:03***Saving seen data "./dancer.seen"
14:13:48 Part ep0ch
14:22:26stripwaxoh well. so it looks like it's going to have to go back to iriver. it's *just* still under warranty. reckon there's any chance of them repairing or replacing it; or reckon they'll return a non-rockboxable player instead?
14:23:15 Join JoeBorn-having [0] (i=jborn@
14:24:52crwlthey'll probably repair it if it's repairable
14:25:34crwlat least that's what they did to a friend's iriver when the main sound chip (that philips something) got somehow broken - the chip was replaced
14:30:15 Join Deadpool [0] (
14:30:34Deadpooli have a question about #rockbox
14:30:48Deadpoolis there any plan for a version for a GMINI XS 202?
14:31:36stripwaxDeadpool hrm, what does it say on the wiki page on
14:31:55Deadpoolit says that it was going to be planned
14:32:07Deadpooland then the team fell apart
14:32:12Deadpooland now the one person who was working on it is gone
14:32:29Deadpoolhowever, that really is no excuse to be an asshole about it
14:32:32Deadpooljust yes or no
14:32:58stripwaxhang on, the answer is "i don't know", am I'm not an asshole
14:33:47Deadpoolthe comment you gave was being an asshole
14:34:36stripwaxyou really think so? i don't know anything about the gmini so I was just directing you to the wiki
14:35:20DeadpoolDeadpool hrm, what does it say on the wiki page on
14:35:22Deadpoolthat's obviously a sarcastic statement
14:35:36amiconn"-ChanServ- [#RockBox] Be sure to check out the FAQ before you look for 'general' information!" <= from the channel join protocol
14:35:51amiconnThat was just a hint, nothing more
14:36:15Deadpooli don't get channel join protocol
14:36:23amiconnRockbox runs on a diversity of platforms, and obviously an arbitrary developer doesn't know about all of them
14:36:37Deadpool−−>|YOU (Deadpool) have joined #rockbox
14:36:47Deadpool=-=Topic for #rockbox is “The 2.5 feature freeze is here! No release just yet”
14:36:58amiconnIn the server window, not in the channel window
14:36:59Deadpool=-=Topic for #rockbox was set by B4gder on Monday, September 05, 2005 11:04:04 AM
14:37:06Deadpoolno chanserv join info
14:37:11Deadpoolat all
14:37:42Deadpool−−-End of /MOTD command.
14:37:49Deadpoolstripwax{#rockbox} Deadpool hrm, what does it say on the wiki page on
14:38:08Deadpoolif you want chalk it up to misunderstanding
14:38:11Deadpoolbut don't be a cock either
14:40:42stripwaxa cock? i was having a cigarette
14:40:47stripwaxlet's start again shall we?
14:41:18 Quit JoeBorn (Read error: 110 (Connection timed out))
14:41:19stripwaxGmini port? I dont' know. I haven't heard anything. I just checked the wiki page and it says the sole developer is missing-in-action, so my guess is it's currently mothballed
14:41:53linuxstbI've noticed that now a couple of plugins (searchplugin and rockboy) are in their own subdirs. I'm developing a new plugin (Sudoku) which I would prefer to develop in a subdir. Would there be any objections?
14:41:54 Join BigDave4 [0] (
14:42:13DBUGEnqueued KICK stripwax
14:42:13stripwaxWhen i joined #rockbox, this appears in my chat window: [10:59] *** -ChanServ- [#RockBox] Be sure to check out the FAQ before you look for 'general' information!
14:42:17BigDave4hello all
14:42:38Deadpooldoesn't appear in mine
14:42:41stripwaxlinuxstb- is that so it can store leveldata?
14:42:45BigDave4can I ask what will likely be a noober question?
14:42:51stripwaxBigDave4 of course
14:42:56linuxstbstripwax: No, I'm talking about the source code.
14:43:08 Join Liehbeth [0] (
14:43:17BigDave4I want shuffle to cross folders and can't seem to get it to work
14:43:19amiconnlinuxstb: Actually I don't like these plugins having separate dirs. For rockboy it is acceptable since it is actually a port of gnuboy which consists of several source files.
14:43:37amiconnFor all others, using several files is artificial, and imho unnecessary
14:43:49stripwaxlinuxstb ah sorry- what's the requirement for a separate source dir for Sudoko?
14:44:07stripwaxBigDave4 - I think you need to create a playlist consisting of those folders, and then shuffle the playlist
14:44:20linuxstbThe main reason is that I'm using code from other projects (e.g. a Sudoku solver), and it would be easier to be able to just use the original .c/.h files from the original application directly.
14:44:33stripwaxBigDave4 - or do you mean, play a folder shuffled, and then play the next folder shuffled, and so on? if so, I don't think we can do that (..yet?)
14:44:35BigDave4I can have Iruvium create those for me
14:44:55BigDave4I mean play a single tune from a folder
14:44:59BigDave4then shuffle to another song in another folders
14:45:03stripwaxBigDave4 cool, then yeah, a playlist should just work there
14:45:37stripwaxBigDave4 - or if you want the entire set of all folders, just go to Playlist options and Create Playlist in root directory −−>root.m3u
14:46:09CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
14:46:09*stripwax thinks on the fly playlists is one of the most compelling reasons to use rockbox in iriver!
14:46:47BigDave4I'll try that too. Thanks stripwax
14:47:07BigDave4Just loaded Rockbox and the shuffle piece is a little cludgy
14:47:38stripwaxlinuxstb - if that's all free sourcecode then I think a separate source folder makes sense- unless there's only a handful (say two or three) such source files, - in which case best to put them just in the plugins source folder
14:49:06stripwaxBigDave4 - I know what you mean −− on iriver, the default shuffle behaviour allows you to shuffle the entire set of songs without needing a playlist first. It's a bit of a pain on rockbox to have to rebuild root.m3u everytime new songs are added... maybe we can come up with a way to just implement 'shuffle everything' without a need for root.m3u ...
14:49:09linuxstbIn fact, the "external" code I'm using isn't that big, so I think I'll be OK with a single .c file if that's the preferred approach.
14:49:17stripwaxlinuxstb I think so
14:50:04 Part Liehbeth
14:50:14BigDave4That would be money. Thanks again. Trying the playlist solution now.
14:52:27linuxstbIf I have a file of pre-generated games, I will just store them in .rockbox (like sokoban.levels and snake2.levels).
14:53:06linuxstbBut is there a convention for user-data stored by a plugin? I will need to save the current state of the game between plays.
14:53:35stripwaxsnake2 stores hiscores, I think ..
14:53:40linuxstbAnd if I give the user the ability to either generate random games, or copy a game from (e.g.) a newspaper, where should those be stored?
14:54:24stripwaxin a directory specified by the user, i'd say
14:55:07stripwaxand store the name of that directory inside the user prefs file
14:55:50 Join Febs [0] (
14:55:58linuxstbRockboy claims the .rockbox/Rockboy directory - it would be easier (and consistent) if I created a .rockbox/Sudoku. Or is Rockboy a special case?
14:56:17amiconnSome plugins (viewer, snake2, vu_meter) store their settings in the same dir as the plugin. Rockboy creates its own
14:56:27amiconnImho both solutions are sub-optimal
14:56:59 Join matsl [0] (
14:57:30linuxstbI'm also undecided whether the "saved state" facility should only store the state for one game, or store the states for every game - i.e. treat the levels file like a book where the user can move from puzzle to puzzle, going back to older ones at a later date if they are not completed.
14:57:59linuxstbBut that gets more complicated when the levels the game knows about are a mixture of "built-in" levels and user-generated levels.
15:00:38stripwaxonce the state is saved, does it matter if the level is built in or not? the state itself must be tiny, and self contained too, I'd have thought
15:01:26linuxstbYes, the state is self-contained. But the question is whether I store one state or many.
15:01:44stripwaxI'd say many, all in one file
15:02:30stripwaxhm, anyone ever successfully raised iRiver european support telephone line on a weekend? was that just wishful thinking?
15:02:35linuxstbYes, I think so to.
15:02:45stripwaxshouldn't be complicated
15:02:47linuxstb(I mean many in one file)
15:03:03linuxstbNo idea about iriver support.
15:03:39linuxstbOK, I think my approach is becoming clearer now.
15:04:43linuxstbThere seems to be a de-facto standard ".ss" file format for storing a Sudoku game. It's used by Simple Sudoku - and that site has zip files full of games that Rockbox users could download and play.
15:05:04stripwaxoh, neat. yeah, definitely use an existing format if there is one
15:05:07amiconnIs that binary or ascii?
15:05:18linuxstbamiconn: Simple plain text
15:05:29linuxstbBasically an ascii-art representation of the game
15:05:36amiconnSounds like a good thing to do
15:05:58amiconnSokoban does similar iirc
15:06:18linuxstbThis solves the problem of my Sudoku probably not having a random game generator - Windows users can use Simple Sudoku to generate random levels, and then copy them to thier iriver.
15:06:31linuxstb(at least initially - I'm sure one will be added in the future)
15:06:38stripwaxlinuxstb - random game generator would be v cool though
15:06:43amiconnUmmomm, that reminds me, wasn't there a sokoban level-caching patch ??
15:07:11stripwax..... yessss... ;-)
15:07:19linuxstbI've found a very nice solver algorithm (in C, GPL'ed), which I'll incorporate. It solves puzzles in 0.000 seconds on my 800MHz iBook.
15:07:41linuxstbNot sure what the memory requirements are though yet.
15:07:47amiconnstripwax: Hmm, that should really get applied; should also help on archos
15:08:02stripwaxamiconn - iirc, you were going to extend it to cache 'as much as possible' that would fit into the plugin memory
15:08:04amiconn(Though I'm not worried about disk spinups on Ondio ;))
15:09:03stripwaxamiconn - my patch as-is only caches if theres > 20k (i think) memory available to the plugin - not sure if that's true on archos or not
15:09:08linuxstbI'm not sure if it's a good idea or not (it probably isn't), but my current version of sudoku automatically saves the current state when the user exits the plugin, or USB is attached.
15:09:13amiconnstripwax: The bad thing is that (ideas * time_needed_to_implement > available_time) is usually true
15:09:19linuxstbFrom a user's point of view I like it, but as I say, I'm not sure it's good practice.
15:10:18 Quit matsl (Remote closed the connection)
15:10:22amiconn...and I am currently working on fixing/improving archos stuff
15:10:40amiconn(and all-target stuff of course)
15:13:00 Quit Deadpool ("Chatzilla [Firefox 1.0.6/20050716]")
15:13:57stripwaxcool. well, i can't commit code, so .. :-(
15:15:24amiconnstripwax: Your patch won't get lost...
15:16:07 Join JoeBorn [0] (i=jborn@
15:17:15linuxstbAre all the current "saved state" files (e.g. hi-scores) ASCII files?
15:17:32linuxstbI'm assuming binary files are frowned upon.
15:17:34amiconnI don't think so
15:18:05amiconnYes, binary files aren't exactly what we want, but there are some plugins that save their settings as binary
15:18:12amiconn(e.g. the clock)
15:20:19linuxstbIf my saved state file contains mulitiple games, then I think I should search it for an existing state when a user views a .ss file.
15:24:53stripwaxlinuxstb - I actually hand in mind that your state would be the .ss file!
15:25:10stripwaxor is there any real difference between state and a .ss file?
15:25:56linuxstbMy plugin needs to know which numbers are part of the starting board (i.e. can't be modified) and which numbers the user has entered (which can be modified).
15:26:14linuxstbI'm not sure if a .ss file can contain anything apart from the starting state.
15:27:01stripwaxinteresting - i see your point
15:28:20stripwaxso i guess your state file would consist of something that begins with the name of a .ss filename, and then has an ascii representation of the initial state followed by the current state i.e. looks like two .ss files concatenated
15:28:47linuxstbI don't think I need to store the .ss filename - it doesn't contain any useful information.
15:29:04stripwaxI thought you said you wanted your plugin to look for saved state when the user opens the .ss file .. .?
15:29:38linuxstbMy idea is to store both the current and starting states in a file of fixed-size ASCII records. When a user opens a .ss file, the plugin will search the state file to see if that game is "in progress".
15:30:14stripwaxright but how will it know if a game is 'in progress' if the ss filename is not stored in your state file somewhere?
15:30:32linuxstbe.g. I could use the letters a-i to store the state, with lower-case meaning a "start number" and upper-case meaning "user number". So I just need 81 bytes (plus a CR) per game.
15:30:54linuxstbSimply by comparing the game in the .ss file with the starting state of each saved state.
15:31:08linuxstbIt's almost as quick as comparing filenames.
15:31:11stripwaxlike the idea of using letters for user numbers tho
15:31:33 Part amiconn
15:33:06stripwaxlinuxstb - if you're storing both the starting and current state in a single 81-byte record, then I can't possibly see how comparing the 'initial' part of that state with the contents of the .ss file is anywhere near as quick as comparing filenames....
15:33:38linuxstbYes, you're right. But it's still a very simple operation, and only needs to be done once - when the user loads a new .ss file.
15:33:52stripwaxok true. it's no worse than the file handling in sokoban anyway ;-)
15:34:15linuxstbSo obviously the disk is already spinning and a state file with 1000 games is only 82,000 bytes in size.
15:34:59 Quit JoeBorn-having (Read error: 110 (Connection timed out))
15:35:14 Join JoeBorn-having [0] (i=jborn@
15:35:18stripwaxjust so long as you can reset to the initial state if the user makes a mistake without going to the disk again. which should be trivial since that's part of your state record
15:35:33stripwaxsounds very nice by the way; i'll be sure to beta test it (if I get my iriver working again!)
15:36:09linuxstbIt doesn't need that much more work - this conversation has helped me work out how it's going to work.
15:36:40linuxstbBut I do want to make it work on the Archos - which is why I've resurrected my old Recorder today (I had lost the charger and USB cables, but I've found them now).
15:49:24 Quit JoeBorn (Read error: 110 (Connection timed out))
15:51:05 Join ender1 [0] (
15:59:04linuxstbI'm getting an error compiling Rockbox for an Archos Recorder (target #2). During the "LD rockbox.elf" stage, I'm getting the error "/usr/local/sh-1/lib/gcc-lib/sh-elf/3.3.6/../../../../sh-elf/bin/ld: error: no memory region specified for loadable section `.rela.dyn'"
16:00:07linuxstb"sh-elf-ld −−version": GNU ld version 2.16.91 20050813 (recent CVS of binutils)
16:00:57 Quit ender` (Read error: 110 (Connection timed out))
16:03:18 Join stripwax_ [0] (
16:05:47 Quit stripwax_ (Client Quit)
16:07:05***Saving seen data "./dancer.seen"
16:07:49 Join ender` [0] (
16:11:46 Join [1]ender [0] (
16:12:28 Quit gromit` (Remote closed the connection)
16:12:38 Join gromit` [0] (n=gromit`
16:16:59 Quit stripwax (Read error: 110 (Connection timed out))
16:22:29 Nick JoeBorn-having is now known as JoeBorn (i=jborn@
16:23:07 Quit ender1 (Read error: 110 (Connection timed out))
16:27:55 Quit ender` (Read error: 110 (Connection timed out))
16:37:26 Quit ashridah ("Leaving")
16:40:56 Nick Sucka is now known as Sucka`away (
16:44:37 Quit BigDave4 ("CGI:IRC (EOF)")
16:55:55 Join XavierGr [0] (
17:04:41 Quit XavierGr ()
17:09:42 Join mirak_ [0] (
17:09:47 Join ]RowaN[ [0] (
17:09:52mirak_bonjour les gens !
17:11:03 Quit JoeBorn (Nick collision from services.)
17:32:54 Quit solex_ (Read error: 110 (Connection timed out))
17:35:11 Join bagawk [0] (i=1000@unaffiliated/bagawk)
17:37:34 Quit mirak_ (Remote closed the connection)
17:37:45 Nick [1]ender is now known as ender` (
18:07:07***Saving seen data "./dancer.seen"
18:16:05 Join linuxstb_ [0] (
18:17:43 Quit linuxstb (Read error: 110 (Connection timed out))
19:08:41 Quit linuxstb_ ("Leaving")
19:13:09 Join Mxm`Pas`Bien [0] (
19:13:09 Quit Maxime (Read error: 104 (Connection reset by peer))
19:27:13 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
19:30:36 Join Nilisco [0] (i=nilisco@
19:35:24 Nick Sucka`away is now known as Sucka (
19:38:06 Join Musicmad [0] (
19:38:46 Join linuxstb [0] (
19:47:00 Join Zagor [0] (i=foobar@pdpc/supporter/sustaining/Zagor)
19:47:56*Zagor updates the web server. there will be a short downtime (and disconnection for irc web clients).
19:50:45 Quit ]RowaN[ ("CGI:IRC (EOF)")
19:51:06 Join Shagnar [0] (
19:51:23Shagnarhey ho whats going on with the page?
19:51:51Zagori'm updating the server. back up in a minute.
20:07:11***Saving seen data "./dancer.seen"
20:15:48 Join TCK- [0] (
20:26:03 Quit bagawk ("Leaving")
20:31:54 Quit TCK (Read error: 110 (Connection timed out))
20:36:56 Join Moos [0] (
20:38:20 Join solex [0] (
21:04:40 Quit Musicmad (Read error: 110 (Connection timed out))
21:24:59 Join ender1 [0] (
21:27:52 Quit ender` (Nick collision from services.)
21:27:57 Nick ender1 is now known as ender` (
21:30:58 Join muesli- [0] (i=muesli_t@
22:00:51 Join blackvd [0] (
22:01:52 Quit samwichse_ ("Leaving")
22:05:33 Quit blackvd ("Leaving")
22:07:12***Saving seen data "./dancer.seen"
22:13:29 Quit Zagor ("Client exiting")
22:16:42 Join hshah [0] (
22:18:53hshahits been sooo long since i updated my iriver
22:19:17hshahthe wps etc has all changed :s
22:24:25linuxstbFebs: The links at the end of your misticriver post here seem to be broken:
22:24:43linuxstb(the ones under "Useful Mistic River Rockbox Threads")
22:26:21hshahthe root to wps patch no longer works :'(
22:32:08 Join paugh [0] (n=pete@2001:5c0:8fff:ffff:8000:0:3e03:6822)
22:32:09 Quit hshah ("CGI:IRC (EOF)")
22:37:18 Join Sucka` [0] (
22:38:56Febslinuxstb, thanks I'll check out that thread and fix the links.
22:40:31FebsI really wish that I could get the MR people to use a wiki. It's a pain to have to find a moderator to open a closed thread.
22:41:01 Join XavierGr [0] (
22:44:23linuxstb :)
22:47:49FebsI know, I know. Actually, I've made many, many updates the RB wiki. But the MR folks asked me to do the sticky thread about Rockbox info.
22:48:01FebsIt was either that, or they were going to do their own Rockbox FAQ.
22:48:30FebsWhich I was firmly opposed to, as two parallel FAQs would be more work and the MR FAQ would undoubtedly get out of date quickly.
22:48:43FebsSo the sticky thread with links to information was the compromise.
22:49:10 Quit Sucka (Read error: 110 (Connection timed out))
22:50:25linuxstbI agree with you. Shame you can't edit your own sticky thread though.
22:52:24FebsI know! And there are no moderators online right now ... which is pretty amazing, actually.
22:55:15 Quit edx (Read error: 110 (Connection timed out))
23:01:46 Join webguest64 [0] (
23:10:48 Quit webguest64 ("CGI:IRC (EOF)")
23:33:30 Join webguest13 [0] (
23:39:53 Quit paugh ("Leaving")
23:45:10 Quit XavierGr ()
23:49:44 Quit webguest13 ("CGI:IRC")
23:50:15 Quit TCK- (Read error: 110 (Connection timed out))

Previous day | Next day