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 2006-01-09

00:01:27 Join HaKuNa [0] (
00:01:35amiconnNot elf? ;)
00:01:53preglowi'm not completely insane
00:02:02preglowif this is ever to work, it needs to be relatively low complex
00:02:21preglowone problem i can see already is that we sometimes use longer section names than coff can handle :P
00:02:36preglowhmm, no, that's just in the intermediate files
00:02:58preglowthe resulting files will only have .text, .data, .bss and the .ixxxx variants, yes?
00:03:12amiconnPlu s.rodata
00:03:20amiconn*Plus .rodata
00:03:32preglowwill do
00:03:37preglowi think the limit is eight characters
00:03:40preglowwhich is lame, but yeah, low complex
00:04:17preglowbut you agree on the macro stuff?
00:04:53amiconnA dynamic loader would be somewhat like dreaded malloc() ...
00:05:17preglowcan't see how it's even comparable
00:06:12amiconnYou could load several plugins/codecs, but how many depends on how large they are, and in what order you load them, provided you unload some of them inbetween
00:06:47preglowahh, yes, but only in the same way we reserve memory for plugins anyway these days
00:06:54preglowwe would have a single buffer where all would have to go
00:06:56amiconnOf course we could define a number of fixed-size buffers
00:07:00preglowconfigurable, of course
00:07:02preglowin size
00:08:02preglowbut yeah, i understand the issue
00:08:03 Quit HaKuNa ("CGI:IRC (EOF)")
00:08:12preglowi just firmly believe this is a problem we're going to have to deal with sooner or later anyway
00:08:15*markun found out that the Gigabeat has SRAM after all. 4kB of it :)
00:08:23preglowi'll probably try making a dynamic loader anyway, just to educate myself
00:08:26preglowmarkun: no iram?
00:08:31preglowmarkun: or is that it?
00:08:51markunIt has 16kB data and 16kB code cache
00:08:57preglowyeah, cache
00:08:59markunno IRAM
00:09:05peturgdnight all
00:09:08preglowpetur: night
00:09:08markunbut you can use the bootup SRAM
00:09:10 Quit petur ("here today, gone tomorrow")
00:09:11preglowthat's a decent amount anyway
00:09:43markunWould be make sense to use the bootup SRAM?
00:10:20preglowwhy no
00:11:06 Quit TCK (Read error: 110 (Connection timed out))
00:11:54markunIt's connected to the NAND flash controller, maybe that makes it slower to access
00:13:08preglowahh, right
00:13:12preglowit probably is
00:13:18preglowshould be quick to test
00:14:32Bagderbootup sram?
00:14:40Bagderisn't that what we call iram usually?
00:14:49markundon't know
00:14:52Bagdera little pieace of ram on the chip itself
00:16:15preglowoh <-
00:16:19preglowi thought it was external
00:17:00Bagderis it?
00:17:09BagderI know my ARM at work has it internal
00:17:15 Quit Kohlrabi ("Leaving")
00:18:00 Join HaKuNa [0] (
00:20:39preglowmakes most sense
00:21:17 Quit Thus0 ("Leaving")
00:21:31lodesijust sent a japanese learning plugin on the patch tracker
00:21:44lodesiif someone want to try
00:24:03preglowsounds cool
00:24:14preglowwish i had this five years ago when i actually tried learning :>
00:24:31 Quit Mulzi-SAW ()
00:24:52lodesithat's why i did it, if i can learn kanjis this way, it would be really cool :)
00:33:57 Join einhirn [0] (
00:34:30 Join lokki [0] (
00:35:09 Join Jungti1234 [0] (n=jungti12@
00:36:00 Join ashridah [0] (
00:38:13preglowi'm out, later
00:38:29lokkiamiconn: how do i adjust the size of the peakmeter? i thought it depends on LCD_WIDTH, changing that to LCD_REMOTE_WIDTH seems not to work
00:41:38 Quit HaKuNa ("CGI:IRC")
00:46:59 Quit ender` (" But there, everything has its drawbacks, as the man said when his mother-in-law died, and they came down upon him for the fu)
00:51:10 Quit lokki ("CGI:IRC (EOF)")
00:51:28 Join hshah_ [0] (
01:00:40 Quit lodesi ("Leaving")
01:09:31 Quit Jungti1234 ()
01:10:44 Quit hshah (Read error: 110 (Connection timed out))
01:16:25 Quit mikearthur (Remote closed the connection)
01:16:27 Join mikearthur [0] (
01:17:15 Quit Moos ("Happy new year @ all")
01:30:15 Join akaidiot [0] (
01:30:54 Quit t0mas (" good night")
01:34:52 Quit ashridah ("Leaving")
01:52:50***Saving seen data "./dancer.seen"
01:53:16 Join TCK [0] (
02:09:16 Join DJDD_ [0] (
02:14:12 Quit akaidiot (Read error: 110 (Connection timed out))
02:18:20 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
02:32:45 Quit mikearthur (Remote closed the connection)
02:36:24 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
02:41:00 Quit ts-x ("CGI:IRC (EOF)")
02:47:32 Quit hshah_ (Read error: 104 (Connection reset by peer))
02:51:15 Part Paul_The_Nerd
03:09:08 Quit DreamTactix291 (Read error: 110 (Connection timed out))
03:10:49 Part Midgey34
03:11:22 Join hotwire [0] (
03:26:22 Join pmmagic [0] (
03:30:59 Quit TCK (Read error: 104 (Connection reset by peer))
03:45:56 Join gtkspert [0] (
03:52:53***Saving seen data "./dancer.seen"
03:54:24 Join akaidiot [0] (
04:05:47 Join ashridah [0] (
04:06:05 Quit hotwire (Read error: 104 (Connection reset by peer))
04:20:59 Quit akaidiot (Read error: 110 (Connection timed out))
04:23:07 Join Membrillo [0] (n=sam_kill@
04:25:34Membrilloany able to add an Ext. Mic option to recording source on H300? Just make it the same as line in but automatically add ADC gain
04:26:00 Join hd [0] (
04:26:04 Quit goa (Read error: 104 (Connection reset by peer))
04:26:31 Nick hd is now known as goa (
04:35:43 Quit Paprica (Read error: 113 (No route to host))
04:40:22 Quit RotAtoR ()
04:40:36 Quit gtkspert (Read error: 110 (Connection timed out))
04:42:21 Quit Membrillo ()
04:44:37 Join Paprica [0] (
04:48:40 Join DreamTactix291 [0] (
04:59:07 Part mc365
05:03:44 Join Jungti1234 [0] (n=jungti12@
05:03:58 Join Rob2222_ [0] (
05:04:52Jungti1234Toshiba PMP
05:06:01Jungti1234 Gigabeat
05:06:20Jungti1234 ZEN PMP
05:20:33 Quit Rob2222 (Read error: 110 (Connection timed out))
05:52:58***Saving seen data "./dancer.seen"
05:54:28 Quit Jungti1234 ()
06:01:11 Join akaidiot [0] (
06:14:41 Quit actionshrimp (Read error: 110 (Connection timed out))
06:19:51 Quit pmmagic ("CGI:IRC (EOF)")
06:23:10 Quit akaidiot (Read error: 110 (Connection timed out))
06:27:27 Quit joshn_454 ("KVIrc 3.2.0 'Realia'")
06:32:41 Quit thegeek (Read error: 104 (Connection reset by peer))
06:35:18 Join thegeek [0] (
06:38:17 Join webguest90 [0] (
06:38:33webguest90hello all
06:40:04 Join mithereal [0] (
06:40:33mitherealis anyone here?
06:40:34 Quit webguest90 (Client Quit)
06:42:41mitherealhas anyone successfully flashed a gigabeat?
06:43:04lostlogicrockbox gigabeat doesn't work yet.
06:43:31mitherealoh ok i was readin a page that said they had an unofficial
06:44:58lostlogicmarkun has been working on it, but I don't think progress is anything beyond just booting and that appears flakey even.
06:46:53 Quit bagawk (
06:46:53 Quit Nilisco (
06:46:53 Quit CoCoLUS (
06:46:53 Quit crash (
06:48:06NJoinCoCoLUS [0] (
06:48:08NJoinNilisco [0] (
06:49:50mitherealwell ne1 know how i can dump the old firmware well my friend that is he wants to try neways
06:51:56 Join crash_ [0] (
06:52:27NJoinbagawk [0] (
06:58:20 Join webguest88 [0] (
07:06:21 Quit uwe_ (Read error: 104 (Connection reset by peer))
07:09:42 Join fobbo [0] (
07:14:49 Part mithereal
07:23:28 Join uwe_ [0] (
07:32:46 Quit fobbo ()
07:40:33Bgerlostlogic, are you here ?
07:44:46 Join B4gder [0] (
07:50:03Bgermorning, B4gder , back to work ? ;)
07:50:18B4gderthe harsh cold reality
07:51:14B4gderno more working in my underware all day :-)
07:52:59***Saving seen data "./dancer.seen"
07:53:23Bgerhehe :)
08:04:07 Join akaidiot [0] (
08:07:23 Join Kohlrabi [0] (
08:26:04 Quit akaidiot (Read error: 110 (Connection timed out))
08:29:03 Join ender` [0] (i=ychat@
08:29:26 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
08:35:46*B4gder enjoys TFTP boot and root fs mounted on nfs
08:38:41webguest88That is a neat trick
08:39:24B4gderits lovely
08:39:45Bgerbootp or ?
08:39:55B4gdernah, plain TFTP
08:40:12B4gdermy board runs u-boot that sucks over the kernel image via tftp
08:43:45preglowahh, i thought this was your workstation setup
08:44:09Bgerah, it isnt ?
08:44:22Bgersome embed. system ?
08:44:23B4gdernah, its my little arm9 cutie
08:44:23preglowmounting root via nfs voluntarily would be insane
08:46:27B4gderyes. but for a tiny flash-based thing it is a dream
08:47:10BgerB4gder what's its purpose if it's not a secret ? (sorry if being too nosy)
08:47:42B4gderit is just the thing I've been working with last year, its a little RFID reader
08:47:59B4gderbased on the atmel at91rm9200, arm920 core
08:48:09B4gderrunning linux
08:48:19B4gderI write just about all sw for it
08:48:31Bgerah, i understand :) sounds very good
08:48:42B4gderit is a neat and fun little toy
08:49:28BgerRFID is the follower of barcodes or i'm messing it up with something other
08:49:54B4gderyes, basically. You can "read" the RFID chip wirelessly
08:50:12Bgeryep, that's what i mean
08:52:07 Join _FireFly_ [0] (
08:54:40 Join LinusN [0] (
08:56:18preglowbah, ofcourse, of course the light had to go bust
08:56:23preglownow my office is dark as well as cold
08:57:51B4gderso you IRC without electricity? ;-(
08:57:55B4gder:-) even
08:58:24BgerUPS ?
08:58:30*B4gder imagines preglow pedaling like a mad man to power his computers...
08:58:31Bgeror mobile phone
09:00:12 Join Paul_The_Nerd [0] (
09:09:58 Join Zagor [0] (
09:10:12B4gdermorrn bjrn
09:14:57 Quit saa[b_r]ider (Read error: 104 (Connection reset by peer))
09:18:41 Join einhirn [0] (
09:22:03 Join saab_rider [0] (n=saab_rid@
09:28:17preglowi am a mad man
09:29:43Paul_The_NerdIn the good way, or the "angry" way?
09:30:31preglowhaha, good way of course
09:31:28Paul_The_NerdGood, good.
09:31:33Paul_The_NerdSo, why are you a mad man today?
09:34:20preglowwhat, i'm always mad
09:34:38preglownot any madder today than other days
09:34:41preglowtireder i am, though
09:37:37preglowrockbox just went mad
09:38:02preglowstarting cycling through files for some reason, couldn't make it stop, then it promptly stopped itself with "codec error"
09:40:48 Quit bagawk (Read error: 110 (Connection timed out))
09:41:09 Join bagawk [0] (
09:43:14Paul_The_NerdThat's a strange happening
09:43:26Paul_The_NerdMine's been skipping on the first song when I start playback
09:43:39Paul_The_NerdUsually there's be a half-second or so gap 3 times in maybe the first 10 seconds of playback.
09:44:58preglowi get that whenever it needs to buffer a new track
09:45:08preglowi think slasheri's tried to fix this about a hundred times now ;)
09:45:42LinusNi never get that
09:46:18Paul_The_NerdIt just started on mine shortly before the holidays.
09:46:26webguest88Here's a fun one: Start playing a song, then add your top-level directory recursively to the playlist
09:46:30webguest88Skipping guaranteed
09:46:39Paul_The_NerdIt hadn't done it before, I don't believe.
09:46:51Paul_The_NerdI'd kinda assumed my player was just creatively aging.
09:48:12preglowevery time i insert a new dir, i get skip on the first track
09:51:13preglowhooray, lights are suddenly working
09:51:43preglowthis just strengthens the theory of me being mad
09:51:44Paul_The_NerdHrm, I usually just play playlists I've set up, but any time I start playback, I get a few skips.
09:52:27 Join TCK [0] (
09:53:00***Saving seen data "./dancer.seen"
09:54:39SlasheriPaul_The_Nerd: interesting. What bitrate the first song is?
09:55:35*markun just received a big envelope from Japan
09:56:42Bgerwoot :)
09:57:19markunThe CD reads: "GPL Source Code for gigabeat MEG-F40"
09:57:37preglowimage, image, quick
09:57:59markuniso or picture?
09:58:35preglowlets hope nothing is missing
09:58:47preglowwouldn't surprise me a second if something turned out to be
09:59:01markunThere is not much on it
09:59:24markunThe only thing I couldn't have downloaded it the USB driver: legna-usb-ohci.tar.gz
09:59:24psiis anyone developing a level indicator for recording mode?
09:59:48preglowtwo hundred megabyte ram dump and a README with 'kekekeke' in it? ;)
09:59:49markunAnd maybe the wlan driver: linux-wlan-ng-0.2.0-gigabeat2-2.0.tar.gz
09:59:49LinusNpsi: what's wrong with the current one?
10:00:23psiusing ext mic through line-in on the iriver, there's no way to tell if you're clipping
10:00:30Paul_The_NerdSlasheri: Sorry 'bout the slow response. It's independent of bitrate or format, as it'll happen on both flacs and MP3s. It's any song on my player, at least.
10:00:48SlasheriPaul_The_Nerd: oh, then that is really weird..
10:01:03Slasherimaybe the disk is worn out or something like that
10:01:22preglowmine too? :/
10:01:24Slasherimarkun: what, gigabeat has a wlan? :o
10:01:26Paul_The_NerdThat's kinda my feeling. The HD is finally beginning to show signs of age and being bumped off a table here or there.
10:01:44markunSlasheri: Maybe you can add a wlan USB NIC or something
10:01:47Paul_The_NerdI've had it since a few weeks after the H120s started showing up in the US.
10:01:52Slasherimarkun: ah, that sounds cool :)
10:02:05Slasherimarkun: i have always wanted a wifi internet radio to my dap ;)
10:02:24LinusNpsi: i wasn't aware that the peak meter didn't show the truth
10:02:49psiLinusN: are we talking about the h120?
10:03:04LinusNor the h300
10:03:46SlasheriPaul_The_Nerd: ah, then that is probably the problem. It could be fixed by increasing the pre-buffering of the first song before starting playback
10:04:26Paul_The_NerdSlasheri: Is that a code change I'd have to make, or is that handled by the anti-skip buffer menu entry?
10:04:30psiLinusN: i was planning on purchasing a refurbished h120 off ebay, and all the reviews i've read from field recorders using the h120 as the recording device say that there is no level indicator with ext mic in... maybe it's different for the internal mic?
10:04:49SlasheriPaul_The_Nerd: nope, that needs a code change
10:04:57LinusNpsi: you're talking about the iriver firmware
10:05:09LinusNpsi: i'm talking about rockbox
10:05:14Paul_The_NerdSlasheri: All right. Perhaps I'll try that then. Though I was planning on just getting one of the 40gb single-platters at some point anyway.
10:05:17psiLinusN: ooh! so rockbox already has a peak meter implemeneted!?
10:05:52SlasheriPaul_The_Nerd: hehe, sounds good :)
10:06:00psiguess i skipped over that one on the features list ;]
10:06:18LinusNpsi: we don't list obvious things as features
10:06:28Slasheriprobably you should do a little modification to playback.c to allow it pre-buffer more at once
10:06:38 Join akaidiot [0] (
10:07:41Paul_The_NerdYeah, it shouldn't be too big of a problem
10:10:55 Join amiconn_ [0] (
10:11:07LinusNyou would believe that the anti-skip setting would help in this case
10:11:43preglowi think the antiskip settings works differently for iriver than for archosd
10:11:48LinusNas a user, i mean
10:12:31 Join gtkspert [0] (
10:12:40preglowyeah, sure, i was just trying to make someone clear up what the differences are :)
10:12:59LinusNi have no idea how the iriver playback works
10:13:13LinusNonly that it seems awfully complicated
10:13:24LinusNcompared to the archos
10:14:10preglowi think the archos one looks pretty complicated as well...
10:14:24preglowLinusN: btw, what's your position on having plugins be coff format? :)
10:14:46LinusNthe archos playback code is far from trivial as well
10:15:14LinusNpreglow: you mean dynamic relocation?
10:15:40LinusNgive me a reason
10:15:41preglowwe'll also get proper section handling
10:15:53preglowloading of multiple plugins of some type
10:15:57preglowfor example dsp plugins
10:16:04preglowwithout doing some ugly linking kludge
10:16:28LinusNah, of course
10:16:51preglowno immediate need, i just think we'll need it sooner or later, and i might want to work on it at some point
10:17:13preglowarchoses will probably just continue to use flat bin files
10:19:16preglowwe already have to work around the static linkage, i believe, in voice ui
10:19:27preglowhaving to swap the entire codecs back and forth
10:19:39 Quit _FireFly_ ("Leaving")
10:20:38LinusNyes, but we would still have to swap the iram
10:21:57preglowbut that has less of an impact, of course
10:22:03LinusNof course
10:22:06preglowbecause it is, after all, iram, and smaller
10:22:12preglowking of the commas
10:22:26LinusNi'm all for an elf loader
10:22:31 Quit saab_rider (Read error: 110 (Connection timed out))
10:22:41preglowme too, if i can find out exactly how much more complex than coff it is
10:23:04preglowit is more complex, but i don't know how much of it we'll be forced to use
10:24:05LinusNi was under the impression that elf would be simpler if it wasn't for the coff compatibility
10:24:15pregloweh? elf is coff compatible?
10:24:39LinusNor i might just be talking out of my a**
10:25:25preglowi'm by no means an expert, but i don't think elf is coff compatible in any way
10:25:39preglowbut that doesn't mean it isn't complex anyway, which i think it is
10:26:26 Join solexx [0] (
10:26:51 Quit solexx_ (Read error: 110 (Connection timed out))
10:27:47 Join Xerion [0] (
10:28:25 Quit webguest88 ("CGI:IRC (Ping timeout)")
10:28:51 Quit amiconn (Read error: 110 (Connection timed out))
10:28:51 Nick amiconn_ is now known as amiconn (
10:30:05Paul_The_NerdI like the bit where it says it may "modify random memory locations at runtime."
10:30:10 Join _FireFly_ [0] (
10:30:11 Quit akaidiot (Read error: 110 (Connection timed out))
10:31:12preglowperfect, it's even got m68k code
10:31:33LinusNc++ ... :-(
10:34:14*preglow uses c++ right now
10:34:48 Join mikearthur [0] (
10:34:49 Quit Paul_The_Nerd ("Leaving.")
10:35:36preglowoh, it _is_ c++.......
10:35:54preglowat least i can pillage their code
10:36:30LinusNfrom what i can see, it is a two-step process
10:36:51LinusNhe preprocesses the elf file
10:37:00LinusNand his loader can only load preprocessed files
10:37:04LinusNfair enough
10:38:14preglowi'd say
10:38:23preglowif it makes the loader complexity less
10:38:46preglowit's quite important the actual relocation step doesn't take too long
10:38:58preglowhmm, or well
10:39:34preglowit's not too important, if we're unlucky and the pcm buffer is low between track change, gapless playback between different codecs isn't that important
10:39:37preglowif at all
10:40:13 Join eli_sherer [0] (
10:41:36*amiconn suddenly noticed that his .bss changes introduced a nasty bug for plugins :-(
10:43:43preglowwhat's that?
10:44:28amiconnplugin_get_buffer() uses the binary size to calculate the start of free RAM.
10:44:43amiconnThis now clashes with the .bss space...
10:45:10amiconnWe'd need a small header to solve this problem (similar to the archos .ovl header)
10:45:49preglowor: *drumroll* a proper plugin format!
10:46:18preglowbut yeah, a header will do for now :)
10:51:19preglowlooks like the elf spec will be tonights bedtime read
10:53:12 Join linuxstb [0] (
10:54:00linuxstbMorning all.
10:54:06preglowgood morning
10:55:31eli_shereri'm working on a color patch for the wps (it's quite easy)
10:55:46eli_shereri'm using 'y' instead of 'x' for color images
10:56:01linuxstbpreglow: Did you get a chance to test the USB code on the Nano? I've discovered that the "reboot to diskmode" code doesn't work on the video - but the USB detection works.
10:56:37B4gdereli_sherer: I would say you should make 'x' support colors
10:56:38preglowno, i'm sorry
10:56:39preglowi can try now
10:56:55preglowif i can find my nano, that is
10:56:58linuxstbNo rush - I was just curious if it worked.
10:57:09preglowno, i'll just do it before i forget it again
10:57:57markunI disassembled a lot more of the gigabeat bootcode:
10:58:07linuxstbI didn't realise some of the Archos models had USB charging.
10:59:58eli_shererB4gder: that would cause me to check the inside of the bitmap
11:00:59B4gderI just find it unlikely that color-LCD people will fiddle much with non-color images
11:01:31B4gderso why not just deal with all images as color
11:05:05eli_shererbecause the buffer for the monochrome bitmaps is too small to handle big 16bit bmps
11:05:18eli_sherer1bit <-> 16bit
11:05:24B4gderwhy would you use a too small buffer?
11:05:33 Quit DJDD_ ("Trillian (")
11:05:35eli_shererto save space ofcourse...
11:05:56B4gderyou just said you can't use that small buffer
11:06:02B4gderthen you don't save any space when you add another buffer
11:06:26B4gderso again, why don't you extend the buffer when you load color images?
11:06:57B4gderof cours non-color targets would continue using the smaller size
11:07:16 Quit linuxstb ("CGI:IRC (Ping timeout)")
11:12:25 Join linuxstb [0] (
11:13:31linuxstbDoes anyone have any thoughts about my patch adding bmp2rb to the build system? IIRC, the main problem was dealing with multiple bitmap formats (main unit and remote, mono and native formats) for the same device. I think we will need a more complicated "SOURCES" format to handle all the permutations efficiently.
11:14:28B4gderI like the idea of having the bitmaps stored in a struct
11:14:40B4gderso that you can read the depth and size from it
11:15:12 Join saa[b_r]ider [0] (n=saab_rid@
11:15:25preglowLinusN: what's supposed to happen on usb plug?
11:15:34linuxstbBut that means that sizes are variables, not #defines - probably not a big problem, but I think it will increase code (and data) size a little.
11:15:46B4gderhm, right
11:15:56linuxstbpreglow: It should reboot into disk mode. Unless you hold menu, in which case it will just charge.
11:16:04preglowi just get the usb logo
11:16:13B4gderlinuxstb: exctly what problem would the "advanced" SOURCES solve? (I'm forgetting)
11:16:26preglowit comes with age
11:16:43preglowlinuxstb: holding menu and inserting, i get a little icon besides the blinking battery
11:17:10amiconnThat's the 'USB power' logo
11:17:18linuxstbB4gder: My memory's fading, but I think it was a) Dealing with different (mono and native) bitmap formats; and b) Using the same bitmaps with different names - e.g. "rockboxlogo" for Archos, "remote_rockboxlogo" for iriver.
11:18:33linuxstbWhen I insert the USB cable on my ipod, I get the USB logo appearing for a split second, then it reboots.
11:18:34B4gderlinuxstb: I could imagine that to start with, the build system builds all bitmap combos but the source files use #ifdef to figure out which one to actually use
11:18:58B4gderthen of course they would use different names
11:19:09linuxstbThat could be a _lot_ of bitmaps.
11:19:20B4gderyes, scratch that
11:19:21preglowlinuxstb: most certainly doesn't reboot here
11:19:33 Quit TCK ("I shall not cease!")
11:20:12B4gderstill, I believe they should use different names
11:20:26B4gderand have a header file or something that #ifdefs the symbols right for each target
11:21:14B4gderI need to get the SDL patch committed so that I can start working with colors!
11:22:29 Quit linuxstb ("CGI:IRC")
11:22:34 Join linuxstb [0] (
11:22:44linuxstbBagder: What do you mean by "different names" ?
11:24:27eli_shererb$dger: if i use the current buffer and extend it. i would need to store white as 0xFFFFFF and black as 0x00000 for monochrome bitmaps and thats a waste for the other machines
11:25:25B4gdereli_sherer: non-color machines should not do that
11:25:25 Quit saa[b_r]ider ()
11:25:53B4gderlinuxstb: I'll try to give it a go during this week and see what I can come up with
11:26:23 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
11:32:20linuxstbpreglow: It's odd that it doesn't even reboot at all. Can you tell if it crashes?
11:33:56 Quit linuxstb ("CGI:IRC")
11:34:02*B4gder has this language-removed-from-configure patch pending...
11:34:13 Join linuxstb [0] (
11:35:20preglowlinuxstb: no, no crash, it even responds when i remove the cable as well
11:35:25preglowgoes back to file view and all is well
11:35:33B4gderI guess no one would object me committing this?
11:36:55preglowwhat's it do?
11:37:05B4gderremoves the language choice from the configure script
11:37:20B4gdermakes the build always do english
11:37:23preglowi never did see what that was good for
11:37:25preglowi'm all for that
11:37:36B4gderit isn't good, it only complicates things
11:37:59B4gderand is hardly ever used by anyone
11:38:28linuxstbSo is there any difference between choosing a language at compile-time, and selecting one at runtime?
11:38:53B4gderthe compile-time builds in the selected language as default
11:38:56B4gderwith no loading needed
11:39:16B4gderbut other than that, no
11:40:14linuxstbSo the only downside is very slightly longer boot time - to load the non-default lang file?
11:41:04B4gderand, I believe there is problems related to the built-in using a non-english language
11:41:12B4gderwhich we can avoid by this
11:41:18linuxstbWell, I always use English, so it doesn't affect me. But I can't imagine many people compiling their own builds just for that.
11:41:44preglowi say get rid of it
11:41:56B4gdercommit coming up
11:42:08B4gderI need to patch the build system too
11:42:11linuxstbTalking about the build system - is it a feature or a bug that the automatic builds didn't realise the ipod builds were failing when Tremor became too big for iram?
11:42:14B4gderI mean on the site
11:42:26B4gderlinuxstb: that's a bug
11:43:52markunWhat happens when someone removes all the language files and there is none built in?
11:44:04B4gderthere is always one built-in
11:44:42B4gderthe only difference now is that the built-in will always be english
11:44:52B4gder(unless you manually edit the root Makefile)
11:45:37linuxstbB4gder: Is the bug Tremor highlighted easy to fix?
11:45:39 Quit Rob2222_ (Read error: 104 (Connection reset by peer))
11:46:14B4gderlinuxstb: what build/URL shows this problem?
11:46:16 Join Rob2222 [0] (
11:46:33linuxstblet me check...
11:47:56linuxstbThe Tremor commit on 28 December (20:42) broke it, I committed a fix on 3 Jan. It affected all three ipod targets.
11:48:46B4gderI'll see what I can do...
11:50:34B4gderadded a pattern for /ld returned (\d+) exit status/
11:52:45 Quit uwe_ (Read error: 110 (Connection timed out))
11:53:01***Saving seen data "./dancer.seen"
11:53:59B4gdershows it being highlighted
11:56:13*ashridah mutters
11:56:27ashridahdamned phone. java app transfers fine, then turns around and says it failed, but won't tell me why
11:59:11 Quit mikearthur (Remote closed the connection)
11:59:13 Join ucan_keci [0] (n=ucan_kec@
11:59:21 Part ucan_keci
12:00:31B4gderwhat debian sdl package do you need for the SDL patch build, anyone?
12:01:18ashridahmight need the mixer one too
12:01:33preglowwhy'd you need that? we've only got one sound source
12:02:21*ashridah shrugs
12:04:22 Join mikearthur [0] (
12:09:07 Join akaidiot [0] (
12:09:07B4gderhey, it is looking good
12:09:36linuxstbWhat's that, the SDL patch?
12:09:53B4gderwhen running the h3x0 sim
12:10:21B4gderits painfully slow over my x11 tunnel over ssh to work, but it works
12:10:48linuxstbWhich patch are you using? I can see two on the patch tracker...
12:12:20B4gdercan't check the sound now though, as this runs on my computer at home and I'm at work ;-)
12:12:26preglowperhaps i could try it
12:12:44B4gderit applied and built cleanly
12:13:01preglowgah, there i am again, lets distract ourselves with rockbox!
12:13:40LinusNi think we should commit it and work from there
12:14:07LinusNif the general approach is right in the code
12:14:13linuxstbSo does the patch create a third simulator?
12:14:30preglowshould do for now, until the day we can remove the other ones
12:14:36linuxstbNo harm in committing it then.
12:14:45markunIt nathan's better?
12:15:28markunpoor guys duplicating each other's work..
12:15:55linuxstbI seem to recall them saying that they needed to steal the best bits from each other - so neither patch is perfect. But I could be wrong.
12:21:22B4gderI'll commit
12:22:42B4gderthere you go
12:28:04linuxstbIt doesn't seem to handle the ipod's byte-swapped LCD - but that should be easy to fix.
12:30:39 Join Nibbler [0] (
12:33:18*LinusN awaits the massive flood of "i can't build the h300 simulator" from the devkit (l)users
12:35:11*LinusN plays bejeweled in the h300 sim
12:35:54markunLinusN: don't do it man, you wont be able to stop..
12:37:09preglowit's hard
12:37:17preglowsomeone make puyo pop for rockbox and i will never code again
12:37:34LinusNis that a threat? ;-)
12:38:04 Quit akaidiot (Read error: 110 (Connection timed out))
12:39:22 Join Moos [0] (
12:41:33linuxstbLinusN: Does audio work OK for you in the SDL sim? I've tried FLAC, AAC and WAV files (in the ipod sim) and they all stop playing after about 2 seconds.
12:42:33 Join Vladoman [0] (
12:43:09LinusNlinuxstb: works fine for me
12:43:26LinusNonly tried mp3 though
12:43:43LinusNoops, stopped after 36 secs
12:44:04preglowso, what about the ifp patch?
12:44:17LinusNpreglow: i'm reluctant
12:44:34linuxstbI just tried a h300 sim, and audio seems much better - so far about 50 seconds of a FLAC file without problems.
12:44:37LinusNmany changes
12:44:55preglowi wish he'd drop in on irc
12:45:00LinusNbut on the other hand, we need to put it in cvs to be able to continue developing
12:45:26B4gderand to keep him happy/encouraged and make it easier for others to join in
12:45:28linuxstbAdding the ifp to cvs would help him a lot - especially with another ARM port going on at the same time.
12:46:07preglowthere aren't any more changes than any other device port would make, are there?
12:46:17linuxstbI'll try and find time to test his patch on the ipod later today - I haven't looked at it yet.
12:46:25preglowi think it should work
12:46:35 Quit ehntoo (Remote closed the connection)
12:47:10preglowi don't get his set_irq_level function
12:47:19preglowshouldn't that set priority level or something?
12:47:25preglowhis implementation just masks either fiq or irq processing
12:48:12LinusNpreglow: that depends on the architecture
12:50:49preglowoh? the name is kind of misleading in this case
12:52:25 Join actionshrimp [0] (
12:52:32LinusNwell, some cpu's have irq levels, some don't
12:53:16preglowi expect this to have irq levels too
12:53:35preglowoh well
12:54:14LinusNare the levels a function of the cpu arch or a separate irq controller?
12:55:04preglowarm arch itself only has that masking functionaloity
12:55:16preglowat least this is the case for portalplayer
12:55:31LinusNthen that should be what set_irq_level() should do
12:55:49preglowthe masking?
12:56:13preglowany reason for that? makes more sense for me to make it speak to the irq controller
12:56:34LinusNin fact, that function could just as well be called disable_irq(), because that's how it is used
12:59:21LinusNthe reason is that the function was developed for the archos, where the irq levels were native to the cpu (or at least the interrupt controller was on-die)
12:59:37LinusNand it translated fine to the coldfire irq levels
13:00:44 Join webguest55 [0] (
13:01:02preglowsure, it's on-die here as well, it's just not part of the core
13:01:05preglowarm core, that is
13:01:39LinusNi'd say use what suits your purposes best
13:02:05LinusNit would of course be nice if the function worked on all arm cores
13:02:29LinusNit has turned out that we haven't had any use for the levels anyway
13:02:49LinusNwe always use MAX_IRQ_LEVEL or 0
13:03:01preglowthen we might as well just let it be
13:03:25LinusNwe could even make it into a disable_irq() function
13:07:30 Quit _FireFly_ ("Leaving")
13:08:45webguest55Can the sdl sim be built for windows?
13:08:52linuxstbpreglow: Have you given any thought about the audio driver for the ipod? IPL uses an 80KB ring buffer in IRAM - which is something we obviously can't duplicate, at least not on my ipod.
13:09:47LinusNwebguest55: yes, but you have to install the sdl libraries
13:12:13 Join petur [0] (
13:12:31webguest55Silly me, I meant if it could be cross-compiled
13:12:56webguest55That is, if you can compile sdl-w32 on linux
13:13:17LinusNshould work...
13:14:05webguest55Will the configure script do it? Can't seem to figure out what I'd need to answer
13:14:06LinusNthe grayscales seem wrong in the h100 sdl sim
13:14:18preglowlinuxstb: ahahahahahahha
13:14:28preglow80kb ring buffer _IN IRAM_
13:14:36*preglow wipes tears
13:14:41webguest55That sounds... extremely retarded
13:14:52LinusNwebguest55: the config script can't do it for you, i'm afraid
13:15:21preglowlinuxstb: but ok, does it use dma?
13:15:31webguest55LinusN: Alright. I don't need it anyway, was just curious
13:17:58preglowlinuxstb: for now i don't think we need to think about that at all
13:18:03preglowlinuxstb: we're just using one core anyway
13:18:21preglowand the iram issue is probably just a way to work around cache discrepansies
13:18:34 Join aliask [0] (
13:20:25linuxstbNo, it doesn't use DMA - the functions have "dma" in their name, but I think that's just because it sort of simulates DMA by using the COP.
13:21:16linuxstbThe problem is that I wasn't able to get the FIQ working on the main CPU.
13:35:34preglowunable to make it call?
13:35:40preglowsure the fiq mask isn't set for the cpu?
13:35:49preglowand then i mean the fiq controller one, not the cpsr one
13:37:17 Quit ashridah (Remote closed the connection)
13:37:41linuxstbIt was a few weeks ago when I tried, but I think that was the problem - the IPL code only does it for the COP, so I don't know the memory addresses to do the same on the main CPU.
13:37:41 Quit webguest55 ("CGI:IRC (EOF)")
13:38:05preglowi think everything you need is in the wiki
13:38:12preglowthe PP5020 page
13:38:21preglowi haven't got time to give it a go now, i'm afraid
13:38:57 Join ashridah [0] (
13:39:04 Quit Nibbler (Read error: 110 (Connection timed out))
13:39:12 Join Nibbler [0] (
13:39:15LinusNanyone else than me who thinks the h100 sdl sim grayscales are wrong?
13:40:30linuxstbpreglow: I'll give it another go and see how far I get this time.
13:41:10preglowdid nathan hand just post... nothing?
13:41:22LinusN00 isn't white and 11 isn't black
13:44:21B4gderthe color magic is in sdl/lcd-x11.c
13:44:38LinusNi have fixed it, but i wonder if i should commit it or not
13:44:50linuxstbWhy not?
13:44:53LinusNmaybe the author intended it to look more like the actual lced
13:45:01B4gderI doubt that
13:45:35B4gderby looking at the code
13:45:46B4gderif so, it would've used a table lookup or similar
13:46:26LinusNi have committed a fix
13:47:36B4gderI think I've fixed the reds too
13:48:35B4gderthose builds didn't pick a simulator type but got the default
13:48:43B4gderand the default is now sdl
13:53:05***Saving seen data "./dancer.seen"
13:58:57preglowthink i like the wps token parser
14:00:48 Quit Rob2222 (Read error: 104 (Connection reset by peer))
14:01:22 Join Rob2222 [0] (
14:18:32 Join akaidiot [0] (
14:19:36 Join lokki [0] (
14:20:21markunIdeal companion for your Kalashnikov:
14:21:47 Join linuxstb_ [0] (
14:22:35lokkianybody here that knows the peakmeter code?
14:23:03 Part LinusN
14:24:09 Quit linuxstb (Read error: 110 (Connection timed out))
14:27:16 Quit lokki ("CGI:IRC (EOF)")
14:27:19 Quit ashridah ("Leaving")
14:29:13 Join Kohlrabi [0] (
14:38:13darklesshow does the replaygain preamp work? does it amplify the signal first and then EQ it down to "replaygain level (eg.79 dB)+preamp"? or does it apply replaygain first and then amp the result?
14:38:55godzirraSomeone yesterday told me the name of a great tag editor they use for their iriver.. anyone know who or what it was?
14:38:58godzirrasomething like TDT?
14:39:30darklessobviously, the second approach might kill some low-volume details compared to the first approach
14:39:39markundarkless: It just adds the preamp value to the replaygain value and scales every same with it
14:40:05darklessmarkun: thanks
14:41:38B4gderno "TDT" in that list
14:43:49darklessI looked through the wiki but couldn't find any info on which crossfeed algorithm is used
14:44:10preglowit's a homebrew one
14:44:21preglowwe're in the process of investigating alternatives
14:44:36darklesspreglow: ah, ok :)
14:44:41markundarkless: do you have some algorithms we could implement?
14:45:03darklessmarkun: no, I was merely curious
14:45:40markunpreglow: do you use crossfeed?
14:47:48 Quit akaidiot (Read error: 110 (Connection timed out))
14:48:42aliaskDoes crossfeed/stereo width work on the H300s?
14:48:48darklessthe crossfeed algorithms developed by meier-audio and used in his products are public domain IIRC
14:49:11darklessand his crossfeed circuits sound rather nice :)
14:49:36markundarkless: it was lightly based on his crossfeed
14:49:49markunaliask: stereo width doesn't
14:50:15aliaskAh ok, I thought I was missing out. How are stereo width and crossfeed different anyway?
14:52:07darklessstereowidth compresses the soundstage, whereas crossfeed adds a little bit of each channel to the opposite (with added delays, etc)
14:53:51darklessI have a Behringer DEQ2496 which allows me to compress or expand the soundstage by adjusting the stereowidth
14:55:01darklessSimon & Garfunkel - Bridge Over Troubled Water has a hard-panned drumkit which is highly annoying to me when listening with headphones
14:56:59darklessI tried to reduce the stereowidth, but I had to turn it all the way down to 0 (mono) in order to not focus on the drumkit. A better solution was to rotate the soundstage by 15% (another feature of the DEQ), thereby putting the drumkit in the center of the soundstage
14:57:40preglowmarkun: no, i don't
14:57:44darklessI tried using the rockbox crossfeed before experimenting with the other options
14:58:04darklessbut the crossfeed wasn't strong enough to compensate
14:58:05preglowi will when i start listening to music that needs it, though
14:59:05darklesspreglow: like the beatles? they have several hard-panned stereotracks
14:59:39markundarkless: preglow and I used LSD a lot for testing (the song, not the drug)
14:59:54darklessgood one
15:00:28aliaskIt's all too common in old songs.
15:00:42preglowdarkless: i know
15:01:00preglowproblem is i haven't listened to any really old music for a while
15:01:23darklesswould it be possible to make a scaling crossfeed using the existing algorithm?
15:01:38markunI also like crossfeed for a lot of newer songs
15:02:18markundarkless: you want to be able to change the crossfeed parameters?
15:04:57 Quit eli_sherer (Read error: 110 (Connection timed out))
15:05:03darklessmarkun: I'd like to make the crossfeed "stronger" on certain tracks. A scaling crossfeed would allow me to do that
15:05:26preglowwe'll probably allow you to adjust the crossfade when we make a better one
15:07:02 Part petur
15:07:03 Quit edx (Read error: 104 (Connection reset by peer))
15:07:16preglowthe current one is very hard to tune
15:07:19preglowso we just dropped it there
15:07:23 Quit aliask ("-yawn-")
15:11:34 Quit Vladoman (Read error: 110 (Connection timed out))
15:11:56 Join Vladoman [0] (
15:15:36 Quit Rob2222 (Read error: 104 (Connection reset by peer))
15:15:55 Quit Febs (Read error: 110 (Connection timed out))
15:16:30 Join Rob2222 [0] (
15:22:02 Join webguest18 [0] (
15:23:02 Join Jungti1234 [0] (n=jungti12@
15:26:27 Quit gtkspert (Read error: 110 (Connection timed out))
15:31:24preglowlostlogic: decided on what to buy?
15:33:12Jungti1234hi all
15:33:17Jungti1234I came back
15:45:57markunpreglow: he bought a US H340 on ebay
15:46:29 Quit psi (Nick collision from services.)
15:49:21 Join muesli- [0] (n=muesli_t@
15:49:25 Join psi` [0] (
15:52:36 Quit muesli- (Client Quit)
15:52:49 Quit Jungti1234 ()
15:53:07***Saving seen data "./dancer.seen"
16:06:14 Quit Zagor ("Client exiting")
16:07:13 Join Febs [0] (
16:08:42 Quit YouCeyE ("Leaving")
16:16:43lostlogicpreglow: ebayed a used US H340
16:17:26 Join Shani}{Coder [0] (i=Coder@Is.A.TclCoder.Org)
16:23:12godzirrahow much?
16:27:40FebsThat's a good price.
16:27:59lostlogicyeah −− I figured if I could get it for that, I'd take it, toehrwise I'd get an IPV or X5L for $300.
16:28:54 Quit webguest18 ("CGI:IRC")
16:30:15 Join akaidiot [0] (
16:35:59 Quit Shani}{Coder (Read error: 60 (Operation timed out))
16:36:22godzirraThats not bad.
16:36:54 Join Shani}{Coder [0] (i=Coder@Is.A.TclCoder.Org)
16:38:04 Join DangerousDan [0] (
16:44:07 Join eli_sherer [0] (
16:45:10*eli_sherer get frustrated from his secret project %(
16:50:32markunThen just release them into the wild :)
16:50:45preglowcan't help you with a secret project
16:55:14 Join DrMoos [0] (
16:56:42 Join Maxime` [0] (
16:58:17 Quit Moos (Read error: 104 (Connection reset by peer))
16:58:27godzirraOoh... a secret project.
16:58:36godzirraI need a white lab coat and electric socket hair for my secret project.
16:59:24 Quit akaidiot (Read error: 110 (Connection timed out))
16:59:54 Quit Maxime (Read error: 104 (Connection reset by peer))
17:01:40 Quit Febs ("CGI:IRC (EOF)")
17:05:50 Join edx [0] (
17:05:52 Quit DangerousDan (Read error: 104 (Connection reset by peer))
17:06:32 Join DangerousDan [0] (
17:09:46 Join muesli__ [0] (i=muesli_t@
17:10:49 Join Mmmm [0] (
17:13:42 Quit B4gder ("time to say moo")
17:15:54markunMmmm: did you work on the metronome plugin?
17:21:06Mmmmyeah...with Shebbs help.. now it purrrs! :D
17:21:23Mmmmthanks for the array!
17:21:49markunBut you didn't get it to work?
17:22:19 Join akaidiot [0] (
17:22:20Mmmmhave a look if you like
17:23:05Mmmmworks gorgeously!
17:24:53markunI'll try it when I get home
17:25:31MmmmI got the sync button working too and fixed the remote support
17:29:07 Quit edx (Read error: 110 (Connection timed out))
17:48:59 Quit akaidiot (Read error: 110 (Connection timed out))
17:53:08***Saving seen data "./dancer.seen"
17:59:39 Join San [0] (
18:01:49 Join gtkspert [0] (
18:03:34 Join hshah [0] (
18:09:33 Join NicoFR [0] (
18:28:02 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
18:28:26 Part hshah ("Leaving")
18:29:42 Join Mark_ [0] (
18:38:32 Quit DreamTactix291 (Read error: 104 (Connection reset by peer))
18:40:08 Join DreamTactix291 [0] (
18:51:11 Join Lear [0] (
18:58:49*preglow explores the murky depths of tiredness
18:59:19markunjust go to sleep man :)
19:00:54 Join linuxstb__ [0] (
19:01:08preglowi just did
19:01:23preglowin a very awkward position while trying to read
19:01:28preglowdidn't work out, and now my neck hurts
19:01:36 Quit linuxstb_ (Read error: 110 (Connection timed out))
19:01:47preglowa few more hours and i'll give in
19:03:03gtkspertwhat time is it there?
19:04:57gtkspertwow, its 2:04 here...
19:11:15 Join edx [0] (
19:16:46markunMmmm: yes, the metronome works very nice
19:17:50MmmmCheers... the only thing is the fading lcd seems to flicker a lot... cpu not being boosted?
19:18:01markunYes, I also noticed that
19:18:04preglowwhich unit?
19:18:09markunH120 here
19:18:22preglowfading lcd flickers?
19:18:30markunonly in the metronome plugin
19:18:31preglowi've never noticed that
19:18:33preglowahh, right
19:18:49preglowwell, the backlight should deal with the boosting issue itself
19:19:01MmmmThats what I thought
19:19:07markunyes, trange
19:19:09preglowso, unless the metronome plugin boosts, which i can't imagine why it should...
19:20:38Mmmmcan you disable lcd fade from within a plugin?
19:24:04 Join webguest34 [0] (
19:24:23webguest34Isn't a flickering backlight-fade signs of lack of yielding?
19:24:54 Join einhirn [0] (
19:25:23 Join lokki [0] (
19:25:59lokkiMmh: concerning your patch, how do i tap the tempo?
19:26:24Mmmmpress the joystick or pretty much any button on the remote
19:26:48 Join petur [0] (
19:26:59 Join linuxstb_ [0] (
19:27:13lokkiyes i did that... but it does not update the tempo, it just begins with a new "1" the tempo stays the same
19:27:44 Join Thus0 [0] (
19:28:08 Quit linuxstb__ (Read error: 110 (Connection timed out))
19:28:22Mmmmwhat do you mean by tap then? I thought that was to sync the beat with whatever you like
19:29:07lokkino i mean tap the joystick for like six or seven times and then adjust the tempo to the tap tempo
19:29:29lokkilike on some delay pedals or similar
19:29:59Mmmmahhh... i see... I didnt think of that. :D I'll look into it later!
19:30:19 Join akaidiot [0] (
19:30:31lokkiit's not so important, just missunderstood you...
19:31:20lokkii need help with a patch though... it's about the peakmeter, anybody?
19:33:31MmmmGot to go and get some dinner...see ya!
19:34:35 Quit Mmmm ("goes to heat the Wok...")
19:43:14 Join fobbo [0] (
19:46:54 Quit webguest34 ("CGI:IRC (EOF)")
19:51:14 Quit Shani}{Coder (Read error: 113 (No route to host))
19:51:57 Join xmixahlx [0] (n=xmixahlx@
19:53:10***Saving seen data "./dancer.seen"
19:55:30mirakmarkun: hi
19:55:39mirakI can't spot why deconding fail
19:55:44 Quit akaidiot (Read error: 110 (Connection timed out))
19:56:12mirakthere must be an endian problem somewhere or something, tough it works on linux ppc also wich is big endian
19:56:15mirakso I am not sure
19:58:47 Quit gtkspert (Read error: 104 (Connection reset by peer))
19:58:58markunmirak: :(
19:59:36mirakit's not easy to trace what happens
20:00:00 Quit Thus0 (Connection timed out)
20:00:28 Join Thus0 [0] (
20:04:26 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
20:07:41peturapps/recorder/recording.c has more ifdefs than source code :(
20:17:29 Nick DrMoos is now known as Moos (
20:21:01lostlogicpetur: perhaps should be separated into different files witih #ifdefs in SOURCES?
20:26:04 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
20:37:04preglowrockbox went mad again
20:37:10preglowstarted reading track after track after track
20:37:14preglownot playing any of them
20:37:42Learwhat kind of files?
20:38:36preglowbut i think it happened with oggs yesterday
20:38:51pregloweach time it happen when i selected a new file as rockbox started buffering the new dir
20:38:55preglowi _THINK_
20:39:00preglowthere's a race condition going on there
20:39:04preglowi'm convinced of it
20:39:28Learah, I seldom do things like that, which would explain why I've never had problems like that.
20:39:39t0maspreglow? steps to reproduce?
20:39:46t0mas1 select a dir and make it start playing
20:39:55t0mas2 immidiately select a new file?
20:40:06preglowi've just played one dir
20:40:16preglowand so, at the very end of the last track, rockbox starts buffering the next dir
20:40:20preglowwhile this happens, i select something else
20:40:22preglowi think that's the key
20:40:55t0masah, you have skip to next dir enabled?
20:41:35t0masthat might explain why I've never seen it
20:41:53preglowthat shouldn't matter, it should happen whener rockbox is loading something new
20:41:55t0mas(it's called auto change directory now right?)
20:42:01preglowgod knows what it's called
20:42:27t0masI'll enable it
20:42:42t0massee if it happens here too... and if I can find out when exactly
20:42:51t0masmaybe you can find the bug then :)
20:43:40 Join linuxstb__ [0] (
20:45:56 Quit linuxstb_ (Read error: 110 (Connection timed out))
20:50:44Moosthis auto change dir feature wasn't finished yet, since len0x seems have deserted Rockbox
20:50:53preglowyes, fine
20:51:23preglowwell, he hasn't been gone that long
20:51:34preglowi hope timid hasn't deserted rockbox :/
20:52:12MoosI have receive one mail from him, and he is fine, just in exams session... and no internet connexion
20:52:32Moosbut he will have soon, and he will be back
20:53:04Mooshis multi screen support wasn't finished too :)
20:53:45preglowyeah, i know
21:01:37 Join linuxstb_ [0] (
21:02:48 Quit linuxstb__ (Read error: 110 (Connection timed out))
21:04:32 Join _FireFly_ [0] (
21:13:44peturpreglow: I recently saw rb do the same thing (constantly skipping to next song) - no hints to what I did to trigger it...
21:15:23 Join vger_ [0] (
21:16:09preglowgood, at least it's not only my player acting the fool
21:21:48 Join lodesi [0] (
21:27:21peturunable to trigger it again, will keep an eye on it...
21:28:54peturwhich of the RB supported players has the smallest screen (character-count-wise)?
21:29:03preglowdepends on the font...
21:29:42peturnevermind, I'll do a special case for the iRivers...
21:36:18 Join akaidiot [0] (
21:36:59 Nick linuxstb_ is now known as linuxstb (
21:39:46 Join jlo [0] (
21:40:26preglownow, there you are
21:40:45jlohi preglow, nice to hear you
21:41:15 Join mlind [0] (
21:41:47jlopreglow, something new about crossfeed ?
21:41:58preglowno, nothing
21:42:04preglowyou had me do a blind test, then vanished :>
21:42:37jlodid you download the samples I prepared and test it ?
21:43:11preglowyes, i did, but i can no longer remember my results...
21:43:43preglownot so weird, since it's over a month ago
21:44:30jloI did 4 types of crossfeed, but I
21:44:53jloI'd like to have you results ;)
21:45:50jlobut if you did it with ABC/HR maybe you saved the results ?
21:46:13 Join Mmmm [0] (
21:46:29mlindHi guys! Anybody compiling on OS X? I'm having trouble setting up things...
21:48:18linuxstbYes, but it was a while ago when I set everything up. I think I installed fink. and then installed the required packages (e.g. gettext) using fink.
21:48:47jlopreglow, do you want me to do other files to test or resend lucy in the sky (its' still available)
21:49:18preglowresend will do nicely
21:49:25preglowand i'll test tomorrow
21:49:39preglowi'm dead tired now and need to sleep desperately, heh
21:50:00mlindI was told I didn't need anything more than gcc, gdb & binutils. What's fink?
21:50:30linuxstbpreglow: You posted your crossfeed findings to IRC on 2005-12-14 at about 16:35 GMT
21:51:03MmmmLokki, Markun: here's a new metronome patch with tap function...HOOORAAAHH, just for you lokki :) Great idea, it was in the original code but I didn't know what it did! (I've kept my sync button too because I think that is also handy)
21:51:09linuxstbYou liked 3.wav best, and found 2.wav to be the worse.
21:51:31 Quit fobbo ()
21:52:26jlothanks linuxstb, files are at, I'll check which is which !!!
21:52:26markunmlind: do you get an error?
21:53:13***Saving seen data "./dancer.seen"
21:54:10linuxstbmlind: - it's a system for installing packages of GNU and other open source software on Mac OS X.
21:54:43jlopreglow and linuxstb : winamp=1 rockbox=2 reaktor=3
21:55:00linuxstbmlind: But as markun says, what specific problem are you having?
21:55:31mlindmarkun: yes. When I try to setup binutils it says "C compilier cannot create executables" See:
21:56:35linuxstbSeems you don't have gcc installed properly.
21:57:30linuxstbHave you installed Xcode?
21:58:53 Quit _FireFly_ ("Leaving")
21:59:57mlindNo Xcode. Doing everything in the order described here:
22:00:04 Join solexx_ [0] (
22:00:38linuxstbMac OS X doesn't have gcc installed by default - you need to install Xcode, which is Apple's development environment.
22:01:15 Join __ste__ [0] (
22:01:18linuxstbWhich version of Mac OS X do you have?
22:01:19preglowjlo: so i liked you algo best, then
22:02:09preglowdidn't you post a schematic for it somewhere?
22:02:10mlindlinuxstb: Ok. Thanks. I kind of thought Xcode just was an alternative to doing it the wiki way.
22:02:16markunpreglow: well, start programming :)
22:03:17Mmmmmarkun: have you tried new metronome patch?
22:04:03jloyou know, I think we could maybe improve some values we need your programming skills ;)
22:05:11preglowi want this crossfeed to be fully parameterisable
22:05:20preglowwith decent defaults, of course
22:05:49Mmmmmarkun: tis quite strange, now the lcd fade seems to be disabled!
22:09:22jloto be in the parameters : delay (10 to 18 samples), Fc between 400Hz and 3KHz, Levels : direct and opposite in high and low (4 parameters)
22:10:36preglowFc? as in both filters use the same cutoff freq?
22:11:13jloyes, I used same frequency for low and high pass
22:12:05 Quit solexx (Read error: 110 (Connection timed out))
22:12:27preglowfirst order filters?
22:12:50jlofirst order but shelving !
22:13:14preglowthen same cutoff is irrelevant
22:13:45preglowstop band attenuation=?
22:14:26 Join [San] [0] (
22:15:40jlostopband attenuation was about 4dB in lows, and 12 dB in highs (I have to exactly check), why do you mean same Fc is irrelevant ?
22:15:58preglowbecause i can't calculate a complementary filter with a simple operation
22:16:37preglowlike i could if the filters were low/highpass
22:16:38jloyou could do two filters independantly if it's easier
22:16:48preglowno, not easier, it's just an optimise thing
22:16:55preglowbut shelving
22:17:03preglowbut ok, the code will be a bit slower than the current one, then
22:17:11preglowbut i guess it's acceptable for the quality
22:17:31 Quit San (Read error: 104 (Connection reset by peer))
22:18:18jlofor shelving : can't you do a filter and add an allpass attenuated by the stopband value ?
22:18:39preglowwhy allpass?
22:18:47preglowthat would give phase problems
22:18:56jlowell not the allpass, just the original signal
22:18:56preglowjust adding the straight input should do, i think
22:19:12jloI think so
22:19:17preglowwe'll see
22:19:22preglowi wont have time for a while anyway
22:21:12jlowhen you have time, .... but in between, I'm ready to encode and try with other musics
22:21:58preglowthat would be nice
22:22:11preglowbut yeah, where is the block diag of your algo?
22:23:01jloI'll do a pdf in you want and post it on my ftp
22:23:57jloand if other want to try, I can give the Reaktor file also
22:25:19preglowdon't have reaktor here
22:25:20preglowbut anyway
22:25:21preglowno time now
22:25:28preglowand now i'm damn near to falling asleep
22:25:31preglowso time for bed
22:25:42 Quit __ste__ (Read error: 110 (Connection timed out))
22:25:49preglownight, all
22:25:58jlonight preglow
22:35:45jlogoodnight everybody
22:35:48 Quit jlo ()
22:45:19 Quit xmixahlx ("blah blah blah")
22:47:24 Quit Moos ("Happy new year @ all")
22:48:06 Quit akaidiot (Connection timed out)
22:49:33 Quit actionshrimp (Read error: 110 (Connection timed out))
22:51:39 Join mirak_ [0] (
22:53:37 Quit Lear ("Chatzilla 0.9.69 [Firefox 1.5/0000000000]")
22:58:53[San]whats xobox?
22:58:59[San]a sim like rockboy?
22:59:11 Quit linuxstb (Read error: 110 (Connection timed out))
22:59:16[San]link is down
22:59:27 Join linuxstb [0] (
23:00:14Papricaeli dont upload the files
23:03:15[San]is it finished?
23:03:25 Join edx__ [0] (
23:03:40[San]but, like released?
23:03:52Papricamm dont know
23:03:54Papricaask eli
23:04:07Papricagood night
23:04:21 Nick Paprica is now known as Paprica[sleep] (
23:06:49 Quit edx (Read error: 110 (Connection timed out))
23:07:15 Quit mirak (Connection timed out)
23:10:30 Quit vger_ ("gtkBitchX for president.")
23:12:44 Quit Mmmm ()
23:15:08 Join ehntoo [0] (
23:23:44 Part linuxstb ("Leaving")
23:27:32 Join linuxstb [0] (
23:29:49 Part mlind
23:31:39 Join sa0 [0] (
23:31:43sa0hi all
23:31:58sa0ok for IriverPort here ?
23:32:49sa0k thx
23:33:24sa0was wondering if it will be possible one day to add the USB OTG Host function to ihp series ?
23:34:23linuxstbYes, but it's a big job, and I don't think anyone has started yet.
23:34:55sa0i wasn't sure, amazing
23:34:58linuxstbOr are you asking about adding USBOTG to the h1x0 ?
23:35:08linuxstbWhich isn't possible.
23:35:09sa0just i've to hope for it so
23:35:36sa0yes that's what i was asking ... ;o(
23:35:41sa0so why not ?
23:36:08linuxstbBecause the h1x0 lacks the hardware to do it.
23:36:16linuxstbThe h3x0 has a special USBOTG chip.
23:37:09sa0oh ok clear
23:37:14sa0thx linuxstb
23:42:32 Join hd [0] (
23:42:39 Join RotAtoR [0] (
23:43:49 Quit sa0 ("( :: NoNameScript 3.81 :: )")
23:53:17***Saving seen data "./dancer.seen"
23:54:41 Quit goa (Read error: 110 (Connection timed out))
23:54:41 Nick hd is now known as goa (
23:55:26 Quit NicoFR ()
23:58:32 Join LinusN [0] (

Previous day | Next day