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 2002-08-08

00:02:28 Quit jedix` (Read error: 110 (Connection timed out))
00:57:48***Saving seen data "./dancer.seen"
01:37:33freshmakeris still someone there?
01:39:53freshmakerZagor: what's the size of the player's LCD?
01:40:21Zagor11x2 characters
01:41:01freshmakerONLY 2 LINES?? thank god, i have a recorder :-)
01:41:37freshmakercan i use all 11 chars in menu entries? or are there some other constraints?
01:42:43Zagorthey scroll, so it's ok. if you can keep it to 10 it's best though
01:43:50freshmakerhmm, what's a good short version of "volume characteristik"?
01:46:21Zagor"Vol scale" maybe?
01:52:22freshmakercan i use the struct global_settings in mpeg.c ?
01:54:04freshmakeris there any way to get an int from global_settings into mpeg_sound_set() w/o changing the API?
01:56:34 Nick dwihno is now known as dwihno|gone (dwihno@Bald067.Baldakinen.Umea.SE)
01:56:34DBUGEnqueued KICK dwihno|gone
01:56:56freshmakerok, then it must be done somewhere else.
01:58:31freshmakeri need a small function in sound_menu.c and wps.c doing that voluem scale conversion, what's the best place to put it
01:58:56Zagorwps? why do you need it in wps?
02:00:00freshmakerwps calls mpeg_sound_set(SOUND_VOLUME, ...)
02:01:47Zagoryeah, but many other places do that too. we need a more central place to do the scaling.
02:03:38freshmakeryes, that would be mpeg_sound_set() itself, but i need to get the info which scale to use; currently i'm storing it in global_settings
02:05:34Zagora new sound setting, then. SOUND_VOL_SCALE
02:06:25Zagori'm off to bed. see you, guys
02:06:30 Quit Zagor ("Client Exiting")
02:27:30freshmakeryeah, nice. but that new toshiba thing is much nicer :-)
02:29:42freshmakerdo you know how large that Mambo-X P350 is?
02:30:00elinenbe3 feet by 2 feet
02:30:24elinenbeit uses a jukebox of old 5 1/4" floppies
02:30:55freshmakerhmm, portable as a backpack?
02:32:14 Quit ironi_ (Read error: 104 (Connection reset by peer))
02:35:20 Join ironi__ [0] (~ironi@
02:57:52***Saving seen data "./dancer.seen"
04:12:48USAG33kd00d, i have no idea how to program at all for rockbox
04:12:57USAG33kbut like, i have a feeling i am going to tear it up eventually
04:19:23 Quit freshmaker ("ChatZilla 0.8.7 [Mozilla rv:1.0.1/20020805]")
04:24:45 Join MeRWiN [0] (~merwin@
04:24:53MeRWiNHow much does HZ equal again?
04:26:41adi|homekernel.h:#define HZ 100 /* number of ticks per second */
04:27:12 Quit USAG33k ("Client Exiting")
04:44:53 Join elinenbe_ [0] (
04:45:02 Quit elinenbe (Read error: 110 (Connection timed out))
04:57:53***Saving seen data "./dancer.seen"
05:03:16 Quit elinenbe_ ("User pushed the X - because it's Xtra, baby")
05:03:16 Quit MeRWiN (Read error: 104 (Connection reset by peer))
05:27:10 Join elinenbe [0] (
05:29:31 Join MeRWiN [0] (~merwin@
05:32:44 Quit elinenbe ()
05:33:40 Join elinenbe [0] (
05:34:04 Quit elinenbe (Remote closed the connection)
05:35:41 Join elinenbe [0] (
06:37:05 Quit elinenbe (Read error: 104 (Connection reset by peer))
06:57:57***Saving seen data "./dancer.seen"
07:08:09 Join rwood [0] (
07:15:42 Quit rwood ()
07:17:52 Join Linus [0] (
07:18:58 Join rwood [0] (
07:22:00rwoodHi, this is Randy Wood, is linus listening?
07:22:19 Join elinenbe [0] (
07:22:35Linusi am
07:23:02Linusi reviewed your patch yesterday
07:23:07rwoodDid you get my email regarding file attributes
07:23:13 Part elinenbe
07:23:19 Join elinenbe [0] (
07:23:25Linusnot that i can remember
07:23:35elinenbeI am heading to sleep.
07:23:44Linussleep tight, elinenbe
07:24:15rwoodi looked at the IRC log from yesterday, it looked like you had a problem and were questioning attributes.
07:24:36Linusi was just wondering how many bits we could use
07:25:06rwoodfat.h has the attribute field a unsigned short, tree.c used char. i changed tree.c to match fat.h
07:25:21Linusyeah i noticed that
07:25:42rwoodwere you able to get it to run?
07:25:47Linusnot on the recorder
07:25:47 Join Synthe` [0] (
07:25:53 Quit Synthe (Read error: 104 (Connection reset by peer))
07:26:07 Nick Synthe` is now known as Synthe (
07:26:15MeRWiNmorning linus
07:27:03rwoodmy guess is that it has something to to with stopping the running app before loading the file or initializing the hardware before transferring to the new app
07:28:02rwoodsomeone that has looked at the hardware could probably give some insight into what i should have done
07:28:48Linusyeah, it's just a matter of resetting to the initial state
07:29:54MeRWiNLinus: too bad it can't make it in 1.2 ... but I'm guessing that there's probably a few hidden bugs that might need to be worked out, so it seems as if it's probably better that it's not in 1.2
07:29:57MeRWiNrwood: great patch though :)
07:31:01rwood i'm considering getting a recorder - harmony computer has the 10gb for $219 and a $30 mail-in rebate
07:31:17MeRWiNLinus: also, something that was done in the past week or so to the code stopped that problem with skipping to the next track immediately after loading it
07:31:18LinusMeRWiN: yeah, for example, i'd like it to work on the recorder :-)
07:31:29MeRWiNLinus: working on both is probably a good thing
07:31:41MeRWiNrwood: where are you from?
07:32:02LinusMeRWiN: what skipping-to-the-next-track problem?
07:32:19rwoodNewport Beach, CA - my company is about a mile from disneyland
07:32:43MeRWiNLinus: in playlists, if you skipped to the next track and then tried to skip to another track within 3 seconds it wouldn't load because the HD was told to spin down
07:32:57MeRWiNrwood: nice :) I'm from sunny washington state
07:35:41Linusand i'm from cloudy Sweden
07:36:06MeRWiNLinus: I'm heading over to europe next week for work... spain, uk, italy, and france
07:36:22LinusBjörn's latest next-file CVS commit wasn't all that good
07:36:32LinusMeRWiN: nice
07:36:41MeRWiNLinus: the next file is working now though... i was suprised
07:36:58Linusit isn't when you have several tracks in memory
07:37:20MeRWiNLinus: like how many? and does this happen with dir and playlist?
07:37:54Linusi have thought a long time of doing that fix, but i haven't because it is very tricky
07:38:24Linusit fails when you have more that just the next track loaded in memory
07:38:34rwoodLinus: are you looking at improving the time to change to the next track?
07:38:41Linuslike when you have a dir full of very short files
07:38:45MeRWiNahh... so if you have a file that is under 1mb
07:38:51MeRWiNit'll load that file and the next also
07:39:04Linusrwood: that was the intention
07:39:29Linusbut Björn rushed it (right before the 1.2 release, shame on him)
07:39:53MeRWiNLinus: were we going to leave random as it is for the 1.2 release? Right now for me it doesn't start with the selected file. I strongly suggest changing that for the release :)
07:41:52MeRWiNis there an easy way to reset all of your files to the cvs files? or is it just rm -rf and a cvs update
07:42:42LinusMeRWiN: remove and update is the best
07:42:59Linusother people strongly suggest that it should play a random file
07:43:05MeRWiNLinus: i'm working on individual patches right now... for AFTER 1.2
07:43:52MeRWiNi figure it's better than a huge patch full of everything
07:45:36rwoodLinus: there has been some discussion of changing the previous/back key to operate like a CD player - restart track is > n seconds, start previous track is < n seconds. what are the developers plans?
07:46:27MeRWiNLinus: I just recompiled the archos.mod from the CVS and the player still displays "Time: XX:XX / XX:XX" instead of "XX:XX/XX:XX" ... weird, someone must have reversed it
07:48:07MeRWiNrwood: wouldn't take too much work to implement... I'll write a patch for it now :)
07:50:28LinusMeRWiN: do you really think it's that easy?
07:50:58rwoodMerWin: that would be great - by the way, thanks for the complimnent earlier - this is the first time that i have been in a chat room and i'm running a step behind
07:51:39MeRWiNrwood: heh... this channel is usually dead from about 5pm to 11pm our time, just to let you know
07:52:10LinusMeRWiN: the time display looks ok in the source code
07:53:09rwoodMeRWiN: i sort of guessed that from the logs on the site - my wife crashes about 9 and i play on the computer til 12 or 1
07:54:00MeRWiN snprintf(buffer,sizeof(buffer), "Time:%3d:%02d/%d:%02d",
07:54:05MeRWiNthat's what's in the current cvs code
07:54:15MeRWiNI just removed my file and updated
07:54:49Linusnow i see
07:56:18LinusMeRWiN: change it too "%d:%02d/%d:%02d " and try again
07:56:42MeRWiNLinus: that will work
07:57:26MeRWiNThat makes it up to 11 characters if it's under 100 minutes
07:57:44Linusi guess that will be enough :-)
08:00:04MeRWiNLinus: i would personally use %02d for all, because it would look nicer, but unfortunately it would limit it to under 100 minutes :(
08:03:11 Join Bagder [0] (
08:04:02BagderI checked yday more closely, the archos stock firmware charges my batteries a lot more
08:04:36Bagderwhen I switch over to rockbox, I got at least 5.55 V
08:04:48Linuslet's harass Hes when he wakes up
08:04:50Bagderwhile rockbox could stop already at 5.1
08:05:37Linushow long did the Archos charge take compared to the rockbox charge?
08:05:40Bagderin the meantime, I'll charge my unit with the stock firmware until it says "charged" and check again with the battery meter display
08:06:00BagderI didn't even bother to wait for the Archos charge to complete
08:06:06Bagderit takes *a lot* longer
08:06:44Linusinteresting, i guess they don't use negative-delta detection
08:07:02Linuswe need to do some more research
08:07:02Bagdervery possible
08:07:11Bagderor perhaps a less strict check
08:07:18LinusBagder: Zagors last mpeg.c fix sucks
08:07:30Linushe must have been tired
08:07:41Bagder's yell at him when he gets here ;-)
08:08:54CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
08:08:54*Bagder laughs...
08:09:10Bagderparts of a search string on the curl site
08:14:49Linussounds desperate to me
08:15:22Bagderto me it sounds like "I don't know what I'm doing here but let's search for some porno at least"
08:17:47BagderLinus: the netscape webmail thing turned out to be a whole can of worms ;-O
08:18:10Bagderbut I'm doing progress
08:18:11Linusi just read on a battery manufacturer's site that they recommend about 2 hours of trickle charging after negative delta detection
08:18:20LinusBagder: surprised?
08:18:40Bagderyes, this is worse than I expected
08:18:47Linuscookie hell?
08:18:54Bagderredirect mania
08:19:00Bagderjavascript hell
08:19:21Bagderduring the login process, my mozilla does 29 GET requests and 3 POSTs
08:19:36Linusit's called "web designer job security"
08:19:58Linusor maybe just plain stupidity
08:20:11BagderI think they've obfuscated this on purpose
08:20:18Linusin the web design business, it's hard to tell the difference
08:21:06BagderLinus: seen the new status page on the rockbox site?
08:21:32Bagderall you want in a single page ;-)
08:22:22Linusi like the bleeding edge builds
08:22:48Bagderthat version define is nice in those
08:23:03Linuswhat is it?
08:23:25Bagderit sets the version string to date and time
08:23:33Bagderthat shows in the version screen
08:24:00Linusbut that isn't exactly new
08:24:14Bagderbut useful in these cases
08:38:33 Quit notch|lunch (Read error: 113 (No route to host))
08:39:55MeRWiNi'm having much more trouble with just restarting the track than i thought it was going to be
08:40:06LinusMeRWiN: i warned you
08:40:20MeRWiNI can get close
08:40:47MeRWiNLinus: shouldn't be too hard, because if you restart the track within the first 3 seconds, it will all still be in memory...
08:41:05MeRWiNyou just need to change the last_dma_tick, right?
08:41:17Linusnot at all
08:41:34Linusyou might have to back the buffer pointers
08:41:47Linusand reinit the dma
08:42:11MeRWiNreinit the dma is easy... back the buffer pointers i don't get
08:42:50Linusif the dma has advanced one block it will have advanced the read pointer as well
08:43:59Linusthat can happpen on high-bitrate files
08:44:41MeRWiNi could do it the old fashioned way and do prev_track() next_track() :-)
08:44:57*Bagder slaps MeRWiN ;-)
08:45:13Linusand you could probably get a TRACK_CHANGE event that you don't want
08:45:47MeRWiNI like this sh1 compiler in cygwin much better than that other one that's on the website
08:46:03Linusit is almost the same problem Zagor didn't care about in his latest "fix"
08:46:32MeRWiNwhat is the problem with his latest fix... the don't load from disk?
08:47:03Linusit skips more than one file
08:47:44Linuswell, it screws up big time, actually :-)
08:48:09MeRWiNI'd consider that a showstopper for the release :P
08:48:31Linusif you ask Zagor he probably wouldn't
08:48:49Linus"not many people have that small files"
08:48:56Linus"i don't"
08:49:04Bagderit is probably true
08:49:31Linusyeah, i guess most people have files over 1.5Mb
08:49:43Bagderyes, 99% at least
08:49:52MeRWiNi've got a bundle of them under 1.5mb
08:50:02Linusthe problem is that his patch also can stop playback accidentally
08:50:11MeRWiNabout 50% of the southpark mp3's are under 1.5mb
08:50:24Linusif the next file is only partially loaded
08:50:56Linuswe have to remember that not all people have only cd-rips on their boxes
08:51:26Linusthey download small files from the net, they record snippets with their recorders
08:51:34Bagderyou mean someone might have DOWNLOADED files? B-)
08:52:06Linusnot me of course, but i have read i the newspapers that some people do that
08:52:15*adi|home gasps
08:52:18adi|homethose sinners
08:52:22*adi|home goes back to bed
08:52:45BagderI wish someone made a good audiogalaxy replacement
08:52:50MeRWiNactually, doing mpeg_prev(); mpeg_next(); seems to work :P It doesn't even seem to up the load time at all.
08:52:54MeRWiNg'nite adi
08:54:08LinusMeRWiN: have you changed the ata_sleep()
08:54:13MeRWiNLinus: nope
08:54:33MeRWiNLinus: i took the stock cvs and added mpeg_prev(); mpeg_next(); to it
08:54:47Linusbut you have zagor's latest "fix"?
08:54:55MeRWiNcurrent cvs
08:56:12Linusthen it loads the previous track and then the next again
08:57:47MeRWiNdoesn't seem to up the load time though... it's really interesting
08:58:01***Saving seen data "./dancer.seen"
08:58:06MeRWiNyou should try it :) It's like a 5 line patch
08:58:12Linusas long as the disk is spinning, and the track is in the disk cache...
08:58:44rwoodMeRWiN: it seems like it should double the load time minus the spinup time
08:59:08Linusnot really
09:00:10Linusit stops the loading of the first track before it has loaded the large part of it (before the low watermark is passed)
09:00:40Linusso it loads 64-128k of the first track and then starts loading the correct track
09:01:06Linusbut it will have to seek both files for the ID3 info
09:01:20MeRWiNLinus: the only weird part is that it says "End of track, Press ON" inbetween restarting
09:01:21rwoodi was just going to ask how much you loaded before starting
09:02:05MeRWiNend of list, press on
09:02:49Linusrwood: it loads in 64K blocks until the low watermark is reached (which is 512K atm)
09:03:04Linusthe it loads the rest in one chunk
09:03:26rwoodLinus: thanks
09:04:40Linusthe low watermark is probably way too high nowadays
09:04:58rwoodis there much difference in performance between ID3-1, ID3-2, and no ID3?
09:05:13Linusnot really
09:05:19MeRWiNit tries to load both of the id3's no matter what, right?
09:05:31Linusthe id3 parser looks for both v1 and v2
09:05:57Linusif we didn't look for v1 of v2 is found, we can gain some performance
09:08:33rwoodthe next track time on the Archos firmware is so small, they must be loading a very small sample before starting
09:08:57MeRWiNwow, Micro$oft added a "Set program access and defaults" where you can supposedly disable media player, ie, etc
09:10:36Linusrwood: do they spin up the disk at all?
09:11:57MeRWiNLinus: maybe we could have a little 512k segment that has the beginning of the next track in it?
09:12:00rwoodmore like they don't spin down immediately - possibly they let the shut-down timer spin down then sleep later?
09:12:02MeRWiNor even 256k
09:12:42Bagderright, we still spin down too fast, don't we?
09:13:08Bagderwhen loading music
09:13:27rwoodi think that a spin down time like the backlight timer might be useful
09:13:45MeRWiNrwood: this was discussed as an option in the menus
09:13:45Bagderrwood: there is one, but the mp3 loader circumvents it ;-)
09:13:47LinusMeRWiN: we always have the next track in memory
09:14:00MeRWiNLinus: then why does it take so long to load it?
09:14:39Linusbecause we don't use the already loaded track when we press NEXT (well, that is what Zagor tried to do, and failed)
09:15:01MeRWiNeven with what zagor did, it still doesn't do it for me
09:15:10rwoodis the reaction time of the disk different for shut-down versus sleep?
09:15:11Linuswe reset the buffer and start over with the next track
09:15:24LinusMeRWiN: that is why i wrote that he failed
09:15:36MeRWiNheh... ok.
09:15:43Linusrwood: on my disk there is hardly any difference
09:15:45MeRWiNhe should really have thought first
09:16:10MeRWiNin any case, bedtime for me... gotta get up in 2.5 hours
09:16:17LinusMeRWiN: night
09:16:48Bagdernight MeRWiN
09:17:18rwoodMeRWin: thanks for trying - night
09:19:21Linusi'll debug zagor's fix now...
09:19:52Linus(gdb rocks)
09:20:20*Bagder hails Linus
09:20:34 Quit wettoad ("Client Exiting")
09:21:03rwoodLinus: do you have a full rs-232 interface to gdb?
09:22:05rwoodare you using both the in and remote for communications?
09:22:45Linusi have connected serial port 0 to the line-in plug
09:23:08 Quit WetFlax (Read error: 104 (Connection reset by peer))
09:23:09 Join WetFlax [0] (
09:24:10rwoodthe pictures on the site show adding two wires - do both go to the line-in connector?
09:25:11rwooddoes the mod work for both player and recorder?
09:25:25Linusyes, but you have to do it differently
09:25:43Linusthe circuit boards don't look the same
09:26:25Linusi was lucky on my recorder, the Line-In was already patched
09:26:41Linusso i just moved the wires
09:27:30Linusi just received 4 Nigeria bank account scams!
09:27:51Linusmajor mail address harvesting
09:28:08rwoodi'm leaving for vacation fri. i'll my re-work girl mod my player week after next.
09:29:00rwoodwe get Nigerian faxes 2 or 3 times a week - it's amazing the number of people that fall for it
09:29:29BagderI find the mount of different versions of the Nigerian mail interesting
09:29:37Bagderthere must be a lot of people involved
09:29:39*adi|home nods
09:29:53adi|homeartcile in the local paper said some couple lost 25k
09:30:31rwoodthe new one is to get you to a different country, then hold you for ransom
09:35:04Bagderhm, I got 6 ones
09:35:10Bagder"DR. VICTOR PETERS"
09:35:23Linussame as i got
09:36:49 Nick dwihno|gone is now known as dwihno (dwihno@Bald067.Baldakinen.Umea.SE)
09:36:49DBUGEnqueued KICK dwihno
09:36:53dwihnoGood morning! :)
09:37:01rwoodit's dark night 30 here - i'm giving up for the night
09:37:20Bagderg'night rwood
09:37:22Linusrwood: good night
09:37:59rwoodLinus: i hope you can correct the error in ROLO for the recorder, after you get 1.2 released.
09:38:41 Quit rwood ()
09:39:38adi|homewe have a version of ROLO working
09:39:54Linusdoes it work for you?
09:39:57Bagderadi|home: we have rwood's patch that loads a mod and runs it
09:40:08CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
09:40:08*adi|home faints
09:40:11adi|homewtf did we get that?
09:40:26*adi|home needs to come back to the project
09:40:43adi|homewell.. as soon as 1.2 goes gold, im putting in the patch for song limit sleep
09:42:32adi|homedamn.. i completely missed that
09:42:40adi|homewhats the bug in it?
09:43:06*Bagder hasn't tried it
09:43:34*adi|home nods
09:43:42*adi|home will wait while others play
09:43:47Linusthe loaded firmware freezes on the recorder
09:43:59adi|homeim prob going to see if one of my buds can give me a hand soldering this damn player
09:44:05adi|homethe reboot is WAY to sensative
09:44:33Linusreboot? sensitive?
09:46:37*adi|home nods
09:46:39adi|homethe battery issue
09:46:44adi|homecontacts and all
09:46:51adi|homeive been dealing with it since i got the unit
09:47:05adi|homeeverything is fine when its plugged in.. but unplug it.. and *poof*
10:07:09Bagderbtw, is there really a need for the config entry 'status bar' ?
10:07:20Bagdergiven that f3 works
10:08:56 Nick mbr|gone is now known as mbr (
10:09:12Bagdermorning mbr
10:09:24mbrBagder: The global setting was on wishlist ..
10:09:36LinusBagder: the F3 setting isnt't saved
10:09:51Bagderso why not let f3 toggle the setting
10:10:14BagderI think it makes more sense
10:10:44Bagdernow it works like a setting to enable the f3 for toggling
10:11:29dwihnoDon't you think it would be better keeping all the configuration in a separate place?
10:11:50BagderI think it should be easy to use
10:12:00Bagderthis makes it harder to use
10:12:03Linusthe relative path handling in playlist.c puts two slashes between the path and the file
10:14:07mbrBagder: ok, i think it makes sense to remove the settings menu. i'll remove it. shall i do it now or wait until 1.2 released?
10:14:31Bagderto keep Zagor happy, it should prolly wait until after
10:14:52Linusi am very keen on removing his silly next-file fix
10:15:16Bagderso correcting it is much work?
10:16:34mbrBagder: One other thing about that. Now I recall, that I introduced the extra menu.
10:17:10mbrI don't wanted to change a global setting every time the status bar is turned off (for menu, usb)
10:17:47mbrMay be a problem if we save to disk ..
10:17:58Bagderbut the global setting needs only to be changed when f3 is pressed
10:18:13mbrYou are right ...
10:18:27Linusthe menus and the USB don't have to care about the status bar setting
10:18:38Linusthey just don't call status_draw()
10:19:15Bagderalso a good point
10:19:58Linusand also a good reason not to have the status bar drawing in a separate thread, as was my idea a while ago
10:20:00Bagderthen again, why isn't there a status bar in the menus?
10:20:23Linusbecause they don't call status_draw()
10:20:36Bagderwell shouldn't it?
10:23:58BagderI think the status bar should be present more or less everywhere
10:24:17Linusme too
10:24:38Linusmaybe not in the games
10:24:41mbrEven in usb mode?
10:24:57Bagderperhaps not in usb mode
10:25:16 Join notch|lunch [0] (
10:25:16Linusit would actually be nice to see the battery status
10:25:24Bagderthat is true
10:25:26 Nick notch|lunch is now known as notch (
10:25:36Linusbut that takes more than just a status_draw()
10:25:41Bagderyou wanna know if it is gonna die while usb'ing
10:25:53Linusbecause the power thread is stopped while the USB is working
10:26:10Bagderbut it doesn't have to be stopped, right?
10:26:11Linusa fairly quick fix though
10:26:17mbrThe problem is, that each lcd owning thread must redraw status bar periodicaly
10:26:54notchLinus: got my remote routine working... however as soon as I send a command the software frezes...
10:27:14Linusmbr: the button_get_w_tmo() was made for that purpose
10:27:17notchI'm going to look into it, however any ideas off the top of your head?
10:27:39Linusnotch: do you clear the interrupt after serving it?
10:27:52Linusie writing 0 to the event bit
10:28:53notchmanuals! who needs them! weve got linus!
10:29:22*Bagder slaps the "manual" sticker on Linus' forhead
10:30:21LinusRTFM? No, ALFGS: Ask Linus For God's Sake
10:31:10Linusnotch: you have to read the event register first, and then clear the bit
10:32:18notchie read the result, then clear the int bit... In the manual it says this is automatically cleared grrrrr..
10:33:04BagderI'm disspointed, you *did* read the manual! ;-P
10:33:20notchfor what good it did me!
10:33:36Linusnotch: i am talking about TDRE
10:33:42 Join Zagor [0] (
10:33:55Bagderok guys, one two... three
10:33:58Linusand RDRF
10:33:59Bagderboooo zagor ;-)
10:34:10DBUGEnqueued KICK Linus
10:34:11Zagorhehe, what?
10:34:12notchbooo (for what?)
10:34:25Linusbecause his next-track-in-ram fix sucks
10:35:04Zagorhaha. well why don't I see a patch for it then?
10:35:32Bagderbecause we rather complain than actually do anything ;-)
10:35:38Linusbecause i haven't found out how to do it properly
10:35:56Zagorwhat's the problem with my method? it works for me.
10:35:58Linusthat is why i havent done that obvious enhancement yet
10:36:17LinusZagor: you probably have only two tracks in memory
10:36:22notchlinus: TDRE (Transmit) or RDRF ????
10:36:24Linusthe current and the next
10:36:55Linusit is cleared when you read a 1 and then write a 0
10:37:05ZagorLinus: and it doesn't work with 3, because?
10:37:29Linusbecause it does two track changes
10:37:52Linusthe DMA IRQ sends a track change event
10:38:03Linusit isn't that easy
10:38:18Linusbut there is another bug too, that i haven't found yet
10:38:19notchchar serial_byte;
10:38:19notch serial_byte = RDR1;
10:38:19DBUGEnqueued KICK notch
10:38:19notch SSR1 &= 0x40; /* Clear RDRF */
10:38:33LinusRDR1 is a bit in SSR1
10:38:49LinusSSR1 = SSR1 & ~0x40;
10:39:31LinusZagor: your fix will stop playback if the next track is only partially loaded
10:39:42Linuslike if only 32K is loaded
10:40:00Zagoryou mean if not a whole dma buffer is loaded
10:40:08Linuswe must set the panding flag, and load more data before starting the DMA
10:40:12notchnotch claims cut and paste immunity!
10:40:23Linusotherwise the disk won't spin up in time
10:40:28ZagorLinus: that kills the whole point of the fix
10:40:42Linusyou mean you want it to stop playback?
10:41:06Linusthat only applies when too little of the next track is loaded
10:41:13Zagorno, but we don't want to load more data unless we're below watermark
10:41:28Linusof course
10:41:49Linusi just mean that the fix is incomplete
10:42:51Linusit isn't as easy as just advancing the read pointer and restart the dma
10:43:07Linusif it was, i would have done it a long time ago
10:45:15Zagorwell, if we perform the track change instead of posting to the queue the dma interrupt won't run it again
10:45:40 Join pyvasene [0] (
10:46:28Linusor we could just let the dma do it
10:46:37Linusand don't do it ourselves
10:47:24Linusbut woth the special case when the next track is below the watermark
10:47:44Linusthen we set the pending flag and start loading data
10:48:31Zagori remember now, when I didn't change track it didn't work. I never got the problem with two track changes.
10:49:09Linusthat's probably because you only had two tracks in memory
10:49:40Zagorthat shouldn't make a difference. we only deal with one track change here.
10:50:22Linusif you have several tracks in memory, the dma thread will send a track change event too
10:51:10Zagorno. if i've already done the track change, the read pointer in the dma check won't match current buffer. (it looks at next track)
10:51:30Linusexactly my point
10:52:01Zagorso... then it won't do the track change. and there is no problem. will you start talking anytime soon?
10:54:55Linushmmmm. maybe the bug is something else....
10:55:07Linusthe GDB output is strange
10:55:19Zagordo you have some nice short tracks I can use for testing?
10:55:39Linusi think i know what it is now
10:56:04Linusmy Farbror Frej tracks are very short
10:57:02Linusinit_dma() doesn't take that into account
10:57:31Linusit inits a 64K buffer regardless of the next track change pointer
10:58:02***Saving seen data "./dancer.seen"
10:58:29 Nick Zagor is now known as Zagor|breakfast (
10:58:50 Join edx [0] (
10:58:52Linusi think we can trick the DMA IRQ to do it by playing the last byte of the current song
10:59:00edxgood morning.
10:59:20Bagdermorning edx
10:59:43BagderI found out that the bleedging edge win32 simulator files don't run
10:59:56Linusvery bleeding then
11:00:08Bagderit might be my cross-compile that doesn't work completely
11:00:13BagderI've never tried them before
11:00:15edxrecorder or player or both?
11:00:27Bagdertry them from the daily build page
11:00:32Bagderat the bottom
11:00:38edxwhat the f...
11:00:45edxthey are like consoles
11:00:48LinusBagder: haven't you tried to run them at all??????
11:00:54edx.. they are supposed to be windows
11:01:04LinusBagder: that's naive
11:01:06Bagderah, I bet that's a reason
11:01:10edxtry those: - they work
11:01:36BagderLinus: no, only if I thought it would work
11:02:09Bagderso they probably don't work since they're not console applications
11:02:11Bagderor something
11:02:26Linusif you didn't think it would work, why did you put them up for download?
11:02:32Bagderwhatever, they still work as a good test that the win32 code compiles
11:02:37BagderLinus: I didn't
11:02:43BagderZagor did
11:02:45edxBagder: yes, exactly
11:03:39Bagderso, let's keep the auto-build of them and remove the download link until we've made them actually work
11:03:52edxkeys dont work in the simulator... *oops*
11:04:07Bagderme break keys ;-)
11:04:08edxi tree.c - what function is called to get the keys there
11:04:44Bagderedx: I pretended to be a win32 hacker the other day, I guess I failed
11:05:01edxhaha ;-)
11:05:34edxyou get a win32 hacker as slowly as you get a Linux hacker :]
11:07:22CtcpIgnored 4 channel CTCP requests in 4 minutes and 9 seconds at the last flood
11:07:22*edx uses his good old debugger :)
11:12:26 Join wettoad [0] (
11:14:18 Quit wettoad (Client Quit)
11:14:28*edx found the mistake (i guess..)
11:15:04edxwhat is supposed to be returned bybutton_get if no button has been pressed?
11:15:17Bagder0 I guess
11:15:26edxbecause the variable is never initialized
11:15:43edxBagder: you did it! :)
11:16:04BagderI know I did
11:16:13Bagderbecause before my fix it didn't even build
11:16:29edxlol yea
11:16:34edxi was about to add the function
11:16:37edxbut you had already
11:16:41edxso i just took yours
11:17:30*edx waits for the next builds of his site to check
11:17:59edxwhe i get back too school again and got some more time i also have to do some fixes with tetris ;)
11:20:18 Nick Zagor|breakfast is now known as Zagor (
11:20:27Zagorwin32 builds unlinked
11:21:06ZagorLinus: did you try the play-last-byte trick?
11:21:58Linuswe still have to fix the peek_next_track() issues
11:22:06Linusit is destructive
11:24:27Zagorhow do you mean destructive?
11:24:54Linusit updates playlist.index
11:25:47Zagorof course. who uses that?
11:26:11dwihnobtw, are you 100% sure mpeg_current_track() is error-free?
11:26:28LinusZagor: playlist_next() uses it
11:26:42Zagordwihno: that function is pretty safe, but the data could be wrong at times
11:26:54Linusthe index advances for each track that the mpeg thread loads
11:27:07 Join wettoad [0] (
11:27:12ZagorLinus: yes, by design. how is that bad?
11:27:41Zagorwe don't call playlist_next() on in-ram next-track
11:27:42Linusno, peek_next_track() was no supposed to advance the index
11:27:57Linushence the name peek_...
11:27:59dwihnoZagor: yeah, the parse display of the wps drops data because of id3->path (I only see the length decreasing the two first times entering and exiting the wps)
11:28:43ZagorLinus: right, but you still haven't told me why this is a problem today
11:28:47LinusZagor: next_track is not the problem, prev_track is
11:29:05Linuslet's say that the mpeg thread loads 4 tracks
11:29:08Zagornew_file needs to use an integer instead of boolean
11:29:08dwihnoZagor: thus, the ".mp3" removal gets screwed removing everything past the last '.'
11:29:19Zagorwe go back the number of tracks we have in ram
11:29:23Linusand then the user presses the PREV key
11:30:09Zagoror even just the next call does that trick
11:30:18Zagorkeeping new_file() unchanged
11:30:43Zagorcan you put up some of those short tracks of your for download somewhere?
11:32:50 Quit wettoad ("Client Exiting")
11:36:03*dwihno gotta get those
11:36:37Bagderw4r3z d00dz
11:37:25dwihnoFarbror Frej is RULEMAN #1! :D
11:40:12dwihnoHej alla barn nu blir det barnprogram! :)
11:40:34LinusZagor: the design idea was that peek_next_track() did NOT update the playlist index
11:40:41dwihnoANSIKTSMASK :D
11:40:52Linusand a separate next_track() function did that
11:41:09Linusthat is why peek_next_track() was invented
11:41:14BagderLinus is right
11:41:41Zagoryes, but now we don't have that any more. and I don't think playlist.c should need to know about how many files mpeg.c have cached or not
11:42:07Zagorit's cleaner if we can make do with just playlist_next(), imho
11:42:10BagderZagor: no, mpeg should pass an index
11:42:26Bagderjust next?
11:42:42Bagderhm, well it could work
11:42:47Zagorplaylist_next() handles both positive and negative "next" :-)
11:43:13Bagderso how would it be called?
11:44:03Zagorplaylist_next( -num_tracks_in_memory() )
11:44:43Linusi like that
11:45:24Bagderand it returns the file name of that index and set the index there?
11:45:39Zagorit already does that. we only need to fix mpeg.c
11:45:56Linusand it should go to the current song if pressed after x seconds have been played of it
11:46:24Bagderthat doesn't involve the playlist then
11:46:27ZagorLinus: that should be handled by mpeg.c, not playlist
11:46:43Bagderthe wps code should check the time
11:47:15Linusinstead of mpeg_prev()
11:47:27Bagdersounds fine
11:47:33Linusthe x second logic shouldn't be in mpeg.c
11:47:48 Part Bagder
11:47:55Linushurry! 1.2 is close
11:47:58 Join Bagder [0] (
11:48:04Linuswelcome back
11:48:05Bagderthe x second logic is wps
11:48:12Zagoryes, and it's post-1.2
11:48:20BagderI hate it when I auto-timout my ssh shell ;-)
11:48:32Linus999999 seconds passed?
11:48:41Bagdermore like 600 or something
11:48:44Zagorthe next_track fix is a bug, so I want that for 1.2
11:48:52LinusZagor: agreed
11:49:39LinusBagder: are you familiar with the relative path handling in playlist_next()
11:49:43Linusit bugs
11:49:58Bagdernot really, I just applied the patch we received
11:50:21BagderI don't use relative paths
11:50:31Linusdir-play does that
11:52:32BagderI don't use dir-play either ;-)
11:52:39Linusi know
11:53:04Bagderso how does it bug?
11:53:49Linusit puts two slashes between the current dir and the filename
11:54:15 Join matsl [0] (
11:54:20Bagdershouldn't the fs code work with that, I mean preferably
11:54:33Bagdernormal unix code does
11:54:35Zagori thought it did
11:54:38LinusBagder: you mean handle it?
11:54:47Linusit does
11:55:00Linusthe filename scroll in wps is ugly
11:57:44Linusi'm on the case
11:59:50matslcurius - whats ugly with the filename scroll?
11:59:58Linusit puts two slashes between the current dir and the filename
12:00:08Bagderwhen doing all-dir play
12:05:27Zagornow next works with Farbror Frej :-)
12:05:43Zagorprev still bugs
12:06:35LinusZagor: no, in-ram stunts on prev, please! :-)
12:06:49dwihnoHå hej med Farbror Frej! Hej alla barn nu är det barnprogram. Titta nu här vad Farbror Frej tar fram. En liten klubba och nu ska ni höra, vad man med sådana klubbor kan göra. Slå den i huvet och dunkeli-dur, så har vi gjort oss en skallfraktur! :)
12:06:51Zagorno, not yet :-)
12:07:52ZagorLinus: num_tracks_in_memory doesn't decrease when playing 2nd, 3rd track etc. is this intentional? this is why my prev doesn't work
12:08:29Linusdoesn't it?
12:08:49Linusit should, when it receives the TRACK_CHANGE event
12:08:58Zagorok. checking.
12:53:04notchZagor: Have you still got the really old IRC logs (december 2001?)
12:53:13notchcan U give me a URL?
12:53:20Zagorno, I don't
12:54:04Bagderwe didn't #rockbox back then
12:54:05Zagornow, finally, the irc log copy script works
12:55:22notchI remeber you having a few logs on the website around this time bit they seem to have been removes
12:56:05BagderI think you remember it wrong
12:56:09notchOh, it's the mailing list I'm thinking of!
12:56:27Bagdersounds reasonable
12:56:30Bagderit started back then
12:58:06***Saving seen data "./dancer.seen"
12:58:45 Quit MeRWiN (Connection timed out)
12:59:08LinusZagor: did you find the track change bug?
13:00:22Bagderbug 581987 "Playlist" could be closed now, right?
13:00:39Linuscould we ever reproduce it?
13:01:19BagderI doubt it
13:01:20ZagorLinus: which one do you mean?
13:02:11Zagorah, no not yet. still working on forward. it still bugs sometimes. claims "no more mp3 data" with 7 fresh tracks loaded...
13:02:33LinusZagor: exactly what i saw
13:03:08*Bagder would like the battery blinking when the level is very low
13:12:29Zagor(gdb) p last_dma_chunk_size
13:12:30Zagor$1 = -247312
13:13:06Zagoron "No more mp3 data"
13:13:16datazonesweet the archos multimedia should be out in 4 days
13:14:32mbrBagder: When should it blink? BATTERY_LEVEL_DANGEROUS?
13:15:04Bagdersounds suitable
13:15:32Bagdermatsl: ?
13:16:04matslBagder: Remember my scroll patch with a seperator string instead of blanks?
13:16:25Bagdernow that you mention it, yes
13:16:45matslWas it considerd a bad idea?
13:17:09mbrBagder: I added it to my todo for post 1.2 era :)
13:17:19Bagdermbr: goodie
13:17:44Zagormatsl: I think it just fell off the track. we'll test it post-1.2
13:18:11Bagdermatsl: if you can re-sync it with cvs and re-post it to the list, it would help us
13:18:12matslBagder: OK. So I don't need to resend it to the list then ;-)
13:19:48mbris it possible, that lcd_invertrect does not invert? Does the same as lcd_fillrect (beside clearing)
13:21:02Bagdermbr: I think you're right
13:22:23ZagorLinus: you here? I got this debug output:
13:22:37ZagorS R:5C60E W:1FFFE SW:1FFFE
13:22:37ZagorNo more MP3 data. Stopping. L:0 R:1FFFE W:1FFFE
13:22:52Zagor(the last ones are last_dma_chunk_size, buf_read, buf_write)
13:23:08Zagoris last_dma_chunk_size ever legally 0?
13:23:23Linusit shouldn't
13:24:11Linusi guess unplayed_space_left goes negative
13:25:29Zagorwhere? it's (of course) 0 when it stops
13:25:39Linusno, before that
13:26:39Linusi see the bug
13:26:47Zagortell me
13:26:49Linuslook at init_dma()
13:27:00Linuslast_dma_chunk_size = MIN(65536, mp3buf_write - mp3buf_read);
13:27:17Linusnegative chunk size
13:27:32Linusinit_dma() only works on empty buffers
13:27:53Zagorat least that line. i can reassign chunk_size after calling init_dma.
13:28:04Linusor compensate
13:29:16Zagori think init_dma should have a proper unplayed_space_left calculation, like in DEI3
13:29:23Linusmake the same adjustments as in DEI3
13:29:47Linusmaybe even make it a common function
13:30:32 Nick Linus is now known as Linus|lunch (
13:30:32DBUGEnqueued KICK Linus|lunch
13:36:32Zagorok, next works good now.
13:36:39Zagorprev still bugs
13:38:22matslbagder: ?
13:38:48matslBagder: Remember my m3u-only patch?
13:39:07 Join freshmaker [0] (
13:39:19matslBagder: Want it again or considered bad ide?
13:39:38Bagderwell, I can't say many people have cried for it to get applied
13:39:49Zagormatsl: filter out mp3 files too?
13:40:05matslThey don't knwo how good it is ;-)
13:40:07Bagderpersonally, I don't think it is a very good idea
13:40:27BagderI'd just move the m3u-files to a separate dir
13:40:33matslOK. Many things have changed so I'll submit it again.
13:41:09matslOOPS! I though you said it was a good ide. Sorry!
13:41:12Bagdermake it extend the mp3/m3u filter then
13:41:37Zagoryes. the filter is probably going to be expanded to an option instead of a boolean
13:41:43matslIt does. (off, on, m3u)
13:41:50Bagderah, ok
13:42:36matslBesides I have all my m3u alreay in one directory. Root ;-)
13:43:10Bagderhaha, so why the need for the filter?
13:43:25matslI don't wnat to see the dirs!
13:43:44Bagderso it scans recursively for all m3us?
13:44:04Bagderor how do you change dir?
13:44:28Bagderhiding the dirs is a bit weird in my mind
13:44:58matslNot if you have all m3us in one dir as you suggested.
13:45:08Bagderright, then you must have them in root
13:45:12Bagderas you can't change dir
13:45:15Zagormatsl: wouldn't it be good enough to change sort order, to show dirs last?
13:45:44matslMaybe that would be almost the same. Yes.
13:46:01Bagdermbr: I broke the revertrect() a long time ago when I made the bitmap use OR instead of XOR
13:48:13matslBagder: To change dir in m3u-only mode you need to go out of that mode.
13:48:15Bagder"bounce" didn't work with XOR ;-)
13:48:15CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
13:48:15*Zagor wonders if his 802.11 link covers the cafe on the ground floor...
13:48:24 Join grummel [0] (
13:48:42Bagdermatsl: sure, but a pretty annoying manouver
13:49:05mbrBagder: OK. on todo ....
13:49:09Bagderand not really something someone with m3us spread all over would use
13:49:17matslBagder: Agreed.
13:49:17Zagori suggest the sort fix rather than filter fix, if that's almost as good
13:49:22Bagderthen again, I don't know if people have
13:50:01matslAs I have the filter fix patch at hand I post it. Then people can look at it again.
13:50:20Bagdersure, posting a patch never hurts
13:54:21BagderI want multi-line scroll support
13:54:42Bagdernot sure if they all would scroll independently of each other tough
13:55:14matslbagder: I experimented with two scroll speeds.
13:55:42matslBoth lines scrolling at the same speed was confusing.
13:56:07BagderI see
13:56:19Bagderthen imagine 4 lines on the recorder ;-)
13:57:16matslI have been thinking of another type of scroll. One word at a time.
13:57:44Bagderseen my attempt?
13:58:07Bagdermeant to become a text-reader for the player
13:58:36Bagderdoesn't work 100%
13:58:49Bagderbut it shows one word at a time
13:58:57Bagderor more words, if they fit within 11 letters
13:59:17 Nick Linus|lunch is now known as Linus (
13:59:17DBUGEnqueued KICK Linus
13:59:48matslOK. Like some radio stations send out to the display? Seen that?
14:00:35Bagderit looks similar to that
14:00:54matslI've been thinking in terms of changeing the wps screen to show id3-info once in a while in that maner.
14:01:18Zagorahh, finally prev works too
14:01:34Bagdermatsl: might work, yes
14:01:52Zagori'm getting brain damage from those test tracks...
14:02:06Bagdereven worse than before? B-]
14:03:07matslbagder: It is hard to do with current scroll implementatiuon. The scroll needs to work better ...
14:03:23matsltogether with the other display stuff.
14:03:26Bagderso don't use it, make a separate impl in the wps
14:03:58LinusBagder: that's a veru good idea
14:04:10matslHmm... OK. Never occured to me. I was more thinking of adding showtext-like funtionality to the scroll-thread.
14:04:27matslLike text-properties or something.
14:04:56Bagderwell, if more than one screen needs it, then that is certainly better
14:05:15Linusactually, the scroll thread could be ravemped to a scroll_line() function
14:05:37Bagderthat would do what?
14:05:48Linusi don't like that the scroll thread updates the LCD independent of other activities
14:05:59LinusBagder: thread vs. function
14:06:20Linusget_button_w_tmo() and scroll in the default case
14:06:32Linusmuch nicer
14:06:32BagderI am +1 on that
14:06:49Linusthe you can scroll several lines if you want
14:07:22matslI got problems with two scrolling lines too since the menu display stuff needed to know about it too. It got messy.
14:09:15Zagori don't. i think get_button_w_tmo() is unfortunate
14:09:35Zagorusing that removes adjustable scroll speed, for one
14:10:12ZagorI think the status bar should go into a thread too, removing the timeouts in wps and tree
14:11:20LinusZagor: only if we get a lcd_update_rect()
14:11:26Zagori want all threads hanging on queue_wait() when nothing is running
14:11:37Bagderupdate_rect() is a piece of cake to add
14:11:42ZagorLinus: sure, but let's fix that then (after 1.2, of course)
14:12:07Bagderespecially if we make it update_row()
14:12:24Bagder8 pixels hight, full line
14:12:26Zagorah, right
14:12:35Linuswe will need a timer_event() functionality if we want all threads to hang in queue_wait()
14:13:05Linusby the way can we print text on an arbitrary pixel location?
14:13:06ZagorLinus: yes, i know. we're not there yet. it's just a vision
14:13:25Zagoryes we can. see Bounce
14:14:07Linusah, lcd_putsxy()
14:14:07BagderI want smooth scroll in 1.3 :-)
14:14:19LinusI want it in 1.2.1
14:14:29Bagderit is really easily added actually
14:14:39BagderI want the new font code in first though
14:14:42Zagori'll work on the font issues after the release
14:15:02Linusand the file system write
14:15:05Zagori hate the fixed-width font
14:15:09Bagderme too
14:15:11ZagorLinus: yes, that's #2
14:15:14matsloops! should be jumjum
14:15:38BagderZagor: I suggest you go with Alex's patch to start with, and then use Greg's code to generate nice internal defaults
14:16:14Zagorsounds goo
14:16:16Bagderand of course remove some memcpy()s from the code ;-)
14:16:17 Join MeRWiN [0] (~merwin@
14:16:22Bagderhey MeRWiN
14:16:37MeRWiNThis is the very first time i've ever been to work 3 hours early :) Usually i'm a good hour late
14:17:01Bagderaha, 3 hours for rockbox-hacking!
14:17:07CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
14:17:07*Bagder grins
14:17:13MeRWiNBagder: heh, yeah... my boss doesn't get in for another 3 hours
14:17:45MeRWiNhmm... is there any way to configure the CVS client for a SOCKS5 firewall? I can't do any cvs activity through this damn firewall at work
14:18:02matslWps question! I'd like next song not to start playing while song is stoped or paused. Keeping the mode moving back and forth. Seems logical. Comments?
14:18:14ZagorMeRWiN: you can run ssh through a http proxy
14:18:18BagderMeRWiN: well, can you get a ssh connection through it?
14:18:29Zagorso I guess there are ways to get it through a socks5 proxy too
14:18:35MeRWiNBagder: yeah... i've got ssh
14:18:57MeRWiNmatsl: how can the next song start playing when it's stopped or paused?
14:19:19MeRWiNmatsl: oh, if it's paused when you switch tracks
14:19:32MeRWiNmatsl: that's a good feature... one for the list of post-1.2
14:19:38BagderMeRWiN: so we add you as the sourceforge project member, then you can checkout over ssh
14:19:43Bagderas a
14:20:39MeRWiNI think I even have a sourceforge account, from some Serial LCD/MP3 Player that I was making a while back :)
14:21:20LinusZagor: how's the next/prev going?
14:22:14grummelI'd like to Fast Froward and REwind
14:22:25MeRWiNgrummel: wouldn't we all
14:22:31Linusgrummel: in time, in time
14:22:38MeRWiN <−−−− Now this is what Archos should be developing
14:22:41ZagorLinus: committed
14:22:43Linusit is pretty tricky to implement
14:22:59Zagoralthough now I seem to have woken up the old "play one track, get next" bug :(
14:23:07*Zagor runs for coffee
14:23:10 Nick Zagor is now known as Zagor|coffee (
14:23:11LinusZagor|coffee: i sent a subscribe to the announce list
14:24:02LinusZagor|coffee: you may want to resynch your CVS
14:24:12Linusmatsl: yes?
14:24:46matslLinus: Any problems with not starting to play song when doing next_song while paused or stopped?
14:24:55MeRWiNI think the sourceforge description needs to be updated to include the Jukebox Stuido as well :)
14:25:45MeRWiNmatsl/Linus: Couldn't we just have it mpeg_pause() after mpeg_play() if it's paused and the next/previous track button is pressed?
14:26:24Linusit depends on how Zagors new fix works
14:26:26BagderMeRWiN: new description there
14:26:43Linuswe can just refrain from starting the DMA
14:26:45MeRWiNBagder: 'scuse me?
14:26:53Bagderon the sourceforge page
14:27:06BagderI edited to include studio
14:27:15Bagderedited it even
14:27:40MeRWiNBagder: Interesting... "The Rockbox project aims to write open source firmware for the Archos Jukebox 6000 and Archos Jukebox Recorder MP3 players"
14:27:45MeRWiNmaybe it takes a bit to update
14:28:02MeRWiNi did
14:28:08Linusi see the new page
14:28:57MeRWiNI can't seem to see it
14:29:59*dwihno is acting phone support :(
14:30:12Linusphony support
14:30:30Bagderpony support!
14:30:46grummelwhat archos unit have YOU got?! I've got the AJBP 6000
14:31:10Linusand AJBP6k
14:32:17grummelHAve u updated your units in any way?!
14:32:25Linusnew hard disk in both
14:32:33dwihnoPoor me :(
14:32:38Linusserial port on Line-In
14:32:41BagderI run rockbox ;-)
14:32:46Linusof course
14:36:31 Join liam__ [0] (
14:36:49MeRWiNwhere is located?
14:37:14 Nick liam__ is now known as jedix` (
14:37:15notchWe have remote control funcionality ! :-)
14:37:15MeRWiNwhat if it's isn't there?
14:37:24notchWill post a diff soon...
14:37:33BagderMeRWiN: then you need to create one
14:37:39MeRWiNwhat do i put in it
14:37:44grummelis it possible to store codecs for WMA for example so that u can play it?!
14:37:48Linusnotch: c00l!
14:37:53BagderMeRWiN: ssh-keygen
14:37:54grummelin the archos.mod
14:37:54MeRWiNgrummel: not until we get a DSP expert
14:38:18notchtook me a while!
14:38:26MeRWiNBagder: what options do i use?
14:38:26Bagdernotch: nice work!
14:38:30grummelmy bacjklight seem to have a loose connection!
14:39:02BagderMeRWiN: it depends... what version of ssh are you running?
14:39:13MeRWiNnewest cygwin.. *shrug* just installed it
14:39:20dwihnoDamn user!
14:39:23dwihnoWhat should I say?
14:39:32MeRWiNdwihno: tell them to go to hell... go to hell and die :)
14:39:34dwihno"Pay me and I'll fix it for you, but I won't give you no telephone support!"
14:40:06BagderMeRWiN: ok, use "-t dsa" and then run again with "-t rsa"
14:40:24BagderMeRWiN: it generates keys to use for SSH v2
14:40:53Bagder(v2 as in the protocol version 2)
14:41:28MeRWiNBagder: now which one of those do I add into sourceforge? both?
14:42:07BagderMeRWiN: you can put both .pub files there, yes
14:42:41Bagderany or both
14:44:08MeRWiNOK, hopefully it worked :) I'm not much of a ssh/cvs guru. Very new to CVS indeed, but getting the hang of it
14:44:45Bagderyou'll learn
14:45:24MeRWiNeSVNC is a godsend... definately. True realtime VNC :)
14:46:11dwihnoMeRWiN: erps?
14:46:17LinusVery Nice Colors?
14:46:26MeRWiNdwihno: erps?
14:46:36LinusVirtual Nintendo Console?
14:46:55dwihnoVelour Night Cap
14:47:02MeRWiNLinus: full folor immediate full screen updates. Virtual Network Computing −−- Like PCAnywhere but free
14:47:13dwihnoMeRWiN: how is it compared to vnc, and tightvnc?
14:47:21Bagdervnc is cool
14:47:40MeRWiNdwihno: it's basically tightvnc with immediate updates.. Over a LAN it's like you're sitting right in front of the computer
14:47:46MeRWiNdwihno: win32 only though
14:48:09dwihnoMeRWiN: ah, okay... do you have URL?
14:48:56dwihnogotta check it out :)
14:49:10dwihnoI thought vnc was cool, but a bit sluggish, even on 100 mbit interfaces.
14:49:26Bagderof course true X rocks ;-)
14:49:29MeRWiNdwihno: yeah, it didn't have good change detection... this is frickin' immediate with no cpu usage
14:49:47MeRWiNBagder: I would use X if my company supported *nix ... stupid telecom companies :)
14:50:35MeRWiNdwihno: where do you work?
14:50:36dwihnoBagder: what window manager do you use?
14:50:49dwihnoMeRWiN: I don't - I'm full-time unemployed :-)
14:51:01dwihnoit's neato
14:51:06Linusdwihno: when are you coming to Stockholm?
14:51:12dwihnoLinus: I'm already here :-)
14:51:13Bagderoh, sawfish I mean
14:51:33MeRWiNwhy was peek_next_track removed?
14:51:38Linustime for a DevCon then!
14:51:49LinusMeRWiN: it was obsolete
14:51:51Bagderdevcon 2002, part 2
14:51:57dwihnoDevCon! DevCon! DevCon! DevCon! DevCon! DevCon!
14:51:59Linus1.2 release party!!!!
14:52:11dwihnoThat is a bit overambitious ;)
14:52:13MeRWiNLinus: oh, since everything is playlist driven now
14:52:30dwihnoLinus: We can bring a laptop and install rockbox and sign the rockbox t-shirts at stureplan ;)
14:52:39Linusgood idea!
14:52:42dwihnoFREE ROCKBOX COPIES! :D
14:52:53Bagderbring your own floppy :-)
14:52:57MeRWiNdwihno: BYOF
14:53:18MeRWiNYou should offer free rockbox installation service also
14:53:37dwihnoOr, we could sell floppies with the ajz and source code for 5 SEK a piece
14:53:57Linusand a (inkjet) printed manual
14:54:02Bagder'll sell *MILLIONS*
14:54:11dwihnoYou're gonna get RICH, Bagder :)
14:54:11MeRWiNdwihno: how much does 5 SEK work out to in USD?
14:54:16dwihnoMeRWiN: $0,5
14:54:29dwihnoA fortune! :D
14:54:38dwihnoSo Bagder is gonna get wealthy
14:54:42MeRWiNbah, sell them on CD then
14:54:48dwihnoYou just read my mind :)
14:54:55dwihnoMost ppl won't even have floppy drives anymore ;)
14:55:10MeRWiNthe most wasted space on a CD ever :) 649.95 MB avaialble of a 650MB disc
14:55:18Bagderwe sell it on 5.25" floppys only
14:55:44dwihno"and to use the floppy, you must buy this EXPENSIVE 5.25" floppy drive" :-)
14:56:02MeRWiNdwihno: now that's thinkin' like Microsoft :)
14:56:37dwihnoThank you, thank you! :)
14:56:44notchIve added a couple of files to the firmware, but when i do a cvs diff these files are not listed as added... Is there a way to add them to my 'local' repositry?
14:56:50Linusyou pay $5 for the CD, and $50 for the Activation Code
14:56:58MeRWiNLinus: *grin*
14:57:04Bagdernotch: only by doing 'cvs add [file]'
14:57:09Linusnotch: don't!
14:57:28Linusnot until the 1.2 release is done
14:57:37MeRWiNhmm... i usually join #linuxhelp as a keepalive for IRC (during the slow times in #rockbox) but this time i joined #linushelp
14:57:55Bagdernotch: and if you've checked out anonymous, you can't add them even to your local dir
14:57:57dwihnoMeRWiN: did you get any help with the linus? :)
14:58:09***Saving seen data "./dancer.seen"
14:58:16Bagderwe could use some help with Linus
14:58:24Bagderhe gets a bit too much at times B)
14:58:35*Linus is offended
14:58:38*Bagder ducks and hides
14:58:56*Linus knows where Bagder lives
14:58:57*dwihno scratches his head and sighs
14:59:08*dwihno knows who lives in Haninge ;)
14:59:17dwihnoor who was it now... ;-)
14:59:35notchso if I wanted to post these diffs to the list then I would attach them as well as sending a diff?
14:59:43notchthe files +diff
14:59:46Linusjust the diff
14:59:52dwihnoLinus: firemountains is a part of it :-) I used to live there myself
15:00:00Linusoh , and the new files, sorry
15:00:18notchbut the diff doesnt include my files as additons as I havent checked them in
15:00:32Linusyeah, sorry. include the new files too
15:00:38notchcheers :-))
15:00:47MeRWiNThis is great... I work for the largest Wireless company in the world and they feel it neccessary to say the following comment... to prove a point, one of the managers went to everyone's desk and stole all of the laptops that were left there overnight. This is the E-Mail that was sent out:
15:00:47MeRWiNEight laptops were left unsecured in cubicles on the 6th floor of Newport 5 last night. If your missing your laptop, check with Kelly Digregorio.
15:00:47MeRWiNEmployees who need a laptop to perform their duties should keep the laptops with them at all times. If you have a laptop and don't actually need one, please turn it in to your manager asap. Any laptops found in cubicles or anywhere else during off-hours will be removed.
15:00:49MeRWiNThis is serious stuff!! Both our company hardware and data assets must be protected at all times. Everyone is expected to do their part. If you see a laptop or any other company assets at risk, act to protect them.
15:01:01MeRWiNGreat office politics, huh?
15:01:37Linusactually, it's the same here
15:01:47MeRWiN*shrug* someone must thing there's a high rate of office-theft in a building where you need a keycard to get in the frickin' elevator :)
15:02:02Linusall employees with laotops have to take them home over night
15:02:12notchMan! starts with a ends with l and has na in the middle!!
15:02:34dwihnooffice theft?
15:02:37Linuswe also have keycards
15:02:38dwihnoMore than just pencils?
15:03:07MeRWiNLinus: they screwed up my keycard so I had to wait outside this morning for like 10 minutes because my after-hours access disappeared :)
15:03:26Bagderwatch this =>
15:03:34Bagdersee a tool you recognize?
15:03:40Bagderread the NDA at the side ;-)
15:04:25dwihnoBagder: Me see
15:04:38dwihnoMisa think that was cool :)
15:04:48Bagderhehe, yeah
15:04:50dwihnoBagder: now when I've kissed your ass, you can do my laundry
15:05:13MeRWiNHow is stuffit/winzip/curl a McAfee beta product?
15:05:21Bagderno idea
15:05:25dwihnoSue them!
15:05:28dwihnoYou can get lots of cash!
15:05:33dwihnoYou will get rich!
15:05:43Bagderand they just link to the sites, they don't offer the tools
15:05:46dwihnoEven richer than Linus with the rockbox 1.2 release ideas ;)
15:06:06 Quit MeRWiN (Remote closed the connection)
15:06:24Linusoh, now they took Merwins laptop too!
15:06:29dwihnoOh no! :(
15:06:33*Bagder laughs
15:06:44Bagderhe probably looked away
15:07:00notchokay how do you send a cvs diff result to a file...
15:07:08Bagder'cvs diff >mooooo'
15:07:13dwihnoWhere did all the honest people go (who only stole the office supplies instead of computers ;)
15:07:31 Join MeRWiN [0] (~merwin@
15:07:43Bagderdwihno: the honest thiefs were stolen ;-)
15:07:46MeRWiNWhich one of you guys wrote cURL?
15:07:54*Bagder raises his hand
15:08:00*Linus raises his
15:08:21MeRWiNall of the haxx brothers then :)
15:08:21dwihnoBagder: heh
15:08:24Bagdereven Björn has patched it ;-)
15:08:53LinusBagder: is the main developer
15:09:04MeRWiNwow, I've been listening to Italian Rock for the last 5 minutes and I haven't even realized that it wasn't in English
15:09:29Bagderyou're probably too busy holding on to your laptop
15:09:31*dwihno hypnotizes Bagder... domyla undry! domyla undry! domyla undry!
15:09:56dwihnoDo I sense a heap of not-so-good laptop theft jokes? ;)
15:10:14MeRWiNBagder: I leave my laptop in the office sometimes... although I'm in the office building that is anal about nothing EXCEPT parking. They hand out tickets for everything
15:10:26MeRWiNIt's the other office building that is anal about office politics
15:16:14Linusi wonder when zagor is going to commit his next/prev fix?
15:16:26MeRWiNdidn't he do that?
15:16:31LinusZagor|coffee: stop drinking coffee!
15:16:40LinusMeRWiN: he said he did
15:16:45MeRWiNFixed bugs in next/prev handling. Replaced peek_next_track() with playlist_next().
15:17:14Linusstrange. i got no commit mail
15:17:18Bagderme neither
15:17:30Linusand cvs diff gave nothing
15:17:34MeRWiNLinus: he committed 3 patches
15:18:36MeRWiNmine updated... peek_next_track is nowhere to be found in my tree.c anymore
15:18:42 Quit edx ("brb")
15:19:13Linusthen we should have no showstoppers
15:20:20MeRWiNif it works :)
15:20:34MeRWiNLinus: you should test it out
15:20:40MeRWiNthis should speed up track changes, right?
15:20:51MeRWiNby a lot?
15:20:59Linusnext, but not prev
15:21:51MeRWiNmost people go forward anyways
15:22:22Linushe hasn't taken the watermark into account
15:23:06MeRWiNLinus: hmm... if there's a low watermark then it doesn't buffer more or something?
15:23:14Linuspossible playback stop
15:23:52Linusit may start the playback of the next song way below the watermark, so the disk won't spin up in time
15:25:17Linusouch again!
15:25:45Linushe calculates the unplayed space too early
15:26:38Linusi wish he could finish his coffee
15:26:50MeRWiNLinus: that's one huge cup of coffee
15:27:00MeRWiNLinus: how prevalent is english in sweeden?
15:28:25Linusthey teach english in school from age 10
15:28:45dwihnoone two tree :-)
15:28:53Linusfree four
15:29:01MeRWiNwow. I was wondering why everyone from Ericsson spoke good english :)
15:29:21Bagderwe're also fed with all the american movies and TV shows, subtitled
15:29:25Linusas nobody else but us speaks swedish, we have no choice
15:29:47MeRWiNThat is true I guess
15:30:17dwihnoWe also broadcast Ricki Lake. She has taught us everything there is to know about the american lifestyle ;)
15:30:40MeRWiNHmm... not sure if Ricki Lake is the best model american. I'd choose Jerry Springer for that
15:30:57Linusor Dave Letterman
15:31:00 Nick Zagor|coffee is now known as Zagor (
15:31:01 Quit WetFlax (Read error: 104 (Connection reset by peer))
15:31:02Zagorbad news
15:31:14dwihnobad news or REALLY bad news?
15:31:22Zagormy wlan doesn't cover the local coffee shop
15:31:35Linusoh my!
15:31:36dwihnoThat's a catastrophe!
15:31:50Linusyou'll have to move the coffee shop
15:32:01Zagortime to buy an external antenna
15:32:11MeRWiNZagor: ouch... We've got wlan on a ton of Starbucks coffee shops in the states :)
15:32:35LinusZagor: your mpeg fix were not entirely correct
15:32:42Zagori know
15:32:48Zagor"I seem to have woken up the old "play one track, get next" bug :("
15:32:55Linusnot only that
15:32:59MeRWiNZagor: one of my co-workers gets the tough job of going to every one of the Starbucks that we have coverage in and testing there drinking coffee all day
15:33:00Zagorwhat else?
15:33:12Linussee my commit for starters
15:33:14MeRWiNZagor: let's just say you won't be drowning in the watermark
15:33:16ZagorMeRWiN: it's a shitty job, but someone's gotta do it :)
15:33:50ZagorLinus: ah, right
15:33:59MeRWiNZagor: yeah, he's the sort of person that drinks about 4 very large cups of coffee a day, so it's perfect for him
15:33:59Linusand you still don't take the watermark into account
15:34:12 Join WetFlax [0] (
15:34:24Zagorwhen does DEI3 run?
15:34:34LinusZagor: is your CVS in sync?
15:34:49Linusi fixed a "play one, get the next" bug
15:35:09LinusDEI3 runs when the DMA buffer is finished
15:35:16 Join ah_ [0] (
15:35:20Zagori think my new_file() change added a new one
15:39:44ah_I try to test daily build for two weeks now and i've a NO FAT 32 PA (can't see more font is big ). Got a jukebox 6000 , internal firmware 4.53 and i replace the HDD with a 30 gb IBM. Ver 1.1 works great.
15:40:24BagderPA ? Partition A... ?
15:41:56ah_the font is really big so the message could be longuer. Jukebox boot , certainly loads the rockbox firmware then i've 'No FAT32 pa'
15:42:58MeRWiNNo FAT32 Partition
15:43:06MeRWiNah_: is the drive formatted as FAT32?
15:43:39MeRWiNLinus/Bagder: That could be another fix, make the font smaller immediately after displaying the Rockbox! message so errors are displayed properly
15:45:21ah_i think that it couldn't work. It works with official firmware and rockbox 1.1
15:45:59Bagder"No partition found, trying to mount sector 0."
15:46:14ah_sorry incomplete message: i think it's fat32 formated
15:46:15Bagderthat's what it says right before the error
15:47:13Bagderfat.c is only changed twice since 1.1
15:47:20matslOff for other things. CU!
15:47:32MeRWiNah_: i believe the original archos firmware only works with fat32 also...
15:47:34LinusBagder: disk.c is involved
15:47:42 Quit matsl ("Liece")
15:48:23Linusah_: is your drive a master or a slave?
15:48:41Zagorah_: do you have a linux machine close?
15:48:47Zagori want to see the partition type id
15:49:04Zagorwe added checking for that, and apparently your partition is not one of the accepted
15:49:35MeRWiNah_: what operating system do you use?
15:50:27ah_here xp , got an old linux pc but no usb on it and my 2.5 - 3.5 converter is at work
15:50:29LinusZagor: maybe an extended partition?
15:50:45ah_will try to find tools
15:50:58Zagormaybe. we need to find out
15:51:01ah_maybe partition magic can do that ?
15:51:09Bagdersounds possible
15:51:27Zagorno, 1.1 didn't handle extended partitions either
15:51:35Zagorit must be the type id
15:51:56 Nick notch is now known as notch|busy (
15:51:56DBUGEnqueued KICK notch|busy
15:53:27ZagorFailed opening dir
15:53:27ZagorCouldn't open file: /Farbror Frejp/Farbror Frej - CD-rom.mp3
15:53:48Zagorhow the heck did i do that :(
15:54:12 Quit notch|busy ()
15:54:16Bagderan extra p?
15:54:39Zagoryou're right
15:54:59LinusZagor: that could be my fault
15:55:06Bagderbad slash fix perhaps
15:55:12Zagori guess so
15:55:31MeRWiN <−−−− shows all known partition identifiers
15:55:38Bagderchange the line under Linus' fix
15:55:51Zagorah, right
15:56:16BagderMeRWiN: good digging
15:56:24 Quit WetFlax ("I like core dumps")
15:56:27Zagorlinux fdisk knows them too :)
15:56:48MeRWiNZagor: yup, but i'm not running linux :)
15:56:50Zagorwe're accepting b and c
15:56:52Bagderwe only need which ah has now
15:57:19LinusBagder: did I put the \0 at the wrong place?
15:57:36ZagorLinus: yes
15:57:38BagderLinus: yes
15:57:43Linusforgot to change that
15:58:01MeRWiNcould be 0B, 0C, 1B, 1C,
15:58:02Bagderit seems as if 1b and 1c are possible too
15:58:22Zagorright. ah_, how's it going?
15:58:44 Join mecraw [0] (
15:58:50MeRWiNBagder: would 0f be a possibility? Extended partition (LBA)?
15:59:04Linusthat would be *so* XP, to create a hidden partition!
15:59:31Bagderperhaps f, yes
15:59:32MeRWiNLinus: never doubt the "creativeness" of microsoft
15:59:33ZagorMeRWiN: we don't support extended partitions at all right now (they require extra parsing)
15:59:33LinusMeRWiN: yeah i thought of that
15:59:44Bagder"Windows NT does not recognize the four W95 types b, c, e, f "
16:00:07MeRWiNBagder: so F is a valid Fat32 type (if the extra parsing were there)
16:00:13LinusBagder: winnt doesn not support fat32 at all
16:00:26Bagderah, right
16:01:21MeRWiNTime to move my car to the parking garage
16:01:24 Nick MeRWiN is now known as MeRWiN|Movin (~merwin@
16:02:56Bagder"Windows 95/98/Me: Partition that uses extended int13 services."
16:03:13Bagderand there's a PARTITION_XINT13_EXTENDED
16:03:58Bagdergod knows what real values they have
16:05:33LinusZagor: you must not start the DMA until the watermark is reached!
16:05:40Linususe play_pending
16:06:20 Nick MeRWiN|Movin is now known as MeRWiN (~merwin@
16:06:33ZagorLinus: you mean if next goes below watermark?
16:07:04ZagorLinus: can you fix this? you're more at home with the dma and watermarks
16:08:34Zagorah_: are you there? i'm preparing a module for you to test. which rom version do you have?
16:09:01Zagorah, he said so. good.
16:16:03MeRWiNah_: is the partition the default one that came with the drive or did you re-partition/reformat?
16:16:11CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
16:16:11*Zagor is anxiously awaiting the result...
16:16:23 Quit elinenbe ()
16:19:10ah_still the same erro
16:19:19Zagorok, hang on
16:19:22BagderZagor: output the found id
16:19:34Bagderand reduce the font first
16:19:39grummeldoes the rockbox charde the batteries in us mode?!
16:19:46grummelusb mode
16:19:57Bagdervery good question
16:20:21MeRWiNgrummel: player or recorder?
16:20:28 Join elinenbe [0] (
16:20:49MeRWiNgrummel: I believe it is. I believe it's always charging when the charger is plugged in
16:21:01MeRWiNgrummel: the charging is hardware-controlled
16:21:02Linusplayer has hardware controlled charger
16:21:05ah_MeRWiN i think that the travelstar comes unformated and i formated it on a win2k server i think , it was a year ago.
16:21:21MeRWiNah_: ahh... ok.
16:21:25ah_using win tools :(
16:21:25Zagorah_: test again
16:22:17grummelok, I know most of us here are from sweden and I'm to(in stockholm), now I'mn goin out to get new batteries, have u got any recommendations and where to buy?!
16:22:37Zagorgrummel: claes ohlsson, teknikmagasinet or some such place.
16:22:45grummelGP batteries?!
16:23:03Linusand get the 1hour charger too. it rocks!
16:23:24Linusand it costs!
16:23:26MeRWiNLinus: 1 hour charger?
16:23:32Linusfast charger
16:23:38Zagorit's lovely
16:23:49Linusand you can fry an egg on the batteries
16:23:58Bagderportable kitchen
16:24:07MeRWiNLinus: can't be very good for battery life though
16:24:26Linusnot really
16:24:46ah_6 0 0 0:No ???????
16:25:02Zagorah_: right. that's the partition types
16:25:03Linusi think Uwe is on to something in the mailing list
16:25:05MeRWiNSo does the max voltage output change if you use a different mAh battery?
16:25:12Zagor6 is fat16
16:25:18ah_hu ?
16:25:22Linusthat might explain why our charger stops too early
16:25:25Zagorare you sure 1.1 worked on this?
16:25:31ah_yes for sure
16:25:34MeRWiNah_: you need to reformat
16:25:47ah_and that windows say that it's a win 32
16:25:51Zagordoes format really alter the partition table?
16:26:01MeRWiNZagor: it should
16:26:15MeRWiNah_: In the properties it actually says FAT32?
16:26:42ah_will try to find a soft like partition magic
16:26:45MeRWiNah_: if you don't have anything too crucial on that drive i'd reformat it
16:26:48LinusZagor: the partition type shouldn't matter
16:26:51grummeldo u know what name the 1 hour chargers has got?!
16:27:11MeRWiNah_: it could have been a windows screwup the first time you formatted it
16:27:16ZagorLinus: true. we should simply try and mount all partitions in order.
16:27:32Linusat least peek at the boot block
16:27:36Zagorgrummel: GP PowerBank Smart
16:27:51Linusi said it costs!
16:28:00MeRWiNgrummel: how much is that in USD?
16:28:03Zagoryeah, it's pricey. mostly useful for us developers, i'd say
16:28:09ZagorMeRWiN: about $70
16:28:20MeRWiN$70?!@#@%?! damn, that's expensive
16:28:52grummelhow long does the 1800 batteries last after a full charge withthat?!
16:30:25Zagorabout 10h with stock firmware
16:30:42grummeland with rockbox?!
16:30:53MeRWiNand about 5h after it's charged 20 times with the rapid charger :-)
16:30:54Zagori don't know. I always mess too much with it to know :)
16:30:56grummelbut teknikmagasinet has got 1900mah batteries!
16:31:37ZagorMeRWiN: not really. nimh batteries can be fast-charged, unlike nicd
16:32:06MeRWiNZagor: hmm.. maybe i should look into one of those fast chargers then
16:32:12MeRWiNhow long does the normal charger take to charge?
16:32:24Zagorabout 6-15 hours
16:32:47 Join notch|busy [0] (
16:35:07Zagorah_: new test
16:35:07 Quit ah_ (Read error: 104 (Connection reset by peer))
16:35:15dwihnoAh, I just thought of something... Will it be possible to crossfade two tracks?
16:35:22Bagderdarned peer
16:35:26Zagordwihno: no
16:35:29dwihnoSankte Peer
16:35:37dwihnoZagor: too bad :/
16:36:09Zagorwe only have one mp3 decoder
16:37:00dwihnoAn idea: To make pressing 'next' more sufficient, maybe the first reads on a "new" file should be made smaller?
16:37:36Zagordwihno: they already are
16:37:51dwihnothey are? :)
16:37:57dwihnoThen why is it so slow?
16:38:19Linusit is the playlist loading
16:38:39Linusand the ID3 parsing
16:38:39 Quit MeRWiN (Remote closed the connection)
16:38:52dwihnoid3 parsing should be optional :)
16:38:57Linusand the drive needs to spin up each time
16:38:58dwihnoAnd playlist loading too ;)
16:39:21dwihnoWhy not keep id3 parsing optional? :)
16:39:25*Zagor thinks the remote patch looks very strange. Why all those #ifdefs?
16:39:42 Join MeRWiN [0] (~merwin@
16:39:45MeRWiNInteresting ... I just read that storing NiMH batteries in the freezer prolongs their shelf-life :)
16:39:56ZagorMeRWiN: really?
16:39:57Linuswe could stop looking for V1 tags if we found a V2 tag
16:40:04*Zagor moves his spare ones to the kitchen
16:40:22dwihnoMeRWiN: freezer? not refridgerator?
16:40:35BagderLinus: I bet the difference won't be noticable though
16:41:10Linusthe longest wait is probably for the spin up
16:41:36dwihnoLinus: why does it have to spin up if I switch track in less than 5 secs?
16:41:38ZagorMeRWiN: oh, it's for keeping the charge
16:41:58Linusdwihno: because someone put an ata_sleep() in the mpeg.c code
16:42:04*Linus points to Zagor
16:42:18Bagderthat's what MeRWiN and I've been on about
16:42:23dwihnoStop blaming Zagor for everything! :)
16:42:39*Linus keeps shut about the fact that he did it first
16:42:56MeRWiNBagder: what have we been on about? ata_sleep?
16:42:58dwihnoLBZ - Let's blame Zagor!
16:43:04Zagorwe might want to wait with the spindown on the first buffer read of a track
16:43:08notch|busyZagor - Feel free to change them - they were for player/recorder compat.
16:43:09dwihnoSo why is there an ata_sleep in the mpeg code?
16:43:14BagderMeRWiN: yes, about mpeg spinning down the disk
16:43:24Linusdwihno: to save batteries
16:43:33MeRWiNBagder: I think the ata_sleep() should disappear from that code.
16:43:36Zagornotch|busy: yes, but you've put identical code in many of them :-)
16:43:37Bagderme too
16:43:53grummelshould I buy the 1 hour 699kr charger or the 2-5hour 499kr?!
16:43:53dwihnoLinus: just move the spindown after 5 secs then :-)
16:43:54MeRWiNLinus: it really doesn't save that much though in the long run. It's only 5 extra seconds of being spun up
16:44:09MeRWiNgrummel: put in the extra cash for the 1 hour
16:44:15notch|busyI thought I would keep them as this may need changing at a later date...
16:44:49notch|busyi.e. I havent tested this on a player :-)
16:45:12ZagorMeRWiN: I smell an option :)
16:45:30dwihnoOptions to the people!
16:45:48dwihnoRockbox is already 10 times more configurable than the stock firmware :)
16:47:57MeRWiNZagor: that's what we were saying before. heh.
16:49:14dwihnoOr as we say in Iran - Frekicheee! :D
16:49:32Zagorooookay, any more known bugs stopping us from releasing?
16:49:56dwihnoparse mode stripping data from the filename
16:50:09Linusdwihno: please explain that one
16:50:37dwihnoI keep my album filenames in the format XX. blablablablabla.mp3
16:50:44dwihno1) I start a track
16:50:52dwihno2) I press 'on' (dirbrowser)
16:51:07dwihno3) I press 'on', return to the WPS and I see only XX
16:51:37dwihno(no id3 tag)
16:52:10MeRWiNWhat is parse mode?
16:52:23Zagorwps mode "parse"
16:52:32Linusi think parse is broken now
16:52:46MeRWiNWhat does the Parse mean though
16:53:27Bagder"extract info from filename"
16:53:27dwihnoMy "ugly fix" involves commenting line 100-101 in wps.c just to make it go away ;)
16:53:28BagderI guess
16:53:37MeRWiNThat will never work :)
16:53:45MeRWiNToo many different ways of naming filenames
16:54:07Bagderwell, if there's no id3 info, showing parts of the path on screen is likely to be more useful than just blank
16:54:26MeRWiNBagder: i'd rather just see the filename if there's no id3 info.
16:54:28Linusyes, but a scrolling filename should be enough
16:54:40BagderI don't think so
16:54:49BagderI'd rather fill up the blanks with static info from the file name
16:54:59Bagderotherwise I get three empty lines
16:55:03Bagderand one scroll
16:55:30MeRWiNBagder: Empty space is not always a bad thing... find something else to fill it with, like the bitrate and filesize or something :)
16:55:33 Join nsauzede [0] (
16:55:35Linusthat is what you get with "parse" now as well, i think it's broken
16:55:51BagderMeRWiN: it's already there, even when the id3 info is complete
16:56:13nsauzede{hi all}
16:56:17Bagderthen again, I have id3 tags on almost every mp3 I own
16:56:24Bagderhey nsauzede
16:56:35MeRWiNBagder: hmm, well, i dunno. I'm personally not much for the parse thing, but if we can find a way to get it to work reliably then it'll work fine for some people i guess
16:56:45dwihnoLinus: I also noticed the id3->path length decreased when leaving&returning to the WPS...
16:57:12Linusthat is probably a strtok() "feature"
16:57:13dwihnoLinus: that's why I got a bit suspicious about the code retrieving the path from the list
16:58:11***Saving seen data "./dancer.seen"
16:58:50Zagori don't think i'll be holding 1.2 for the parse mode. i've always considered it a test
16:59:13BagderI'm on your side Zagor
16:59:13MeRWiNZagor: i think you should disable parse mode for the 1.2 release
16:59:23Linuswe have made a lot of changes today
16:59:24Bagderyeah, you can #ifdef the option
16:59:37Zagorthat's a good idea
16:59:42Linuswe should wait with the release until tomorrow
16:59:46ZagorLinus: i agree
17:00:06Bagder#ifdef now, though, then we can all update cvs and run
17:00:20Linuswe should fix the double-height error messages
17:00:21Bagderand check for the last minute breaks
17:00:22Zagori won't include the partition type fix either
17:00:28Linusor are they double-height?
17:00:35Bagdernah, fix that after 1.2
17:01:04Bagdersmall error messages is a new feature ;-)
17:01:12dwihnoBagder: you need to modify your signature, it still has the old rockbox URL :)
17:01:40nsauzedebagder: nice job with the rockbox manual
17:01:40Linusgotta catch the train now
17:01:48*Bagder bows
17:01:49Linusbye guys!
17:01:53Bagdersee ya Linus
17:02:14Zagoris the charger enabled on rec20 only, or on all?
17:02:24Bagdernsauzede: it still needs a lot of improvements though
17:02:28Linusonly recorder of course
17:02:47Zagordoh, right. the others can't be controlled...
17:02:51nsauzedebagder : I just found 1 dead link,
17:02:59 Part Linus
17:03:02Bagdernsauzede: url?
17:03:37nsauzedelink : back to main page
17:03:44 Join ah_ [0] (
17:03:50nsauzedeoops user manual front page indeed
17:03:54Zagorah_: can you test again?
17:03:59ah_Ok thanks for all, got a long car trip tomorow ( 2000 km) , so no times to reformate my drive , when i come from vacation will make some test before transfert and format my drive. Sounds crazy but i think that the drive is fat32 formated with a bad id type.
17:04:03ah_zagor yes
17:04:27ah_have been disconeted
17:04:39ah_can you give me the url back
17:04:48ah_you remove the fat32 test ?
17:05:02Zagoryes, that version tries to mount all partitions
17:05:13nsauzedebagder: 1 question plz
17:05:29Zagornsauzede: now there should be no bad links
17:05:29Bagderfire away
17:05:44nsauzedezagor : cool !
17:06:06nsauzedebagder : do u remember when you asked me to produce details about
17:06:21nsauzedehow to compile win32 sim with lcc-win32 ?
17:06:45dwihnoCheck that out nsauzede
17:06:58dwihnoit's not lcc, but mingw
17:07:12nsauzedeerrm, thanks dwi,
17:07:28nsauzedebut I *DID* try to compile with lcc-win32 :-)
17:07:43nsauzede(which is another free win32 C compiler)
17:08:00ah_Zagor works fine
17:08:22BagderZagor: that fix should probably go in though
17:08:23Zagorah_: nice!
17:08:57ZagorBagder: maybe. i'm undecided...
17:09:10 Quit mecraw ()
17:09:12Bagdernsauzede: so what's the q?
17:09:29BagderZagor: I think so, for ah's sake and others in his position
17:09:42nsauzedebagder: ok : so I recently re-tried to compile using lcc the recent cvs win32 sim,
17:09:59ah_hum what does that mean ? that the partition is fat32 but that the id type is wrong ?
17:10:45nsauzedebagder: but I encountered problems, I believe that file dependencies have been added in Makefile in the mean time,
17:11:13MeRWiNah_: yeah
17:11:15ah_hey cool i didn't sea the modification since 1.1 , usb without on/off is really great
17:11:21Bagdernsauzede: what makefile are you using
17:11:25nsauzedebagder: and the problem is that lcc is not a gcc clone, and don't offer as many command line switches...
17:11:44nsauzedethe one created by configure
17:11:49nsauzedeI guess :)
17:12:14nsauzede(modified a little to fit lcc)
17:12:34Bagderso if it is recent code, it should work properly
17:12:59nsauzedebagder: indeed, my question is : as someone offered ming-win32 possibility to compile the win32 sim,
17:13:17Zagorah_: yes, it's fat32 formatted but the id says it's fat16
17:13:29nsauzedeis it worth trying to maintain lcc compatability ?
17:13:58nsauzedeming-win32 is a clone of gcc right, what is not lcc
17:14:03nsauzedeyour opinion ?
17:15:06Bagderif mingw32 users on windows can use the mingw-makefile put there right now it would be the coolest
17:15:12Bagderbut right now I'm not sure about that
17:15:21Bagderand if not, we have neither mingw nor lcc makefiles in the tree
17:15:31Bagderso either one would be an addition
17:15:31ah_Zagor, win tools reconize it as fat32 , they maybe use other way that type id to check the file system and that the disk manager tools version i used a year ago to formated is bugged.
17:16:15nsauzedebagder: ok, I'll see what I can do then..
17:17:02Zagorah_: they only check the file system, not the partition type
17:17:28 Join edx [0] (
17:20:15ah_Hum type id check is a good idea. I'll reformat my disk when i come back from vacation or try to change the type id manualy. Hope that i'm was the only guy to use this bugged version of win2k disk manager :p
17:22:22ZagorBagder: i think i'll hold the mount fix. it could introduce problems, and ah_ is the first i've heard with the problem
17:22:34Bagderyou're right
17:23:16Bagderthe mailing list is getting quite busy lately
17:23:17ah_so thanks for that 'personal build' :p
17:23:27Zagorah_: no problem. thanks for helping
17:23:53MeRWiNBagder: I've got a pre-built Cygwin package for the SH1 cross compiler... seems as if it may be easier to have the windows users just install cygwin and drop in my package
17:24:30Bagderfor cross-compiles indeed
17:24:34Bagderbut for building the sim?
17:24:59MeRWiNBagder: how do you build the sim? just go into the sim folder and hit make?
17:25:15Bagdernot really
17:25:18MeRWiNor actually, it would be a different makefile
17:25:33Bagderyou can run the configure script, right?
17:25:44MeRWiNi was smoking crack a second ago :)
17:25:56Bagderif you run that and select win32-sim, it makes a makefile suitable for mingw32 builds...
17:26:21Bagderbut we could pretty easy make the compiler selectable and run either mingw or cygwin
17:27:04nsauzedecygwin for producing win32 binary ?
17:27:39Bagderat least I guess so
17:27:48BagderI'm not the one who'll do it ;-)
17:27:57nsauzedewhat you mean by : cygwin-compiler ?
17:28:11Bagdernsauzede: if you install cygwin you get a compiler
17:28:32nsauzedeok, and ... errm
17:28:38nsauzedeI'm confused here??
17:28:58nsauzede=> I'll check that now :-)
17:29:03nsauzedegreat idea !
17:29:58nsauzedebagder : (gcc −−version returned me 2.95.3-4 => I'll see what I can do with it) −− bye all !
17:30:25Bagderhehe, bye nsauzede
17:30:33 Quit nsauzede ()
17:32:55 Quit PiotR ("El software es como el sexo: es mejor si es gratis. -- Linus Torvalds")
17:32:55Bagderdwihno: you mind if I steal your "shirt" outline for my upcoming design contribution?
17:36:32Zagori'd say it's time to link to the /manual/. it's not complete by any means, but it's the best we've got.
17:36:57Bagderthrowing people at it is the best way to make it improve as well
17:37:25Zagoryes. plus a stern order that anyone who changes the UI *must* also change /manual/
17:37:51Bagderwe need more screen dumps
17:38:02Zagorhow about that ui function? :-)
17:38:17Bagderuse gimp ;-)
17:38:37Zagori know
17:40:26Bagdersome screens aren't possible to get using the sim though
17:40:30Bagderlike wps
17:40:31Bagderand usb
17:40:53Zagorwe had wps before my vacation. what happened with that?
17:41:13BagderI don't know
17:45:37 Quit pyvasene ("using sirc version 2.211+KSIRC/1.2.1")
17:54:01*edx needs photos of an archos device...
17:54:11edxare there any photos on the websites
17:55:04Zagori'll add them to the photo page too, so they're easier to find
17:57:41edxyea - ive looked ther ebut didnt find them
18:06:27Zagorhow did you find that?
18:06:43Bagdernah, googling
18:06:54Zagorlink searching?
18:07:24Bagderjust 'rockbox archos'
18:08:33Bagder, listed under "Custom/Specialized/Crazy Computer Systems" ;-)
18:08:46BagderI like crazy
18:09:06BagderI'll continue this from home instead, see ya soon
18:09:07 Part Bagder
18:17:44Zagorbjorn:~# apt-get -u upgrade
18:17:51Zagor197 packages upgraded, 0 newly installed, 0 to remove and 3 not upgraded.
18:17:51ZagorNeed to get 81.2MB of archives.
18:18:48*Zagor likes fat pipes (eta: 4m6s)
18:21:16notch|busynotch is jealous....
18:25:08notch|busybye all...
18:25:13 Quit notch|busy ()
18:34:51 Join Bagder [0] (
18:35:41ZagorBagder: the sim wps seems to bug due to missing lcd_update()
18:36:49Zagoror maybe because mpeg_is_playing never returns true
18:41:07Zagoryeah, that's it
18:41:21 Quit Bagder ("disconnecting from stoned server.")
18:41:25 Join Bagder2 [0] (
18:41:29Zagorbah, it still bugs bad :-(
18:42:01Zagorcode rots fast around here :-)
18:42:08Bagder2it sure does
18:42:12DEBUGEOF from server (Success)
18:42:12***Saving seen data "./dancer.seen"
18:42:17***Started Dancer V4.16p1
18:42:17***Connected to on port 6667
18:42:17***Logfile for #rockbox started
18:42:23***Server message 501: 'logbot :Unknown MODE flag'
18:42:23Mode"logbot :+i" by logbot
18:42:23 Join logbot [242] (
18:42:23 Join Bagder2 [0] (
18:42:23 Join edx [0] (
18:42:23 Join ah_ [0] (
18:42:23 Join MeRWiN [0] (~merwin@
18:42:23 Join elinenbe [0] (
18:42:23 Join jedix` [0] (
18:42:23 Join grummel [0] (
18:42:23 Join freshmaker [0] (
18:42:23 Join Zagor [0] (
18:42:23 Join Synthe [0] (
18:42:23 Join ironi__ [0] (~ironi@
18:42:23 Join adi|home [0] (
18:42:23 Join fragglet [0] (
18:42:23 Join datazone [0] ([0x72LyCYj@
18:42:23 Join webmind [0] (
18:42:23 Join matp|work [0] (
18:42:23 Join mbr [0] (
18:42:23 Join Tumm [0] (
18:42:23 Join adi|work [0] (~adiamas@
18:42:23 Join Hadaka [0] (
18:42:23 Join dwihno [0] (dwihno@Bald067.Baldakinen.Umea.SE)
18:42:23 Join Hes [0] (
18:42:23 Join PsycoXul [0] (
18:42:32Zagorah, welcome back logbot :-)
18:43:24Bagder2can gimp scale a single layer ?
18:43:26Zagorbtw, I installed 'nullidentd' yesterday. a 150-line dummy ident server that gets rid of the annoying delay when connecting to irc servers
18:43:33ZagorBagder2: no idea
18:44:11Zagoreverything is soooo easy with debian. i'm in love :)
18:44:46Zagori just updated 197 packages in ~10 minutes
19:00:49 Quit ah_ (Read error: 104 (Connection reset by peer))
19:00:54Bagder2my contribution added
19:01:42Zagori like the tagline. "How hard can it be?"
19:02:51Zagorah, now sim wps works. sweet
19:04:09 Nick Bagder2 is now known as Bagder (
19:04:30Zagorprev/next doesn't work, though
19:04:53Bagderdoesn't surprise me
19:05:12Bagdermakes a good screen dump anyway
19:05:15Zagorno, not really. but it shouldn't be too hard to make it work
19:07:48MeRWiNBagder: hehe.. i like your contribution
19:08:14BagderI like it with more attitude ;-)
19:17:03 Join MeRWiN- [0] (~merwin@
19:17:41 Quit MeRWiN (Remote closed the connection)
19:20:11Zagorwhat is the difference, charging-wise, between player-old and player-new ?
19:20:23Zagoris there any?
19:20:40Bagdercharging wise?
19:21:07Zagoryes, i remember linus mentioning a difference between them but I can't remember what
19:21:26Bagderno clue
19:21:34Zagornewer players apparently start up the disk when the batteries are full, while the older ones don't.
19:21:47Zagorneither can turn off charging
19:25:07 Nick MeRWiN- is now known as MeRWiN (~merwin@
19:26:27elinenbebug: in the keylock on screen on the recorder the name of the track still scrolls
19:26:33 Part elinenbe
19:26:42 Join elinenbe [0] (
19:27:51Zagorelinenbe: yeah, that's because we haven't added a scroll_pause/resume yet
19:28:55Zagorlet's call it a feature until then :-)
19:29:08elinenbea good thing would be to add a page to the site for unofficial patches.
19:29:33Bagderwe could perhaps use the patch tracker on sourceforge
19:29:35ZagorHes: do you know about the player chargers?
19:29:39elinenbeI uses phpbb on one of my sites, and they have a page like that (for phpbb) it is great.
19:29:45ZagorBagder: indeed
19:30:08Zagorstill, with this much movement patches would go out of date FAST
19:30:24Bagderthey either go in or grow old
19:31:16elinenbeoff to fix my car! the passenger windo came off the track in my 99 VW golf!@
19:31:25elinenbe#@*($@@!#@@* DAMN IT!
19:31:41elinenbewhat a pain in the ass... I'm gonna have a fun afternoon (sarcasm!)
19:55:23BagderZagor: make a screen dump of the wps for the manual! ;-)
19:55:24 Quit MeRWiN (Read error: 104 (Connection reset by peer))
19:55:41*Bagder curses javascript
19:56:53Zagori think we should kill the status bar menu config and just store changes done with f3
19:57:09Bagder*exactly* what I proposed this morning
19:57:10Zagora menu item is overkill
19:57:14Zagoroh, ok
19:57:25Bagdermbr said he was gonna make it so after 1.2 is released
19:59:58Zagortetris segfaults the simulator
20:00:03dwihnoYou are _SO_ evil :)
20:00:16dwihnoAm I the only one wanting everything configurable in the configuration menu?
20:00:33Bagderthere's no point in having the status bar on/off in the config menu
20:00:41Bagderit isn't a config issue
20:01:24dwihnoSure it is :)
20:01:37BagderI don't think so
20:01:42Bagderits a key
20:01:46Bagdernot a config
20:01:58Bagderf1+down toggles key lock
20:02:04Bagderf3 toggles status bar
20:02:21 Join MeRWiN [0] (~merwin@
20:02:21Zagori agree
20:02:26MeRWiNZagor: I added the scroll_pause/resume once (about a week ago), but Linus never committed it
20:02:45MeRWiNZagor: I think it was supposed to be in but never made it
20:03:02ZagorMeRWiN: ok. well it's not a big enough issue to put it in 1.2
20:03:21MeRWiNI've got the patch if you still want it :)
20:03:36Zagorsend it to the list if you haven't already
20:03:50Zagorit's good to have everything on the list, since it's nicely archived
20:03:51BagderMeRWiN: you got cvs access now, you can commit it yourself after 1.2 is out
20:04:09MeRWiNBagder: true true.. when are we pushing 1.2?
20:04:12*Bagder avoids work ;-)
20:04:25ZagorMeRWiN: i plan for release tomorrow
20:04:28*MeRWiN looks stoned in his passport photo
20:04:41MeRWiNZagor: perfect. :-)
20:04:59Bagderwe'll see a flood of commits the minutes after the release :)
20:05:16MeRWiNZagor: 1.1 is a letdown compared to 1.2 ... I think people will be nicely suprised
20:06:14MeRWiNI remember the good old days (about 2 months ago) when MP3's barely played back :-)
20:06:31Zagorhehe. those were the times... :)
20:06:55MeRWiNThen I didn't come back to the site for a month and suddenly it was release software
20:07:00*Zagor is reliving The Times using Sidplay and a huge SID collection...
20:07:16MeRWiNwhat would sidplay be?
20:07:25 Quit grummel ()
20:07:31Zagorc64 sound chip emulator. plays old game and demo tunes.
20:07:59MeRWiNI miss demos... they're just not like they used to be anymoe
20:08:50Bagderwe were pretty active demo creators on the C64
20:09:04Bagderback in the late 80s
20:09:37MeRWiNI was an active demo watcher in the early 90's :-) All of 12 years old
20:09:58Bagderkiddo ;-)
20:10:58MeRWiNI'm too smart for my own good i guess. *grin*. I've gotta let my age catch up to my knowledge.
20:13:33MeRWiNSo what happens with Feature commits (assuming there's no freeze)? Do we all agree on it then commit it? (That's what I would assume)
20:13:58Bagderbasicly yes
20:14:26Bagdersmall things we just commit and discuss afterwards, of course risking having to change it
20:15:42Bagderalso, after a commit it is considered polite to check the build status and fix any non-green status
20:16:15MeRWiNK.. I'll have the patch for scroll_pause/resume in ready for commit on a timer to execute tomorrow at 8pm Pacific Standard Time. :)
20:24:54Zagorahhh! buggggg!
20:25:15Zagordirs with playlists start the wrong mp3 when you press PLAY on it
20:28:15Bagderbtw, check out this =>
20:28:23Bagderlogo contest ;-)
20:28:47Bagderthe funnest thing is how you vote
20:28:55Bagdercheck out the directory, edit the file, commit
20:29:00Bagderusing svn
20:30:21Zagorsome cool submissions
20:30:27Bagder#2 is in the lead
20:30:43Bagderwith 16 and 7 following
20:33:07Hadakahaha, svn trickling onto this channel as well :)
20:33:23Bagderof course
20:33:29Hadakasayy, would you want to convert the rockbox repos to svn? ;)
20:34:30Bagdernot yet, I'd say
20:35:17Hadakammh, yeah, you might want to wait beyond the phase when boostrapping is a necessity atlesat
20:36:04Bagderthere are times when being an early adoptor is cool and there are times when it is less cool ;-)
20:37:00Hadakawell it's a bit bad to require everybody to do svn bootstrapping just to download rockbox sources from it's version control system
20:37:28Bagderand then update svn every now and then when more or less annoying bugs are fixed ;-)
20:37:58Hadakayeah, tracking HEAD is kinda mandatory still
20:38:26Hadakaso maybe we'll have this conversation again in two months or so ,)
20:38:52Bagdersvn is cool, I would really like to run a project on it
20:39:14Hadakadon't - once you touch it, you'll never wanna go back :)
20:40:20BagderI have touched it you know ;-)
20:40:23MeRWiNZagor: Do you have random on by chance?
20:40:36Bagdergrep daniel COMMITTERS
20:40:58Hadakagrep naked COMMITTERS
20:41:25ZagorMeRWiN: no, off
20:42:03MeRWiNZagor: hmm... weird. It plays the wrong mp3 when you press PLAY on a playlist, or just any dir with a playlist in it?
20:42:21***Saving seen data "./dancer.seen"
20:42:22Zagornot on a playlist, when I press play on an mp3
20:42:34Zagorit plays the next
20:43:01MeRWiNZagor: whoops... might wanna fix that before the release :)
20:43:07Zagorheh, yeah
20:43:22MeRWiNdidn't you say there was a mpeg_next/prev problem or something?
20:43:34Zagorit's repeatable in the simulator so it should be a small deal tracking it down
20:43:40Zagorno, it's in the playlist build I think
20:43:59Bagderits probably in the code that skips non-mp3 files or something
20:44:23Zagoryes. it's off by one, although I don't see why
20:46:27Bagderis there one playlist above in this case?
20:47:00Bagderbut there's one?
20:47:04BagderI mean not two
20:47:24Zagorno playlist necessary, just a dir that contains both folders and files
20:47:36 Join ripnetuk [0] (
20:47:36Bagderand it is always off by one?
20:47:50Zagorno, now I get more. i thought it was one
20:48:07Zagorin one dir, I'm offset by 3
20:48:56Bagderyou need to step into build_playlist() and I'm sure it'll be visible
20:53:22Bagdersee ya
20:53:24 Part Bagder
20:53:37 Quit ripnetuk ()
20:57:36Zagor*dang* found it
20:59:48MeRWiNwhat was it?
20:59:55dwihnothe evil bug!
21:00:01Zagorsilly indexing in build_playlist
21:00:17MeRWiNdamn silly indexing
21:01:44dwihnosilly silly indexing
21:01:48 Quit Synthe ("Connection Lost - Excess Blood")
21:07:25Zagorah, tetris works again
21:14:02 Join Synthe [0] (
21:39:39 Quit adi|work ("[x]chat")
21:40:50MeRWiNZagor: I didn't know it wasn't working
21:41:02Zagorit crashed for me in the sim
21:58:52 Quit MeRWiN ()
22:26:28 Join Linus [0] (
22:27:04LinusYour latest wrap fix in playlist_next() isn't very good, is it?
22:27:28Zagorwrap fix?
22:27:34Linusisn't it possible that the index is less than -1?
22:27:38Zagorit's a negative fix. and it solves the problem.
22:27:40 Join d728r [0] (
22:27:48Zagorwhen would it?
22:28:30Linusplaylist_next(-number_of_files()); or whatever the function is called
22:28:43Zagorthat is one of the bugs reported today "play the first file, go left. go right. it still plays the first file"
22:29:02Zagorthat should still never make the *index* negative. only the 'steps' parameter.
22:29:24Linusif index=0 and steps=-3?
22:29:36Zagorthen something is wrong. that should never happen.
22:29:48Linusnot is repeat=true
22:30:15Linusplay the last song, wrap around and preload the first song again
22:30:17Zagoryou mean just because we repeat, we should repeat *backwards* too?
22:30:33Zagorwell, that's an option of course
22:30:36Linusyou still don't get it?
22:31:08Linusimagine starting to play the last Farbror Frej song
22:31:11Zagorit is my *indention* to not be able to go to any previous track from the first. the first is as much previous as you get
22:31:21Linusit wraps and preloads the next 5 songs
22:31:38Linuspress <left>
22:31:56Zagorok, now i get it.
22:32:00Linusthank you
22:32:06Zagoryeah, we need wrap if repeat is active
22:32:08Zagori'll fix it
22:32:33Linusthat is probably why the modulo operator is used in the first place
22:32:52Zagorwell modulo doesn't help much for negatives
22:33:06Linusi guess the programmer didn't know that
22:33:53Linuswhat happens if the variable is unsigned?
22:34:45Zagorumm, from -1 it will become 4 billion something. then we modulo it down, and probably won't get the last track :-)
22:35:56Zagoralso I don't want it to wrap unless repeat is on
22:37:45Linuswell, i just wanted to whine about the playlist index
22:37:54Linusi'm going to bed
22:38:35Zagorwhat is the difference in charging circuit btwn new and old player?
22:38:44freshmakerHej, there seems to be a new model available: AJB Recorder 15 w/ black(sic!) bumpers:
22:39:11LinusZagor: i don't know
22:39:41ZagorLinus: ok. but they both can just read the voltage and then don't do much more?
22:39:54Linusi guess so
22:39:57Zagorfreshmaker: looks like a Recorder 10 with a bigger disk. USB 1.1
22:40:09Zagorblack looks nice, though
22:40:20Linusi think the newplayer(tm) can shut off the disk as well
22:40:29freshmakeryup, and w/o a real umass ata bridge
22:40:31Linusi haven't figured out how...
22:40:40ZagorLinus: shut off the disk?
22:40:48Zagorah, you mean the ata power
22:40:49Linuspower off, like the recorder
22:41:43Linusgotta hit the sack
22:41:57 Part Linus
22:42:25***Saving seen data "./dancer.seen"
23:00:09 Quit d728r ("Leaving")
23:10:04 Join Bagder [0] (
23:10:29Zagorhey, bagder
23:10:35Bagderhey ho
23:11:06ZagorBagder: what's the the "magic" in ppp log line "[LCP EchoReq id=0xa magic=0xddf631f c3 5c c3 5f" ?
23:11:15Zagorjust a key, or some info?
23:11:35Bagderjust a key
23:11:53Zagorwhat's it for?
23:12:00Bagdermade to detect loopback
23:12:06dwihnohey, Zagor, both can ppp bigger to be playlist modulo voltage
23:12:11Bagderif I recall it correctly
23:15:13Zagordwihno: random tech talk? :)
23:15:41dwihnotechtalker(tm) :)
23:15:43Bagdertried my button? ;-)
23:15:54dwihnoSounds like a song
23:17:00dwihno"mobile telephone silly me take a walk"
23:17:17dwihnothat wraps it up quite nicely
23:17:56 Quit freshmaker ("ChatZilla 0.8.7 [Mozilla rv:1.0.1/20020805]")
23:18:00dwihnoHuman interaction is better than using the mobile phone for communication! :)
23:18:38dwihnotake a walk is in every goddamn sentence :)
23:19:29Bagderveeeery advanced AI engine there
23:20:07dwihnoI bet it is :)
23:20:13dwihnoYou should include it in rockbox
23:20:23dwihnowhen showing the boot screen
23:20:39dwihno"rockbox take a walk assembler MPL mobile phone"
23:20:43Bagderlike a fortune cookie
23:20:45dwihno+ "pardon me
23:20:58 Join ripnetuk [0] (
23:21:30Bagderhm, I should add Rockbox to the words!
23:22:42dwihnoBagder: why didn't you do my laundry? Now I have to get up @ 6.30 tomorrow to do it! :(
23:22:56Bagderah I forgot
23:23:19Bagdersurely you can re-use some of the least dirty ones a few more days? B-P
23:23:42dwihnoI'm not that much of a computer enthusiast ;)
23:23:56dwihnoI don't even wear glasses!
23:26:22 Nick dwihno is now known as dwihno|zzz (dwihno@Bald067.Baldakinen.Umea.SE)
23:26:30dwihno|zzzI'm off to catch some sleeps0r
23:26:43dwihno|zzz"mobile telephone silly me take a walk"
23:26:57Bagdernight dwihno|zzz
23:27:21dwihno|zzzyou should sleep too - another day of work ahead
23:27:44Bagdermore bugs to fix, more patches to apply ;-)
23:29:26 Join mecraw [0] (
23:34:49ZagorBagder: if i can't get a screendump function, I at least want the black border to be on all four sides of the screen :-)
23:35:17Bagderit was my intention, I'm not really sure why it isn't
23:35:48Zagorlcd_clear_screen doing one pixel too far?
23:36:00Bagdercould be
23:36:41*Bagder digs around
23:38:46Zagorhmmm, didn't I receive player simulator icons before my vacation...
23:38:50*Zagor digs around too
23:40:48BagderI think it is drawn below the window
23:40:52 Quit ripnetuk ()
23:41:21Bagderthat's it
23:42:38Bagderperhaps we should remove the border all together instead?
23:43:04Bagderor if we keep it, adjust it when we build a player sim
23:55:19Zagorthat's a good idea
23:55:38Bagderyou mean keep it and resize
23:55:39Zagoris it hard making the whole window smaller for player-sim?
23:55:48*Zagor is an x11 clewbie
23:55:55Bagderwe #ifdef anyway
23:56:06Zagorthat'd be best, i think
23:56:18Bagder "*geometry:120x70", /* this should be .geometry, but nooooo... */
23:57:28Bagderok, so we resize and remove the border then
23:58:38Zagori kinda like the border... :)
23:58:43Bagderso what about the player "icons", shall I spare 8 pixels above and below to allow us to add them?
23:58:51Bagderok, border stays then

Previous day | Next day