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 2006-08-02

00:02:16barrywardelli sent of my H10 to iriver today, so in a couple of weeks, I will be able to get back to it
00:04:54 Join safetydan [0] (
00:05:46markunBagder, LinusN: are the SID files supposed to be listened to in mono?
00:06:00preglowbarrywardell: sure they'll send you one back, though? :>
00:06:13preglowmarkun: you'd need more than one sid chip for stereo sound
00:06:28preglowthe chip itself is mono
00:07:08markunBut we could playback the 4 channels in different 'stereo positions'
00:07:10barrywardellpreglow: I'm not sure, but hope so! iriver are quite good about RMAs
00:07:23preglowmarkun: why?
00:07:28preglowmarkun: besides, it's three channels
00:07:35Mikachupan them around like crazy
00:07:38 Join Gibbed [0] (
00:07:38 Quit Rick (Nick collision from services.)
00:07:50 Nick Gibbed is now known as Rick (
00:07:52LinusNBg3r: there?
00:08:04preglowthree oscillators, i think some people kind amplitude modulating the output as a fourth channel, but i really wouldn't count it as such
00:08:09safetydanpreglow, would it be a crazy idea to use the existing EQ UI for the ipod 5g hw eq? I was thinking having an option somewhere that says "Software or Hardware EQ" and then having the UI limit the possible settings based on that
00:08:11 Quit joe2 (Connection timed out)
00:08:22preglowsafetydan: don't see why not
00:08:53 Quit klrspz (Read error: 110 (Connection timed out))
00:09:39safetydanwell that should be easy enough then if someone implements the interface to the hardware
00:10:40preglowshouldn't be too hard
00:10:43preglowmake linuxstb do it!
00:12:09*safetydan looks at linuxstb
00:12:37safetydanthough how would it work with an eq graph based interface like the one Febs proposed?
00:12:41preglowyou need to tickle him for maximum effect
00:12:48preglowsafetydan: not at all, heh
00:13:09preglowi don't actually know the response curve, so it'd all just be educated guessing
00:13:20preglowthough it is measurable, of course
00:13:58Bagdernow it builds at least
00:14:31safetydanLarge parts of Rockbox are based on educated guessing so it wouldn't be the first time :)
00:15:44preglowi'm guessing at least the peaking filters are exactly what i use
00:15:51preglowyou can't really do too many variations on these things
00:17:25*LinusN spots a sandisk commit
00:17:42Bagdermore coming up
00:18:01Bagderbasically to get a reaaaally basic bootloader to get built
00:18:42preglowi want portalplayer info!"¤
00:18:45Bagder"We are working on getting you the development board with JTAG to help accomplish this...."
00:20:22preglowif so, stuff that thing in a clean room and touch it only with one foot planted halfway into solid conducting wet soil
00:21:49 Quit lodesi ("Parti")
00:22:11 Quit Kohlrabi ("Quit")
00:22:53fraggstawait what, SID playback?!
00:24:08fraggstais there MOD playback?
00:27:10preglowahh, mod hasn't been commited yet?
00:27:25fraggstaI found a patch, but it looks a little limited
00:27:49preglowthe current codec system isn't really well suited to that kind of format
00:27:53barrywardellBagder: was that a reply from sandisk?
00:28:19Bagderand now there's a big commit
00:28:35*preglow does the new target dance
00:28:57fraggstawhy is there the 512 Kb sample limit? are plugins limited in memory usage?
00:29:17Bagderbarrywardell: I would appreciate if you could merge with CVS now and update your patch. I hope I didn't mess it up for you too much
00:29:42barrywardellwill do
00:29:52barrywardelljust doing a cvs update now
00:31:54Bagderbarrywardell: I forgot to define TARGET_TREE in config-h10.h
00:32:46 Part fraggsta ("Ex-Chat")
00:32:49Bagderhm, and all x5 builders will get an annoying range of warnings until configure is re-run
00:36:05 Join TCK [0] (
00:36:15 Join midkay [0] (n=midkay@rockbox/developer/midkay)
00:36:43LinusNBagder: i named the ata.c file in the x5 directory ata-x5.c
00:36:57LinusNnot because it is necessary
00:37:02LinusNbut for convenience
00:37:14Bagder5000 points
00:37:18Bagdernow that's a lot
00:37:29LinusNi think ata-e200.c would be a good idea
00:37:35BagderI agree
00:37:41Bagderwill fix
00:37:49LinusNsame goes for the other files
00:39:27 Join tvelocity [0] (
00:45:54LinusNnite all
00:45:59 Quit petur ("here today, gone tomorrow")
00:46:01 Part LinusN
00:50:32amiconnArchos gamma curve redone. Looking quite nice now too :)
01:00:35 Quit obo ("bye")
01:00:36markunamiconn: btw, what was the doom "headache bug" you were talking about?
01:01:08amiconnWell, doom on H1x0 causes a headache, because it's way too dark
01:01:44preglowwhy would anyone play doom on h1x0? :V
01:01:59markunit was way to blurry for me
01:02:03amiconnAsk that anyone ;)
01:04:06***Saving seen data "./dancer.seen"
01:05:48BigMaci dont even play doom on my ipod anymore
01:06:00BigMacit was fun for a day then lost all fun
01:07:17 Quit scottder (Read error: 104 (Connection reset by peer))
01:08:22*amiconn would have chosen slightly different names for the sansa target dirs
01:09:17markunif we switch to svn we can always rename folders and files.. :)
01:11:07 Quit Rick (Read error: 110 (Connection timed out))
01:13:19 Join mikearthur [0] (
01:13:52barrywardellBagder: in bootloader/SOURCES, #elif defined(IRIVER_H10) is never reached because CONFIG_CPU==PP5020 is reached first.
01:15:35 Join Rick [0] (
01:17:59*linuxstb returns
01:23:31barrywardellwould it be better to check for IPOD_ARCH==1 instead of doing a cpu check?
01:25:55Astrodammit, my X5 just crushed me at chess :(
01:26:13linuxstbbarrywardell: Yes - bootloader/SOURCES should check IPOD_ARCH...
01:26:33Astroand I used to consider myself at least semi-decent....the AI isn't bad
01:27:06preglowit isn't super either
01:27:36AstroI'm sure it's not, but it was better than me this time :)
01:28:45linuxstbbarrywardell: Fixed in CVS.
01:28:57barrywardellthanks linuxstb
01:29:24linuxstbbarrywardell: Any news on your H10? Is it now with iriver?
01:29:39barrywardelli only got around to posting it off today
01:29:49barrywardell3-5 days for delivery to iriver
01:30:37barrywardellmy brother sent his to them for repair a while ago and it took about two weeks
01:30:54 Quit mikearthur (Remote closed the connection)
01:31:14linuxstbSo you can brick your brother's H10 now?
01:32:35 Join mikearthur [0] (
01:33:11 Quit mikearthur (Client Quit)
01:33:16barrywardelllol. he wouldn't be happy
01:33:32preglowwhat did you do to brick it anyway?
01:34:42linuxstbI exaggerate... Sounds like a coincidental hardware failure to me.
01:35:13barrywardelli wish i knew. was running a modified bootloader. but i had done it several times before without problems
01:35:26barrywardellmay have been a coincidence
01:35:58barrywardellby bootloader, I mean rockbox ipod bootloader
01:36:28preglownothing special?
01:37:15barrywardelli was reading button presses from gpio and writing them to disk
01:38:39preglowdoesn't sound very perilous
01:39:29barrywardellmy h10 was in a dismantled state at the time
01:40:25Astroquick question: what's the difference between the X5 and the X5V ? V = video?
01:41:27preglowgreater battery capacity, i believe
01:41:49Astrohmm...that's more commonly known as X5L, I believe
01:42:00preglowright, then ignore me
01:42:05preglowi don't even own one
01:42:14Astronever seen the term X5V before, that's why I'm asking...
01:42:36amiconnThe X5V is a simple version of the X5 afaik
01:42:41amiconnNo radio, no recording
01:42:52Astroah, ok
01:53:22 Join rus1 [0] (
01:54:31 Join enzyme [0] (n=dave_o@
01:55:04enzymeis it possible to put a macosx on my laptop
01:56:19rus1why are you asking that here?
01:57:01 Join JdGordon [0] (
01:57:24enzymei dunno
01:57:33enzymei was just woundering
01:58:37dan_a*sigh* I give up on this multi-CPU thing... for a day or two at least...
02:01:34 Quit enzyme ()
02:02:11 Quit mantono ("sleep!!")
02:03:08 Quit dan_a ()
02:06:39 Part Astro
02:06:49 Quit barrywardell ()
02:10:00*linuxstb sleeps
02:17:14 Join russ [0] (
02:22:40 Join _Lucretia_ [0] (
02:31:14 Join BloodySorcerer [0] (
02:34:50 Quit rus1 (Connection timed out)
02:39:37 Quit russ (Connection timed out)
02:41:47 Part pixelma
02:55:29 Quit spiorf ("Read error: 104 (Connection reset by peer)")
02:58:07 Quit MrStaticVoid ("james /was/ here")
03:00:03 Join Gibbed [0] (
03:00:05 Quit Rick (Nick collision from services.)
03:00:18 Nick Gibbed is now known as Rick (
03:04:08***Saving seen data "./dancer.seen"
03:13:24 Join excitatory [0] (
03:14:42 Quit tvelocity ("Ex-Chat")
03:26:33 Join Davide-NYC [0] (
03:33:04 Quit TeaSea ("Leaving")
04:07:49 Quit actionshrimp ("( :: NoNameScript 4.01 :: )")
04:09:11 Quit Davide-NYC ("Chatzilla 0.9.74 [Firefox]")
04:11:30 Join jn [0] (
04:11:48jnhey i have plenty of room on my ipod but its telling me that there is no space left
04:12:31jn3949 items, totalling 14.6 GB and it's a 30gb ipod photo...
04:12:32jncheck this out
04:12:36jncp: cannot create directory `/mnt/ipod/metal/Killwhitneydead-So_Pretty_So_Plastic_(Chopped_and_Screwed)-(Special_Edition)-TRACKFIX-2006-ARK': No space left on device
04:13:28BHSPitMonkeywhat does the command "df -h" say about the iPod?
04:14:27jn/dev/sda2 28G 28G 0 100% /mnt/ipod
04:14:32jnbut that's not right..
04:15:33Mikachutry fsck.vfat maybe
04:18:37jnwhat would the line be?
04:18:46jnfsck.vfat -a /dev/sda2?
04:18:50Mikachui would try -n first
04:19:21jnThere are differences between boot sector and its backup.
04:19:37jnFree cluster summary wrong (0 vs. really 3461445) Auto-correcting.
04:19:42jnthats it i think
04:19:47Mikachuthat would be the one, yes
04:20:02jnmm still says no space left
04:20:12Mikachu-n means don't do anything
04:20:19Mikachuso you want to run with -a now
04:20:30jnwill it mess up the bootsector?
04:20:46Mikachuyou can use -r if you want and only repair the second thing i suppose
04:21:07Mikachubut i don't think anything uses the bootsector of the filesystem
04:21:07jnwhat do you recommend?
04:21:15jnso -a then?
04:23:19 Join Davide-NYC [0] (
04:26:04 Quit jn ("Lost terminal")
04:34:38 Join jhMikeS [0] (
04:47:32 Join piroko [0] (
04:50:58 Quit ^jhMikeS^ (Read error: 110 (Connection timed out))
04:54:25 Quit Davide-NYC ("Chatzilla 0.9.74 [Firefox]")
05:01:09 Join Rob2222_ [0] (
05:03:40 Quit jhMikeS (Read error: 104 (Connection reset by peer))
05:03:46 Nick lukaswayne9_zzz is now known as lukaswayne9 (
05:04:10***Saving seen data "./dancer.seen"
05:05:21 Join jhMikeS [0] (
05:07:59 Join Davide-NYC [0] (
05:17:20 Part piroko ("Leaving")
05:18:22 Quit Rob2222 (Read error: 110 (Connection timed out))
05:23:48 Quit Criamos ("( :: NoNameScript 4.02 :: )")
05:33:49 Quit lukaswayne9 ("Ex-Chat")
05:41:40 Join RoC_MM [0] (
06:01:13 Part BigMac
06:05:55 Quit RoC_MM ("Leaving")
06:08:28 Quit Siku (Read error: 145 (Connection timed out))
06:28:22 Join lukaswayne9 [0] (
06:34:15 Join JoeBorn [0] (n=jborn@
06:41:21 Join infamis [0] (
06:43:09infamisfellow rockboxers
06:44:21infamisanybody ever think about what a "rock"-"box" would actually look like?
06:44:41goffame when thinking of an updated logo
06:44:52goffabut haven't thought of any ideas.. and everyone seems to like the current logo
06:44:57goffaso the idea passed :)
06:45:15infamisok, so 1 person's awoke...
06:46:02infamisor 'awake'...whatever the right word is
06:46:13goffahe he he
06:46:30goffaonly 10:46pm here
06:46:45scorchemany of us are...just some choose to pipe in at certain times =)
06:46:52BloodySorcererrocken boxen
06:47:09infamissame here plus an hour...well I guess it wouldn't be 'same here' if it was plus an hour...ah whatever
06:47:34BloodySorcerer00:47 here.
06:47:38BloodySorcererinsomnia here.
06:47:56infamisgo to sleep!
06:48:33BloodySorcereri have to wake up at 7:30
06:49:16infamisI wake up at 9am, so I got about an hour 'til it's needs my full 8hrs
06:49:28goffayeah... last month i had to be at work at 7am... so going in at 8 is like sleeping in
06:49:44goffai usually end up with around 6
06:50:06goffai'd like more.. but there's not enough time in a day
06:50:09BloodySorcereri usually end up with "not enough"
06:51:07infamisseems this place is kickin' when it's like 3am/4am over here...translates to morning time over in the "europe's"
06:51:49infamisbut anyways, main reason for coming here was to see if anyone experienced any odd playback issues recently
06:51:49safetydanit's around 3pm here
06:52:43safetydaninfamis, no haven't noticed any problems, but then again my build is like a week old
06:52:56scorcheinfamis: what problems?
06:53:48infamisbasically playback-unrecoverability(?) sometimes when fast-forwarding or rewinding.
06:54:01infamisonly fix is to shutdown
06:54:21 Join Harry [0] (
06:54:23Harryjoo when i fast-forward/rewind, the player ______
06:54:33scorchefill in the blank =P
06:55:10infamislol, the player ___is not responsive, and the only button that works is the power-off button____
06:55:10safetydaninfamis, a fix went in for some buffering issues a few days ago. Is your build more recent than that?
06:55:57scorcheso it is a total freeze...
06:56:12infamisbuild is as of aug 1, 1:00 cvs; the latest commit I have for this build is: 1 Aug 01:00Linus Nielsen Feltzingfirmware/target/coldfire/iaudio/x5/lcd-x5.c 1.11 Patch #5740 by Mike Sevakis - X5 LCD performance enhancements
06:57:16infamisI do see a 'align on 32 bit boundaries' recent commit, maybe that's a fix, but I haven't built yet
06:58:02Harryhow do i reg nick
06:58:24infamisit may be a total freeze, I think I waited up to a whole minute to see if it would recover to no avail
06:58:37scorcheHarry: we have told you to read the freenode FAQ many times...please do not come in here and repeatedly ask the same questions
06:58:49scorcheinfamis: sounds like it
06:59:00Harryscorche just tell me the command
06:59:09scorchefigure it out yourself
06:59:17goffaisn't it alt+f4?
06:59:24scorcheor go ask in #freenode or #freenode-help or whatever
06:59:25infamisyou mean ctrl+alt+del twice?
06:59:50 Nick Harry is now known as Harry| (
07:00:08safetydan infamis, what target?
07:00:19infamisthe x5
07:00:22scorcheinfamis: that does affect the audio buffer
07:00:39infamisyeah.....x5.... :)
07:00:55BloodySorcererwhat're we talking aboot?
07:01:05scorchebut, the x5 has 16megs of ram
07:01:12infamisfast-forwarding/rewinding sometimes crashing rb
07:01:21scorchehave you tried varying amounts of skipping?
07:01:27Harry|how do u spell regester
07:01:33scorcheeg, skipping a short bit, long bit
07:01:55scorcheHarry|: /ns help register
07:01:55safetydaninfamis, I don't know then, but I think similar things have been an issue for a while now
07:02:07scorchenow go away and leave us be!
07:02:26infamisthe thing is, it's a pretty recent issue....didn't have this problem before the recent playback.c commits
07:02:43infamis*if those are the culprit
07:02:49scorcheinfamis: can you try what i asked above? =)
07:03:07safetydanYou could try and summon Slasheri or lostlogic since they're the playback.c gurus
07:03:22infamisyea I'll try it isn't really that easy to reproduce, sometimes it happens, sometimes it doesn't.......for the most part it doesn't
07:03:37scorchewell, i will be back...
07:03:43infamisI'd say out of 3 hours of playback tonight, it happened about 4 times.
07:04:12***Saving seen data "./dancer.seen"
07:04:51infamisinsta-skip (tm) is awesome though....
07:04:52 Quit Harry| (Read error: 104 (Connection reset by peer))
07:07:24 Quit TCK (Read error: 54 (Connection reset by peer))
07:08:48 Quit infamis ("CGI:IRC (EOF)")
07:29:20 Join Hideo_Kuze [0] (i=irc@gateway/tor/x-422248dc96f4574c)
07:30:00Hideo_KuzeHi guys, is it just me or are there no colon and semi-colon in the rockbox keyboard?
07:30:18Mikachuyou can make your own keyboard layout
07:30:28Mikachuput the stuff in a file and call it something.kbd and click on it
07:30:52Mikachui am not saying ; and : shouldn't be in the keyboard
07:30:53Hideo_Kuzeand I can place it anywhere in the .rockbox directory?
07:31:00Mikachui believe so
07:31:38Hideo_Kuzeyeah : and ; should be in there, all sorts of weird chars are in there while : and ; are more common
07:37:46 Join jn [0] (
07:37:55jnis there a way to see free space left in rb
07:38:24Mikachuand then one of the three items there
07:38:51Mikachudid your fsck work?
07:39:48jndidnt free up space tho
07:39:59jnmaybe it really is full and the ipod (apple side) lies
07:40:05 Quit Davide-NYC ("Chatzilla 0.9.74 [Firefox]")
07:40:24Mikachuwhat does du -hs /mnt/ipod tell you then?
07:40:44Mikachuwill take a little while to finish probably
07:40:47jn15G /mnt/ipod
07:40:57jnits a 30 gig ipod
07:41:27Mikachuone course of action would be to move all files off the ipod, mkfs.vfat -F 32 it, and move the files abck
07:43:50jnrockbox claims disk only has 52.1MB free
07:43:54jnsomething's up
07:44:04Mikachuand df -h?
07:44:19Mikachuand fsck says everything is fine now?
07:45:27jnwait nope
07:45:36jndosfsck 2.11, 12 Mar 2005, FAT32, LFN
07:45:36jnFree cluster summary wrong (13345 vs. really 3474790) Auto-correcting.
07:45:36jnLeaving file system unchanged.
07:45:36DBUGEnqueued KICK jn
07:45:36jn/dev/sda2: 4367 files, 3826530/7301320 clusters
07:45:42jnit keeps doing that
07:45:56jni fixed the boot sector and that but it comes back
07:45:58Mikachuah, you may have to say -r
07:46:08Mikachu<jn> Leaving file system unchanged.
07:46:10jni already have
07:46:12jni know
07:46:15jni'll try it again
07:46:24Mikachudosfstools are not my friend
07:46:31 Join cal_rifken [0] (
07:46:37jnPerform changes ? (y/n) y
07:46:37jn/dev/sda2: 4367 files, 3826530/7301320 clusters
07:48:00jnjn@jn-laptop:/$ cp -r /dump/MP3/Jurassic_5-Feedback-2006-ESC/ /mnt/ipod/hiphop
07:48:00jncp: writing `/mnt/ipod/hiphop/15-jurassic_5-canto_de_ossanha.mp3': No space left on device
07:48:17Mikachuyou could try chkdsk if you have a windows computer handy i suppose
07:48:20jnjn@jn-laptop:/$ fsck.vfat -n /dev/sda2 dosfsck 2.11, 12 Mar 2005, FAT32, LFN
07:48:20jnFree cluster summary wrong (0 vs. really 3461445)
07:48:32jni do.
07:48:52jnwhat about doing the mkfs.vfat -F 32
07:48:57jnthink that would do the trick?
07:49:04Mikachui hope you have not been using -n the whole time
07:49:16jni haven't
07:49:35Mikachuthe downside to mkfs is you have to backup all files first
07:49:41jni dont mind
07:49:51jni'll just do that
07:50:36jncp -r /mnt/ipod/ /dump/TEMP/ipod
07:50:46Bg3rmorning ;)
07:50:49jnam i gonna have to extract apple fw and all that again after this?
07:50:59 Join Siku [0] (
07:51:01Mikachuthat stuff all goes in sda1
07:51:03jncool cool
07:51:09jnhope i dont run into this problem again
07:51:29Mikachufat isn't the most reliable file system on earth
07:51:44jnvery true
07:54:02dwihno\o/ wee! morning!
07:58:11 Part safetydan
08:02:03Hideo_Kuzefsck.msdos ruined my partition when I tried it over usb1
08:02:21Hideo_Kuzetoo slow for it apparently
08:03:07scorcheit has worked fine for me...just takes a while =)
08:05:36jnwhat does fsck.msdos do?
08:06:00Hideo_Kuzesame thing fsck.vfat does
08:06:01goffasends email to bill gates telling him to fsck off
08:07:23jni have usb2 :)
08:09:09Hideo_Kuzewell rub it in why don't ya
08:11:29 Quit kidd3ckz (Read error: 110 (Connection timed out))
08:22:19 Quit _Veseliq_ (Read error: 113 (No route to host))
08:22:33 Join _Veseliq_ [0] (n=veseliq@
08:29:54 Quit _Veseliq_ (Remote closed the connection)
08:31:52 Quit jn ("adios")
08:35:31 Join jn [0] (
08:36:19jnhey what was the exact cmd to run with fsck to format the ipod
08:38:05jnis it fsck.vfat -F /dev/sda2 ?
08:38:15jnis it fsck.vfat -F /dev/sda2 32
08:38:16 Join BHSPitLappy [0] (
08:40:03CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
08:40:03*astecp would guess mkfs.vfat -F 32 /dev/sdX
08:40:24astecpbut that's almost entirely different to what you're suggesting
08:40:28astecpand it's only a guess
08:40:41jnlooks right
08:41:24astecpnote /dev/sdX instead of /dev/sdXY ... I'm making the assumption that the ipod like most flash devices doesn't actually want a partition table
08:41:42SlasheriLinusN: The problem with buffer_alloc in tagcache/dircache is that it isn't known how much buffer we really need. So using it requires a dirty syntax: some_buffer = buffer_alloc(0); ...; buffer_alloc(real_size);
08:42:13amiconnastecp: The ipod absolutely needs a partition table, and the first partition, while being of type 0, _must_ not be deleted
08:42:39amiconnIt's the firmware partition, without it the ipod will no longer boot (but can be restored)
08:42:55Bagderthe sansas use partitions as well
08:44:04amiconnThe Ondio also uses parttions (and we need to keep that because of our cfg sector), and _all_ MMCs I've seen are factory formatted using a partition table
08:44:20Bagderso then perhaps "most flash devices" actually use partitions
08:44:52jnthanks astecp
08:46:59 Join muesli|delhi [0] (n=muesli_t@
08:51:39amiconnSlasheri: Are you planning to do something against the "-1. xxxx" in tagcache?
08:53:08*amiconn would also like to see track number guessing removed
08:54:28amiconnI have a some tracks which have no track number, on purpose. Some of them are by "666", and they are evaluated as having track number 66....
08:55:01Slasheriamiconn: yes, that is not because of track number guessing but the new format definition restrictions of the tagnavi.config
08:55:29Slasherithere should be a way to define different syntaxes when tracknum > 0 and when it's not
08:55:40Slasheri-1 just means the track has no track number
08:55:56Slasherihmm, interesting
08:56:00amiconnI know these are 2 different problems.
08:56:37miraklinux finds flash device partitions, but windows often just see the first one
08:57:37Slasheriamiconn: i was even thinking that the entries in tagcache could each be a "rockbox script" that is executed to display the results
08:57:53Slasherithat would give the best customization capabilities
08:57:55amiconnSlasheri: On another matter, undefining HAVE_TC_RAMCACHE for archos saves almost 2KB
08:58:02Bagderthe sansa even has the actual firmware it loads on the second partition
08:58:11Slasheriamiconn: sounds great
08:58:23amiconn(but fired a warning I just fixed)
09:00:54amiconnWe could probably save some more when removing the use_ram parameter completely when HAVE_TC_RAMCACHE isn't defined, in a similar manner as the IF_MV macros in fat.c work
09:01:56 Join mikearthur [0] (
09:02:50amiconnSlasheri: Talking about IF_MV, is there a reason why dircache doesn't work together with multivolume? (Mikachu wanted to know)
09:03:49Slasheriamiconn: hmm, i have never tried it with multivolume and haven't used the IF_MV correctly in the code so that's probably the reason.. Hmm, maybe it could be fixed
09:04:15***Saving seen data "./dancer.seen"
09:04:34amiconnWhen hotplug is also defined, it will get really ugly.
09:04:38Slasherithat's mainly because i don't have any devices with multivolume capability
09:04:50amiconnmultivolume is possible on all targets
09:05:04amiconnJust create 2 (or more) parttions
09:05:14Slasheriah, interesting
09:05:20 Join dan_a [0] (n=dan-mirc@
09:05:37Slasherii thought that was only for the units with different card slots or something like that
09:05:45amiconnHotplug is a different matter. With hotplug, we'll need a way to exclude volumes from dircache
09:06:48 Join petur [0] (
09:07:53amiconnSlasheri: It was invented for devices with multiple "disks" (Ondio internal flash and MMC), but works for multiple partitions on all targets
09:08:59amiconnOn H300, multivolume and hotplug will come into play when usbotg gets implemented
09:09:15amiconnThe hotplugged volumes must not be dircached
09:15:11muesli|delhiwhats the difference between unicode utf-7 16 32 big/little endian etc? what should i use in thunderbird for german characters, does somebody knows?
09:17:03markunmuesli|delhi: I use UTF-8 for international chars in mails
09:17:09Galoisnobody uses utf7
09:17:16Galoisfor the rest, see
09:24:12 Join aliask [0] (
09:25:01amiconnBagder: Do you think it would be possible to make the langv2 scripts check for features?
09:30:48amiconnThe current approach has disadvantages when adding more and more targets, depending on the type of string of course
09:31:16amiconnFor strings which are indeed model dependent (like NAVI for iriver) it works very well,
09:31:43amiconnbut for others, e.g. strings for all swcodec targets, it's cumbersome
09:32:00 Quit Daishi (Remote closed the connection)
09:33:08amiconnSame goes for other features like backlight (to be included for all targets except Ondio, unless the Ondio has the backlight mod), or alarm (to be included for fmrecorder and recorder v2, and recorder with alarm mod)
09:33:49BagderI don't think we need to focus much on the mods versions
09:34:01Bagderthey are extremely rare
09:34:16Bagderand I bet they can live with untranslated strings for those parts
09:34:18amiconnEven when they're rare - missing strings would be very ugly
09:34:33amiconnThey wouldn't be untranslated, they would be empty...
09:34:35Bagderhm, right they would be missing
09:34:55amiconnSome strings even need checking for multiple features (e.g. clipping light needs to check for both recording and backlight)
09:36:01amiconnThe question is how fine grained we want to apply langv2.
09:37:27amiconnWe could say mod versions are rare so we ignore them - but then we'd have to include the strings for possible mods in the standard version, and that again means including unused strings for the masses
09:38:24 Join Mordov|hardlywo [0] (
09:38:24Bagderor we just require the mod-people to run patched versions
09:38:56BagderI think we could perhaps solve this if we give it some thoughts
09:39:45amiconnThat's what mod people do.... but today building a patched version just means adding some EXTRA_DEFINE(s)
09:40:34amiconnThe feature check would be very useful for standard builds as well. Imagine a string needed on all swcodec targets.
09:40:45amiconnWith the current approach, there are 2 options:
09:41:04amiconn*: "swcodec text"
09:41:18 Join lodesi [0] (
09:41:21amiconnplayer, recorder*, fmrecorder, ondio*: ""
09:41:35amiconn*: ""
09:41:59amiconnh100,h120,h300,ipod*,x5: "swcodec text"
09:42:21amiconn...and if you add a target, all those lang entries need updating....
09:44:04 Quit BHSPitLappy (Remote closed the connection)
09:45:58 Join _FireFly_ [0] (
09:49:37 Quit mikearthur ("Konversation terminated!")
09:50:21 Quit dongs (Remote closed the connection)
09:50:24 Join dongs [0] (
09:55:38 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
09:56:08LinusNSlasheri: how about allocating a fairly large buffer when the size is unknown?
09:56:44LinusNsure, it will leave the user with a smaller audio buffer, but that's only the first time after enabling the option, right?
09:57:32amiconnDircache and tagcache buffer size can vary vastly afaik
09:57:41amiconnAnything from a few KB to several MB
10:01:11LinusNbut would it be a huge problem if we allocated 3mbytes when the option is first turned on?
10:01:24LinusNto avoid a reboot
10:01:47 Join scott666_ [0] (
10:02:28amiconnDepends on how much ram the target has
10:02:33muesli|delhiis there already dual boot for iaudio possible?
10:02:38LinusNon another topic, does anyone know why the bookmark files are stored in the parent folder?
10:02:47Bagdermuesli|delhi: there's a patch for that, yes
10:03:03muesli|delhiBagder ok, thx ;)
10:03:18amiconnIirc the maximum dircache size is 6MB; if it's the same for tagcache allocating the maximum will _severely_ hit the audio buffer on e.g. h100 ...
10:03:20LinusNamiconn: sure, but do you store any of the caches in ram on archos?
10:03:59amiconnCvs allows tagcache in RAM for archos
10:04:00LinusNamiconn: yes, it will severly hit the audio buffer, but only the first time you enable the option
10:04:38amiconn...but I'll probably disable this soon because it doesn't make much sense, and disabling saves almost 2KB on binary size
10:05:02LinusNamiconn: sounds fair
10:05:04amiconnOn fmrecorder, we're very near the rombox limit again (a few 100 bytes) :(
10:05:09 Quit dan_a (Read error: 110 (Connection timed out))
10:05:43LinusNi tried to test the voice buffer issue yesterday, and the MS voices were too large :-(
10:06:00LinusNmy buffer was 1.4MB and the voice files 1.5MB
10:06:21amiconnDo you run rockbox from flash?
10:06:37LinusNwell, not rombox
10:06:48amiconnI mean rombox...
10:06:56*amiconn is always running rombox on archos
10:07:07amiconnLinusN: Why not?
10:07:46amiconnRombox gives >200KB extra buffer on my recorder...
10:08:18LinusNso blind people would have to user rombox?
10:08:27 Join BHSPitLappy [0] (
10:08:58amiconnHmm, that'd be a problem
10:09:20LinusNbtw, shouldn't we offer target-specific voice files?
10:09:45amiconnYes, as soon as langv2 is borught a bit further
10:10:18 Quit BHSPitLappy (Client Quit)
10:10:37amiconntalk.c needs some work as well (introducing a target marker and the lang version in the header)
10:13:31LinusNto protect against loading the wrong voice file?
10:14:44amiconnYes, because then (a) voice can be all mixed up and (b) it may not even load at all
10:15:06amiconnI plan to scrap the voice clip bitswap for swcodec targets
10:15:20amiconnIf we do that, voice building should be put in cvs
10:15:54amiconnIt will still require a windows box, unless you can make wine run both SAPI5 and vbscript, which I think might be impossible
10:18:56LinusNi plan to add automatic festival voice builds on the build server
10:20:33 Join dan_a [0] (
10:22:05*amiconn hopes the build server has enough disk capacity to spare
10:22:31amiconnWhat languages are supported by festival (in bearable quality)?
10:22:45amiconnHow many voices are there per language?
10:26:20LinusNhaven't checked closely, but i don't expect much
10:27:05Bagderthe build server has many gigabytes to offer
10:27:19Bagderand if not, we can just add a new drive
10:30:20amiconnHmm, just an estimation: let there be 6 usable voices (languages * voices_per_language). One voice file is 1.4MB. We have 21 targets, and the build server stores the last 30 dailies. That means 6 * 1.4MB * 21 * 30 =~ 5.2GB
10:30:43 Join safetydan [0] (
10:30:57Bagderwell, we don't have to keep 30 days for these if we don't want to
10:31:17Bagderbut 5GB is not a problem
10:31:26Bagder50 might be
10:32:43amiconnBuilding >120 voice files will also take a while
10:33:16amiconnIirc building one voice file (windows, SAPI5) takes 1..2 minutes here
10:33:33*amiconn didn't do this for quite a while now :/
10:36:31 Quit lodesi (Read error: 110 (Connection timed out))
10:36:33 Join lodesi [0] (
10:36:46amiconnYou could apply some trickery by not building voices for certain targets, just symlinking instead. Some voice files will be identical
10:38:04amiconn(a) fmrecorder and recorderv2. (b) h100 and h120. (c) mini1g and mini2g. (d) ipod 3g and 4g (probably)
10:39:27 Join A_M [0] (
10:39:28 Quit A_M (Client Quit)
10:43:54 Quit lodesi ("Parti")
10:44:26 Join lodesi [0] (
11:04:16***Saving seen data "./dancer.seen"
11:04:50 Quit midkay (Read error: 104 (Connection reset by peer))
11:07:18 Join midkay [0] (n=midkay@rockbox/developer/midkay)
11:08:13 Quit _Lucretia_ ("Leaving")
11:18:47 Part Hideo_Kuze
11:26:23 Join Jungti1234 [0] (n=jungti12@
11:29:51aliaskHi Jungti1234
11:29:55markungood day Jungti1234
11:30:06Jungti1234:) hi all
11:31:29 Join zsr [0] (
11:31:39zsris it possible to add a playlist to the tagnavi?
11:31:44zsrso itll run the playlists directly
11:33:46 Quit Bg3r (Read error: 110 (Connection timed out))
11:43:49 Join tvelocity [0] (
11:45:58 Join mantono [0] (
11:49:19 Quit safetydan (Read error: 110 (Connection timed out))
11:50:05 Join safetydan [0] (
11:52:42 Join spiorf [0] (
11:53:01 Quit zsr ("CGI:IRC")
11:57:25 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
11:58:03linuxstbsafetydan: Reading your question from last night - yes, I'ld like to see the current EQ screen adapted to use the 5g's hardware EQ.
11:59:29safetydanlinuxstb, cool. shouldn't be too hard once there's an interface to the hardware eq
11:59:50safetydanjust clamp the available settings to what the hardware can support
12:00:23linuxstbOne possible complication - the hardware supports the following gains: -12db, -1db, 0db, +1db, +2db, ... (1db steps), ..+12db
12:01:07safetydanso... only two negative dB gains?
12:01:43safetydanbut again, easy enough to support
12:01:48LinusNhow silly
12:02:07 Quit amiconn (Nick collision from services.)
12:02:08 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
12:02:09 Join _Lucretia_ [0] (
12:03:07 Quit pike ()
12:03:15 Join pike [0] (i=amiga@
12:03:20linuxstbBut looking again at the datasheet, I may be wrong and it's possible it does support -11db to -2db,...
12:03:53linuxstb(I think they've just forgotten to list them in the table...)
12:04:45linuxstbSo I think we can assume it can go from -12db to +12db in 1db steps.
12:06:12linuxstbYes, I've just found other places in the datasheet where it confirms the EQ can gain from -12dB to +12dB in 1dB steps, so that's fine.
12:06:24*linuxstb will now stop talking to himself.
12:07:15amiconnDoes the hardware eq have selectable frequencies and Q?
12:08:39linuxstbFor bands 1 and 5, it has a selectable cut-off frequency. For bands 2, 3 and 4, it has a selectable centre frequency and two settings "narrow bandwidth" and "wide bandwidth". The frequencies are limited to a choice of 4 for each setting.
12:09:04linuxstbSo no where near as flexible as the software EQ.
12:09:42amiconnNot too shabby. Now if we knew what Q values "narrow bandwidth" and "wide bandwidth" mean...
12:11:32linuxstbThe datasheet has lots of graphs documenting how the EQ works - but I don't really undertand EQs very well.
12:11:41linuxstbIf anyone is interested, the datasheet is here:
12:13:53 Join Bg3r [0] (n=bager@rockbox/developer/Bger)
12:18:02 Join bluebrother [0] (
12:18:46safetydanwell given that data sheet, I could probably code something up now that would follow the limits.
12:19:38linuxstbCool. If you tell me what API you need from the audio driver, it should be very quick to implement.
12:21:43safetydanwell at the moment it just calls functions in dsp.c which then pull the eq settings from global_settings
12:21:45JdGordonLinusN: you around?
12:21:57Jungti1234Does iriverzone exist in the usa?
12:22:39JdGordonhey, "In this particular case, I think we want to reserve the Play button for special stuff, like taking you to the WPS. " <- does that mean that you think the play button should always goto wps? even from menus?
12:22:58linuxstbsafetydan: But dsp.c is in apps/ - firmware/ code (where the audio driver lives) can't access the global_settings struct directly.
12:24:53LinusNJdGordon: isn't that what people want?
12:24:53preglowhah, cool
12:24:57preglowit's got a programmable iir filter
12:25:14linuxstbsafetydan: Seems we'll need to add the EQ settings to sound_settings_table[] in firmware/sound.c - I'm not very familiar with that code though.
12:25:22JdGordonLinusN: im not sure, but if thats workable then lets do it..
12:25:47LinusNbesides, i think it felt odd that Play could be used to select menu items
12:26:15preglowsounds weird indeed
12:26:43LinusNi think part of the consistency is that it is clear what the buttons do
12:26:53amiconnWell, using Play (or the equivalent) for going to the wps from the menu might be convenient, althoug it's a bit against the rockbox concept
12:27:04amiconnAlso, what would Play do if nothing is playing?
12:27:07LinusNif Play can be used to select menu items, you expect Play to select songs as well
12:27:16JdGordonk, also, removing the clear_button_que() call, do you think that may cause any problems on the other targets?
12:27:29LinusNJdGordon: no
12:27:43amiconnJdGordon: Clearing the button queue is generally _bad_ anyway
12:27:45LinusNamiconn: nothing, i guess
12:28:03linuxstbDo some plugins use it though? I seem to recall seeing it.
12:28:24amiconnPlugins may use it occasionally, but the core should never use it imho
12:28:26safetydanlinuxstb, I'm not sure how well this will map as I'm assuming it's easier for you to deal with a request like "set eq band 1 to frequency setting 2" rather than "set eq band 1 to frequency 300Hz"
12:28:28LinusNlinuxstb: yes, especially those who poll the buttons
12:28:35 Quit tvelocity ("Ex-Chat")
12:29:38linuxstbsafetydan: It doesn't make much difference - I could just use a select statement based on the frequency.
12:30:03LinusNJdGordon: will you be around the next hour or so?
12:30:16JdGordonLinusN: ok, so can I add a MENU_QUIT_ALL or something define to force all menus to quit if that comes in from a callback? because atm you can only go up one level
12:30:19linuxstbsafetydan: But it seems more logical for the frequency to be stored as one of a list of four values, rather than in Hz.
12:30:20JdGordonLinusN: ye, probably
12:30:35LinusNgotta grab a bite
12:30:44JdGordonhehe, bon apatite
12:30:49LinusNcu around
12:30:54amiconnI think a button to completely leave the menu makes more sense than a button that goes to wps from menu
12:31:07LinusNamiconn: perhaps
12:31:10*linuxstb agrees with amiconn
12:31:16JdGordonamiconn: i tinhk your right, but it can be the same button...
12:31:46linuxstbIMO, the button to enter the menu should be the same button to "quick-exit" the menu.
12:32:19JdGordonso what do we do with the play button? or allow both to do that?
12:32:54JdGordoni tinhk on most targets the menu button is used in the menu settings to cancel and return to the menu, so that might not work
12:35:16 Quit nudelyn2 ("At Argon, we're working to keep your money.")
12:36:16amiconn(1) On many targets SETTINGS_CANCEL is not the menu button. (2) Even if it is, why would it be a problem?
12:37:18JdGordonye dw, im tired and got confused
12:37:19*amiconn obviously likes numbering statements
12:37:40linuxstb1) I would be happy to change that on the ipods - 2) e.g. PLAY may be a better choice, as 3) it's also mapped to STOP (as a long-press), which is 4) vaguely similar to "cancel"...
12:38:15linuxstbThat would leave MENU purely for entering and leaving the menu.
12:38:46amiconnI have no problem with MENU being cancel on ipod, and still having MENU leave the menu
12:38:56amiconnLeaving the menu is also some kind of cancelling
12:39:28linuxstbThe problem with that is that pressing MENU from within a settings screen won't quick-exit.
12:39:39preglowthe q factors of the peak filters seem to be around 0.5 and 0.25
12:39:49preglowthe latter especially sounds really useless
12:39:52amiconnlinuxstb: No, but then settings are different from the menu anyway
12:40:14linuxstbamiconn: We know that, but does a user see a difference?
12:40:16amiconnI wouldn't expect to leave a setting _and_ the menu with one press - and it wouldn't make much sense
12:40:42amiconnWhy would I enter the setting at all if I then cancel it and leave the whole menu?
12:41:02linuxstbIn which case, I think MENU could do nothing in a settings screen.
12:41:42amiconnImo SETTINGS_CANCEL isn't used often, SETTINGS_ACCEPT is used way more
12:42:05safetydanAre we going to support using the hardware eq and the software eq at the same time?
12:42:07amiconn...and SETTINGS_ACCEPT also leaves the setting
12:42:58linuxstbsafetydan: If you can build an understandable UI for it :) But the idea sounds very confusing to me...
12:43:46amiconnImho this would only make sense when using the hw eq just for treble/bass, but that's also a bit against the rockbox concept
12:44:15safetydanlinuxstb, the only way I can think of to support is another eq ui screen that's the same as the current one but says "hardware eq" instead
12:44:17amiconnRockbox tries to offer the full potential of the hardware when possible (e.g. MDB on archos, etc)
12:46:22linuxstbsafetydan: Yes, that's what I was thinking. There's no reason to force a user to only use one, even if we think it makes no sense.
12:47:38preglowi think both should be supported
12:48:02linuxstbSo two settings - HW EQ on/off, and SW EQ on/off ?
12:48:44safetydanpretty much
12:49:17safetydanthough how does that interact if Rockbox starts using the hw eq for treble/bass?
12:50:12JdGordonso everyone is happy having one button to explicitly cancel a setting change and pretty much any other button will accept the change and return to menu?
12:50:49LinusNJdGordon: did we say that?
12:51:01JdGordondidnt we?
12:52:20JdGordonah i rmemeber, you liked left being cancel and others here wanted it as accept, (or it was the other way waround) ?
12:52:24LinusNi wouldn't expect Rec to leave the setting, for instance
12:52:38LinusNthe other way around
12:52:50LinusNbut i can learn
12:52:59JdGordonor we can add an option...
12:53:06LinusNi think it is important that it is consistent
12:54:10JdGordonin the menu and the settings std_ok is accept and _std_cancel is cancel , as expected, but depending on the target, accept is usually at least 2 buttons...
12:54:25JdGordonso i shuolndt have said any other button,..
12:54:59linuxstbMy proposal for ipods would be: MENU enters/leaves the menu, LEFT/RIGHT navigates up and down the hierarchy, and whilst in a settings screen, SELECT would confirm and PLAY would cancel. i.e. every button has exactly one purpose.
12:55:37JdGordonlinuxstb: L/R naviagtes? not the wheel?
12:55:50 Quit muesli|delhi ("ich will Khe!!!")
12:56:15JdGordonunless heiracy you mean enter/exit menus?
12:56:17linuxstbL/R enters and leaves sub-menus. The wheel would still be used to scroll up and down the list.
12:56:22JdGordonok, sorry
12:56:44safetydanlinuxstb, I think the eq only needs two functions to work, wmcodec_enable_equalizer(bool) and wmcodec_set_eq(int band, int frequency, int bw) where frequency is 0..3 and bw is 0..1
12:57:01JdGordonso.. what about the standard setting screen, does left cancel/accept/do nothing?
12:57:19linuxstbBut I could also be persuaded to carry on using MENU as cancel, leaving PLAY unused (or maybe as a shortcut to WPS).
12:57:21safetydanlinuxstb, let the user interface deal with mapping frequencies on to the available settings in the chip
12:59:26lexwow, replaygain did it :)
13:00:05markunlex: did what?
13:00:19lexi hate when the s-letters whisper
13:00:30lexwhen the volume is up and using my in-ear-monitors
13:00:59*JdGordon doesnt like the define ACTION_STD_WPS, could it be called ACTION_STD_RETURN which would goto wps from tree, or tree/wps from menu depenign where you called the menu from?
13:01:03lexlike "whiSHSSHSHHper"
13:01:04linuxstbsafetydan: OK, let's start with that. Although there's no bandwidth parameter for bands 1 and 5 - just the cut-off frequency and gain.
13:01:17lexor "catCCHHCHHHHH"
13:01:26markunlex: clipping?
13:01:47safetydanlinuxstb, yeah I was hoping the function would just ignore the bw value for band 1 and 5
13:03:47linuxstbThe only thing I'm not sure about is how to implement the enable/disable EQ function. The datasheet says that the DAC needs to be disabled for that setting to be changed. But I'll give it a go.
13:04:22***Saving seen data "./dancer.seen"
13:04:27linuxstbi.e. disabling/enable the HW EQ may require playback to be stopped.
13:04:47lexmarkun: hmm :o
13:05:21 Join Criamos [0] (
13:05:22LinusNJdGordon: ACTION_STD_RETURN is probably not suitable, not if we want the MENU button to return as well
13:05:50LinusNi mean, it might not be possible to have a standard action for it
13:05:56 Quit goffa (Read error: 110 (Connection timed out))
13:06:30JdGordonthen i think it should be called ACTION_TRRE_WPS instead, seen as it does only go [back] to the wps from the tree?
13:06:35 Join actionshrimp [0] (
13:06:44safetydanlinuxstb, does it have any effect if the eq is enable but set to 0db gain across the board? maybe off can just reset things back to 0
13:08:05preglowshould have no effect
13:08:26 Join lightyear [0] (
13:08:39preglowbut there's always the power issue
13:09:27preglowrestarting the dac might just give a minor glitch anyway
13:10:24linuxstbI'll try it, and see what happens. I think I would like to keep the option to completely bypass the EQ hardware, rather than setting it to 0dB.
13:11:13 Join TeaSea [0] (
13:12:34linuxstbAlthough looking at the signal path, it's obviously a digital EQ, so as preglow said, it shouldn't make any difference...
13:13:41 Join Febs [0] (
13:14:04LinusNJdGordon: yes, and then we could add ACTION_MENU_WPS if we want to add the feature to the menu
13:17:04preglowlinuxstb: i sincerely doubt you'll see any codec makers doing analogue filters these days, unless strictly necessary
13:17:19JdGordonLinusN: ok, done
13:19:15preglowlinuxstb, safetydan: as a matter of fact, based on internal precision, an eq with 0 db gain might very well alter the signal a bit
13:21:09safetydanwell I was only thinking of it as an option if the dac restart was problematic
13:22:44linuxstbpreglow: So it wouldn't simply do the equivalent of "if (gain==0) { do nothing }" ?
13:23:04preglowunless they've made a special case out of it, no
13:23:07preglowwhich i doubt they have
13:23:27linuxstbOK, so that reinforces my initial thoughts - disable if possible.
13:23:36JdGordonlinuxstb: menu is cancel, select is accept, play is unused/wps?
13:23:38preglowand anyway, switching from +6dB gain, then to 0dB gain, then just having the codec chip cut off eq processing would give an audible glitch
13:24:22preglowrecursive filters have transients that need to settle each time you change coefficients
13:24:50linuxstbJdGordon: Sounds OK to me.
13:25:22preglowmy guess is that it just more or less what the software eq does. in the case of 0 dB gain, all filter coefficients are still non-zero, but just end up cancling each others effects out
13:26:02safetydanlinuxstb, are you going to expose the "3D Stereo Enhancment" function as well, since that's enabled when the eq is?
13:26:10linuxstbWhy not...
13:26:21linuxstbWhatever that means.
13:26:41linuxstbIs that similar to the "stereo width" setting?
13:27:50safetydanI think so.
13:27:56linuxstbAh yes, "The WM8983 has an advanced digital 3-D enhancement feature which can be used to vary the perceived stereo separation of the left and right channels"
13:28:14safetydanNot that I'd use it if it was there.
13:28:19safetydan(mostly because I don't have an iPod :)
13:28:29Bagderhow do left + right become 3D?
13:28:30linuxstbNo, but as amiconn said, it's nice to expose all the hardware features.
13:28:44linuxstbBagder: Ask their salesmen...
13:30:08 Join Gluc [0] (n=Staff@unaffiliated/gluc)
13:30:10 Part Gluc
13:31:17linuxstbpreglow: Do you know what the "DAC digital output limiter" is all about?
13:33:06preglowBagder: perhaps they've included head-related transfer functions for the entire population1"¤!"
13:33:13 Join TCK [0] (
13:33:35Bagdervery skilled people those guys!
13:33:58preglowour crossfeed also qualifies as crude 3d, in a way
13:34:36*linuxstb wants 5.1 AC3 virtual surround-sound
13:34:50preglowcompletely doable!
13:35:02linuxstbI'll buy you beer.
13:35:42linuxstbpetur: Triggered again?
13:35:49peturyeps :p
13:36:02filRbeer trigger?
13:36:57markunBagder: isn't left + right all we have anyway?
13:37:16linuxstbYou can sense direction though.
13:37:17Bagderbut in my world that's two dimensions
13:37:28*linuxstb shuts up...
13:37:40 Join sando [0] (i=lolsteam@
13:38:14*Bagder curses his board and gets back poking at it
13:38:41markunBagder: is it not possible to hear if something is above or under you?
13:39:00Bagderwith sound only from left and right?
13:39:06FebsIsn't "3D enhancement" usually "we screw with phasing to make your music sound like crap"?
13:39:29peturit is
13:39:32markunyes :)
13:39:45markunI remember it from a very old mod player I had
13:39:48safetydanBagder, it's entirely possible. The shape of your ears plays a big part in determining where sounds come from.
13:40:46safetydanlinuxstb, the DAC output limiter feature almost sounds like replaygain's prevent clipping setting
13:41:24linuxstbIt seems more than that - similar to the dynamic range compression popular with pop music radio stations..
13:42:03*dan_a shudders at the amount of compression radio stations use
13:42:32*petur is still looking for a good one to run on his recordings
13:45:06 Part safetydan ("Ex-Chat")
13:53:44 Join scott666 [0] (
13:53:44 Quit scott666_ (Read error: 104 (Connection reset by peer))
13:59:58 Quit bluebrother ("CGI:IRC 0.5.7 (2005/06/19)")
14:02:33 Part LinusN
14:07:45 Join tucoz [0] (
14:08:34tucozlinuxstb, preglow: Why not have support for both the sw and hw equalizers. But not at the same time. I.e. Equalizer HW/SW/OFF ?
14:08:53linuxstbWhy limit the user?
14:09:27tucozlinuxtb: I wouldn't call that limiting. Who would expect to have two equalizers running at the same time on a dap?
14:09:51linuxstbI could imagine using the HW EQ for some bands, but if you wanted more control, use the SW eq for other bands.
14:10:16tucozThis is of course if the two are nearly identical. Which I don't know if they are.
14:11:36tucozhmm, the processing penalty of running hw and sw is not that big.
14:11:42tucoz(I would assume)
14:12:33tucozgotta go. See you
14:12:36 Part tucoz
14:13:57 Quit JdGordon (Remote closed the connection)
14:16:37preglowBagder: it's completely doable to do real life 3d with only 2 channels
14:16:48preglowwe only have two ears after all
14:17:00preglowit's just very complex and requires plenty of filters which depend on your head
14:17:12preglowwell, not very complex, it just requires a bunch of measurements
14:18:17 Quit BloodySorcerer (Read error: 54 (Connection reset by peer))
14:18:18 Quit Febs (Read error: 110 (Connection timed out))
14:20:52*linuxstb offers to send his head to preglow
14:21:39preglowi'll just relay it to the proper measurement labs
14:21:55linuxstbNot something to do at home then?
14:22:43preglowyou'd pretty much need an anechoic room, some very clever ear pickups and a motorised sound source capable of moving in a sphere around you :-)
14:23:06linuxstbRules out my home at least.
14:23:27amiconnReminds me of artificial head stereophony
14:23:28preglowin the case of your head not making too much of a fuss, i guess we could just move a speaker around over the space of a couple of days
14:23:49amiconnWas somewhat popular in the analog time, and didn't require special measuring equipment
14:23:59preglowamiconn: how does that work?
14:24:53amiconnThe recording was done with microphones mounted into an artificial head
14:25:05linuxstbbinaural recording?
14:25:40amiconnIf your head isn't too different from the average, it works fairly well
14:25:41linuxstbSome live bootleggers like that - they mount the mics next to their own ears, pointing outwards.
14:26:07peturthere are already in-ear mics for that
14:26:16peturlook like headphones but are mics
14:26:22preglowamiconn: that is pretty much exactly what i'm talking about
14:26:26 Join Kohlrabi [0] (
14:26:35preglowamiconn: it just approximates the hrtf to a general average head
14:27:21preglowa dummy head is far better at sitting still than, say, linuxstb is
14:27:26preglowlinuxstb: head-related transfer function
14:27:35*linuxstb just asked google...
14:27:50preglowit's a filter that models the effects of your head and shoulder on the incoming sound
14:28:39peturso he should send his shoulders too
14:28:48preglowideally, he should just send his torso
14:29:04preglowbut i don't know how his lower body would cope with the subsequent boredom
14:29:09peturif he sends his arm and hand too you could both have a beer :)
14:29:12linuxstbSending my head was fine, but I draw the line at my torso...
14:29:53preglowi'm on the phone right now with one of london's finest surgeons
14:30:25preglowbut i'll just hang up then
14:30:27 Join Febs [0] (
14:30:57linuxstbpreglow: When you said "completely doable!", you failed to mention these practicalities...
14:31:20preglowthe exclamation mark clearly expresses irony
14:31:54linuxstbOK, then I'll wander the streets looking for someone with the same physical characteristics to me and send you his body parts.
14:32:14 Quit Febs (Client Quit)
14:32:15preglowthere's bound to be labs doing this kind of measurement, so i just need to code up a convolution engine
14:32:22preglowwith luck, we might have enough cpu left to do mp3s
14:32:38linuxstbIt's OK, I'll buy a gigabeat.
14:33:25 Join Febs [0] (
14:33:50Bagder... it seems Zune won't do playsforsure
14:34:10 Quit TCK (Read error: 104 (Connection reset by peer))
14:34:57preglowi'm practically itching to hack it and rename it playfoshizzle in rockbox
14:35:02 Quit Febs (Client Quit)
14:35:11*linuxstb google Zune
14:35:30*petur googles playfoshizzle
14:35:37aliaskMicrosoft branded mp3 player - doesn't look that promising
14:35:39*linuxstb registeres
14:35:44compubombdoes rockbox play video ?
14:35:58preglownot in the way you probably mean, no
14:36:03preglowit does low tech video on archos units
14:36:09compubombcan you play xvid on rockbox ?
14:36:11Bagderthere are several videos _of_ rockbox though! ;-P
14:36:19markuncompubomb: no
14:36:48markuncompubomb: do you know how to program?
14:36:54linuxstbWhy is there so much fuss over yet another DRM-encumbered music store/player/system/whatever?
14:37:03Bagderbecause its MS
14:37:36dan_aWhen will the first "When will Rockbox run on Zune" posts appear, then?
14:37:38linuxstbWell, as long as CDs continue to be sold, I'm happy.
14:37:44Bagderand its kind of fun they introduce a new thing, not using the one they're already backing
14:37:45compubombmarkun: uhm, i can code php, pretty damn well, but i'm still a newb with ANSI C/C++
14:38:21aliaskg'night all
14:38:22markuncompubomb: you could take a look at this:
14:38:29 Quit aliask ("sleep")
14:38:29markunaliask: good night
14:39:49markunit would be nice if people would shut up about video support after it's implemented, but we will probably only get more questions about it then
14:40:13markun'can I play wmv? how about mov or real video?'
14:40:34Jungti1234I want avi
14:40:42markunor avi..
14:40:46 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
14:40:57linuxstb"avi" covers everything though...
14:41:34markunlinuxstb: I think many people don't know it's a container format
14:42:02linuxstbSo what's usually in a .avi?
14:42:12Bagdera video!
14:42:15*Bagder ducks
14:42:18linuxstband sound!
14:43:55filR95% of the time avi will be xvid or divx with mp3 or ac3. :)
14:44:33peturif you're downloading illegal movies, yes
14:44:36amiconn_As linuxstb said, avi can be just about everything
14:44:57filRpetur: or if you encode your own legal movies.
14:47:37 Quit amiconn (Nick collision from services.)
14:47:41 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
14:48:35Jungti1234Unified codecs...
14:57:08 Join ^jhMikeS^ [0] (
14:58:07 Quit jhMikeS (Read error: 110 (Connection timed out))
15:04:25***Saving seen data "./dancer.seen"
15:04:59 Join Febs [0] (
15:05:09 Quit Jungti1234 ()
15:06:01 Join Jungti1234 [0] (n=jungti12@
15:06:12 Join gursikh [0] (
15:06:31 Part gursikh
15:10:24 Join jhMikeS [0] (
15:10:30 Quit ^jhMikeS^ (Read error: 104 (Connection reset by peer))
15:14:45 Join Paul_The_Nerd [0] (
15:28:53 Join dionoea_work [0] (
15:30:11Paul_The_Nerdlinuxstb: You here?
15:31:00Paul_The_Nerdlinuxstb: With that file, on a Nano, I get glitches at 1:00 and 1:08 exactly as described (a drop in the left channel, right channel is fine)
15:31:47linuxstbWeird. I must be deaf...
15:31:48 Join goffa [0] (n=goffa@
15:32:09peturin that case, no need to send your head :)
15:32:17Paul_The_NerdEither that, or some setting prevents it?
15:32:39linuxstbpetur: It's still useful as a beer receptical.
15:32:46preglowPaul_The_Nerd: could you do an analysis on the file and tell me what order that particular part of the file is encoded in?
15:32:53linuxstbpreglow: It's -8
15:33:02preglowlinuxstb: that doesn't exactly tell me much
15:33:08preglowthat particular part might still be order 4
15:33:21 Join Mahesh [0] (
15:33:35preglowi'm just trying to find out if this is a bug in my lpc asm
15:33:44linuxstbpreglow: I was thinking the same.
15:33:51MaheshHi every body, I am very new to Rockbox
15:34:12*linuxstb tries to work out how many 4608 sample frames there are in a minute of audio...
15:34:19Maheshi am planning to install rockbox today
15:34:34Paul_The_NerdShould I leave it to linuxstb then, or what did you need me to do?
15:34:36Maheshso will it damage my existing data on IPOD?
15:34:50linuxstbPaul_The_Nerd: Yes, I've got it, thanks.
15:35:01Paul_The_NerdMahesh: No.
15:35:04 Join sharpe [0] (
15:35:09MikachuMahesh: you'll still be able to start the original firmware at any time
15:35:24Maheshthanx a lot guys
15:35:28Maheshand one more doubt
15:35:36Maheshcan i copy songs from ipod to PC ?
15:35:57Mikachusongs you put on with itunes is a bit hard, but with rockbox you just copy files back and forth normally
15:36:13Mikachuit would be easier to remove them and recopy them if you still have them on your computer
15:36:44Paul_The_NerdI'm not 100% sure I'll have internet access between now and Sunday (beyond what I get on my phone) so advanced warning that I won't be around to yell at people in the forums. ;)
15:37:16Maheshso if i start rockbox firmware would i be able to access my songs which are previously on my IPOD?
15:37:19Paul_The_NerdWell, "now" being "When I leave the channel in a while"
15:38:00Paul_The_NerdMahesh: Yes and no. iTunes changes the names of the songs, but we have a feature called "Tagcache" that can scan your player for songs, and create a list that you can access with an Album / Artist / Genre menu.
15:38:01dionoea_workMahesh: you'd be able to do that using Tagcache (which works kind of like the ipod firmware)
15:38:47Maheshis it a pugin for rockbox or any other software?
15:39:07Paul_The_NerdIt's part of Rockbox
15:39:12Paul_The_NerdIt's explained further in the manual
15:39:30Maheshso i dont need to worry about my current songs on ipod before installing rockbox
15:39:50petura wise man keeps backups :)
15:40:05linuxstbpreglow: The file seems to contain all orders, up to 12... Unless you want to debug another way, I'll write a quick "flac2wav" plugin we can use to check the output.
15:40:17Maheshyeah i do, but this itunes is not allowing me to copy songs from ipod to PC :(
15:41:03Paul_The_NerdIt's odd that it seems to be just the left channel that does it.
15:42:02Maheshi once used redchair software anapod which created a big mess on my IPOD and i lost all my data
15:42:43Maheshit created a new database which started showing 2 files of a single song on ipod screen
15:44:00peturMahesh: rockbox doesn't mess with the itunes files
15:44:02Paul_The_NerdMahesh: You are aware that this is not software for the PC, but rather an entire replacement firmware.
15:44:06Paul_The_NerdIt does not interact with the iPod's database.
15:44:33dionoea_workexcept that you can just remove all the ipod database crap once you've got rockbox runing :)
15:44:53 Join tvelocity [0] (
15:45:09sharpedionoea_work: : shhh. :-D
15:45:29Maheshthats really great work guys, i wll sure donate for this software
15:45:43dionoea_worki wonder how many people here have an ipod and never used the apple firmware. (i did)
15:46:16sharpei used it, and just have some season shows on it...
15:46:36Paul_The_NerdI used it for Audible format Audiobooks, briefly
15:48:16dan_aI use it, to listen to MP3s without any glitching... (damn my old, low spec iPod!)
15:48:57Maheshcan i get a less size of e-books as it is 1 GB :(
15:49:04 Quit chendo (Remote closed the connection)
15:49:10Maheshmy IPOD is 6GB
15:49:12 Join chendo [0] (
15:49:14 Quit chendo (Read error: 104 (Connection reset by peer))
15:49:20 Join chendo [0] (
15:49:22 Quit chendo (Remote closed the connection)
15:49:25preglowlinuxstb: no uniform order around the place where silence occurs?
15:49:35 Join chendo [0] (
15:49:37 Quit chendo (Remote closed the connection)
15:51:25linuxstbpreglow: Nothing obvious. I've found an old flac2wav plugin I wrote, so I'm just trying that now.
15:52:37preglowlinuxstb: might be faster to just disable the asm
15:53:57 Join rconan [0] (n=richard@
15:54:11linuxstbI'm decoding the file now... 170% realtime without boost, and writing to WAV at the same time.
15:54:33Paul_The_NerdSo, in other words, "FLAC is plenty fast" ;)
15:55:15Jungti1234what is format in web irc option page?
15:55:26filRthat is one of the advantages of flac compared to other lossless formats - simple and cpu unintensive decoding. that is what it says on the homepage anway. :)
15:55:59filR(and shown by several different tests)
15:56:52Jungti1234anyone don't know?
15:57:47 Quit dionoea_work ("CGI:IRC (EOF)")
15:57:55 Join Poka64 [0] (
15:58:53 Join dionoea_work [0] (
16:01:10 Join Naked [0] (
16:01:26 Nick Naked is now known as Hadaka (
16:03:28 Quit JoeBorn ("Leaving")
16:05:42 Quit _Lucretia_ ("Leaving")
16:06:02 Part Mahesh
16:07:13 Quit akaias (Read error: 110 (Connection timed out))
16:07:19 Join webguest34 [0] (
16:07:28webguest34web irc is unicode support?
16:07:47Bagderwe speak english here, without unicode
16:08:00Jungti1234hehe. guest34 is me
16:08:06 Quit webguest34 (Client Quit)
16:08:21Jungti1234what is format in web irc option page?
16:08:31Bagdertry it and you'll see
16:08:38Bagderor read up in the irc client
16:08:51 Join webguest47 [0] (
16:09:14filRunicode: äöüß
16:09:37filRiso latin 1:
16:09:40Jungti1234oops korean
16:09:50Jungti1234can't see :)
16:10:00filRno, just german. :)
16:10:29Jungti1234I seem it by Korean and Chinese character. hehe
16:11:07 Part webguest47
16:11:08filRclient setting i guess. but it shouldn't really matter unless you want to learn german (or korean) via irc.
16:11:43 Join webguest78 [0] (
16:11:48preglowlinuxstb: anything?
16:12:00Jungti1234ah.. format is..... irc window theme?
16:12:09linuxstbpreglow: My flac2wav plugin is acting strange. So I'm now debugging that...
16:13:04 Quit webguest78 (Client Quit)
16:14:45MikachuBagder: some english words have accents :)
16:15:07Jungti1234morning mikachu
16:15:16Mikachuanyong haseyo
16:15:54Jungti1234yea, annyeong hapnida. :)
16:16:27linuxstbMikachu: You mean we've stolen a few French words?
16:17:12filRyou stole french words? watch out or someone will "zidane" you!
16:17:14MikachuJungti1234: what does that mean?
16:17:20 Part tvelocity ("Ex-Chat")
16:17:30Jungti1234annyeong haseyo is hello
16:17:44Jungti1234yea, annyeong hapnida is yes, hi
16:17:48filRannyeong? arrested development? :)
16:18:08Jungti1234or yes, I'm fine.
16:23:38 Join Daishi [0] (
16:26:23 Join webguest78 [0] (
16:30:55 Quit petur ("worrrk")
16:31:09 Quit webguest78 (Client Quit)
16:33:51 Join _Lucretia_ [0] (
16:35:19 Quit Febs ("CGI:IRC (Ping timeout)")
16:37:43 Join ehour [0] (
16:38:09ehour!sen dj-fu
16:38:13ehour.seen dj-fu
16:40:48 Join rp- [0] (
16:41:27 Quit Jungti1234 ()
16:42:53 Join pondlife [0] (
16:47:15 Join chendo [0] (
16:48:26 Part Paul_The_Nerd
16:48:26amiconnehour: Use /msg logbot_ seen <nickname>
16:48:48 Quit _Lucretia_ ("Leaving")
16:50:47 Join _Lucretia_ [0] (
16:52:55***Server message 505: 'logbot_ :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:52:55***Server message 505: 'logbot_ :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:53:32***Server message 505: 'logbot_ :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:53:38***Server message 505: 'logbot_ :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:56:05 Quit rconan (Client Quit)
16:57:11***Server message 505: 'logbot_ :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:57:18***Server message 505: 'logbot_ :Private messages from unregistered users are currently blocked due to spam problems, but you can always message a staffer. Please register! ( )'
16:59:04preglowlinuxstb: like i said, just disabling the asm and seeing if the silence is gone is also clever
17:04:28***Saving seen data "./dancer.seen"
17:05:46 Quit TeaSea ("Leaving")
17:06:44 Join akaias [0] (
17:06:46goffadid anything change with usb on the x5 since july 23rd?
17:06:58goffai can only get it to mount read only now
17:07:33 Join TeaSea [0] (
17:08:02Bagderthat's not controlled by Rockbox
17:08:12Bagderusing linux?
17:08:56 Quit pondlife ("byebye")
17:09:20goffawouldn't think so... but i had to ask
17:09:27 Part scott666
17:09:28Bagderthen check your kernel log, linux disables write-access when it finds file system errors
17:09:30goffaprobably just going to have to do the ms fix
17:09:42 Join webguest15 [0] (
17:10:29 Quit webguest15 (Client Quit)
17:11:23 Join Sinbios [0] (
17:11:55goffanot seeing anything in there
17:12:59 Quit Mordov|hardlywo (Remote closed the connection)
17:14:14goffaoh well.. when i get home i'll reboot and see what happens
17:14:29goffai should try on windows here too
17:16:02goffadamn.. left the subpack at home
17:16:18linuxstbpreglow: That would have been easier.... I can't work on it now, but I'm going to try and fix my flac2wav plugin this evening - I trust md5sum more than my ears...
17:18:19 Join marcoz [0] (
17:19:15 Quit marcoz (Client Quit)
17:20:59 Join ManicSheep [0] (
17:21:33ManicSheephey there people
17:21:52ManicSheepi just wanna know if its posible to load doom in color on a nano?
17:23:04linuxstbYes, Doom runs on the Nano. And yes, it's in colour.
17:24:12ManicSheepwhere do i get it..?
17:24:28ManicSheepi checked the rockbox wiki didnt find n e thing...
17:24:39linuxstbYou didn't find the PluginDoom page?
17:24:50ManicSheepill check again...
17:25:12linuxstbDoom itself is included with Rockbox, but you need to install the WAD files - see that wiki page.
17:25:29ManicSheepok thank u
17:27:43 Quit secleinteer ()
17:31:25 Quit Poka64 ("XChat 2.6.6 -")
17:31:56 Join Poka64 [0] (
17:33:11 Quit akaias (Read error: 104 (Connection reset by peer))
17:33:56 Join akaias [0] (
17:35:27 Join klrspz [0] (n=klrSpz@
17:45:59 Quit ManicSheep ("User pushed the X - because it's Xtra, baby")
17:48:07 Join secleinteer [0] (
17:52:02linuxstbpreglow: I think I can confirm a bug. In the sim, my flac2wav plugin is giving an identical WAV file to the official FLAC decoder (matching md5 checksum). On my ipod, the WAV file sounds OK, but parts of the .wav file differs.
17:55:16amiconnWhat about iriver?
17:56:51linuxstbI haven't tested yet, but am about to.
17:58:25 Quit Criamos (Read error: 110 (Connection timed out))
18:00:49linuxstbpreglow: I think it's the case of LPC order=1 - that appears in 4 frames (and for the left channel only).
18:04:53 Join mordov [0] (
18:08:53 Quit Sinbios ("If the definition of a klutz is someone who doesn't have eyes on their ass, then yes, I suppose I am a klutz.")
18:09:18 Join sucka [0] (
18:10:31 Quit akaias (Read error: 110 (Connection timed out))
18:11:45 Join akaias [0] (
18:14:40linuxstbamiconn: Good news, the iriver decoder is perfect. So the suspect is definitely the lpc_decode_arm function.
18:17:44preglowfor order 1?
18:18:24preglowthat would be a more likely suspect
18:23:01amiconnpreglow: Hmm, maybe the r3 increment is missing for order1?
18:23:19 Join TeaSeaLancs [0] (
18:23:20amiconnorder2 and up use ldmia r3!, ....
18:24:38amiconnorder1 just uses ldr
18:26:23preglowit is incremented at line 211
18:26:29*amiconn is still not really familiar with arm asm
18:26:32 Quit _FireFly_ ("Leaving")
18:27:13 Quit dan_a (Read error: 110 (Connection timed out))
18:27:37amiconnDoes ldmia r3! preincrement or postincrement?
18:27:57preglowia = increment after
18:28:01preglowso post
18:29:27 Join mikearthur [0] (
18:29:42amiconnHmm, looking at the structure of the various loops now. Shouldn't the .loop1 label be placed one line up from where it is?
18:30:20pregloworder 1 does look slightly different from the others to avoid a reload
18:30:27preglowi'm looking at it now, i just need to wake up my brain
18:30:40 Quit actionshrimp (Read error: 110 (Connection timed out))
18:31:32amiconnAlso, I can't see how line 211 increments r3
18:31:33 Quit Poka64 ("brb lol")
18:31:51amiconnThe #4 is just an offset, isn't it?
18:32:07preglow str r2, [r3], #4
18:32:12preglowit's that one, righT?
18:32:27preglowno, offsets are [r3, #4]
18:32:43preglow[r3], #4 means address r3, then postincrement 4
18:32:48preglowunless i'm completely asleep
18:33:59 Join bluebrother [0] (
18:33:59preglowi think i see it now
18:34:24linuxstbShould the asr be on the mul, and not the add?
18:34:45preglowno, i don't
18:34:53preglowmul doesn't take asr, and no, it shouldn't
18:35:42amiconnMeh, arm asm causes me a headache :/
18:36:20 Quit TeaSea (Read error: 110 (Connection timed out))
18:37:16 Join Sinbios [0] (
18:37:35 Join Poka64 [0] (
18:38:08 Join tvelocity [0] (
18:40:46amiconnpreglow: ldr and str [r3], #4 is post increment as well?
18:41:27amiconnIf so, it might be that the str operation writes to the wrong place
18:41:44preglowno, i don't think so
18:42:06amiconnIn order 2 and up, you're reading n bytes with post increment, i.e. r3 points past the data you loaded
18:42:06preglowthe whole thing is just y[n] += coef*y[n - 1], where y[n - 1] is always in r4
18:42:23amiconnAh, no
18:43:06 Quit chendo ("Leaving")
18:43:08preglowin the other orders, i should also do the writing one past the locations i've loaded
18:45:05preglowi seriously can't see what's wrong with this :/
18:45:07*preglow gets coffee
18:45:50 Join dan_a [0] (n=dan-mirc@
18:45:55preglowit's also weird that it yields silence
18:46:20 Quit dionoea_work ("CGI:IRC (EOF)")
18:46:43*amiconn wonders how that order1 loop works with using r4 twice, but without reloading
18:47:10preglowthe result i calculate is the sample i'd be loading the next iteration anyway
18:47:24amiconnYour caluclation result ends up in r2, not r4
18:47:56preglowno wonder this is quiet
18:48:03preglowi'm using the residue instead of the result
18:48:45 Join chendo [0] (
18:50:37preglowusing r4 as first argument in the add and str should fix it, i think
18:53:14amiconnMeh, why is the argument order so weird on arm?
18:53:22preglowi don't think it's so weird
18:53:29preglowit's mathematical order, more or less
18:53:43preglowit's less logical on x86, were you seldom have more than two operands
18:53:57amiconnImho it is. It's reversed from all other asm dialects I know - but not always
18:54:13amiconnThe store instructions use source, dest
18:54:34preglowldr/str always take address last
18:54:42preglowbut yes, it is a special case
18:54:56preglowx86 asm was the first asm i learnt, so i guess i'm kind of biased
18:55:48amiconnI don't really know x86
18:56:03preglowno big loss
18:56:12preglowit's the most horrible language i know
18:56:14amiconnz80 also was dest, source - but that was a looong time ago
18:56:32preglowand 6502, i think
18:56:35preglowhmm, no, it wasn't
18:56:58amiconnI think the coldfire order1 loop can be sped up
18:57:15amiconnAs it is, it does 2 loads and 1 store per iteration
18:58:15preglowlinuxstb: able to test?
18:58:23amiconnadd.l %d2, (%a0) is read-modify-write
18:58:38 Join jaczehack [0] (
18:58:43preglowamiconn: feel free to modify it :>
18:59:06preglowthere are ways to speed up the lpc routines more, i just haven't focused much on it
18:59:09preglowflac being very fast as it is
18:59:53amiconnMaybe I'll even do it sooner or later in order to understand it more
19:00:04*amiconn s thinking about flac on SH1
19:00:25preglowgood luck :>
19:01:14linuxstbpreglow: Just running it now.
19:04:10linuxstbMatching md5sum :)
19:04:29***Saving seen data "./dancer.seen"
19:04:36preglowok, i'll commit it then
19:04:45preglowwith some updated comments for this special case
19:06:07 Quit sucka ("( :: NoNameScript 4.01 :: )")
19:06:40 Join Terinjokes [0] (
19:06:56preglowi kept doing svn diff, wondering why it didn't work!
19:07:16preglowBagder: can you tell me why it didn't work?
19:07:51 Quit mordov (Read error: 104 (Connection reset by peer))
19:08:11Mikachufunction svn() { if test -d .svn; then command svn "$@"; elif test -d CVS; then cvs "$@"; fi}
19:08:53linuxstbWhat do you do for "svn rename" ?
19:09:10preglowwatch shit fail
19:09:16Mikachuget help output for cvs
19:09:38linuxstbI think we need some kind of server-side solution...
19:09:42Mikachui just have a function for svn up/cvs up
19:10:05Mikachuthe one above was mostly a joke though
19:12:21preglowi'm a bit surprised order 1 is used often enough that anyone noticed this
19:12:37amiconnpreglow: Am I right that the coefficient can have max. 15 significant bits?
19:12:49preglowamiconn: that's what i think i can remembert
19:12:57preglowamiconn: i think the sh1 mac instruction will work
19:13:15amiconnsh1 mac is only 16x16->32, which might not be enough
19:13:22preglowi think it will be
19:13:34amiconnThe decoded data can be >16 bits before shifting right by qlevel
19:13:52preglowi did a version of coldfire lpc asm once where i used the mac in 16x16 mode
19:13:55preglowworked just fine
19:14:03preglowthough i didn't test it extensively
19:14:13preglowand then i lost the source
19:14:20amiconnBut it should be possible to implement a 16x32->32 multiplication
19:15:22preglowyou should instead focus your efforts on hoping you won't need one :)
19:21:31amiconnI think we'd want 16bit output on SH1
19:23:22 Join einhirn [0] (
19:23:22 Quit einhirn (Client Quit)
19:37:53preglowi don't get this apparent need to display every piece of info in the wps
19:38:16 Join petur [0] (n=petur@rockbox/developer/petur)
19:38:22Terinjokesis there any pages on the wiki on how to make a theme, i didn't see one
19:38:56sharpepreglow: eh, it's for comfort i'd imagine.
19:40:40sharpei mean, are you secure in yourself not knowing what format the currently playing song is encoded in? along with it's bitrate? eh?
19:41:29 Quit dan_a (Read error: 110 (Connection timed out))
19:41:32preglowhell no, i even start to question my sexuality in cases like that
19:41:34Mikachui think we need a wps tag for the phase of the moon too, with 10 different possible bitmaps
19:41:35preglowi also forget my name
19:42:16Terinjokescan we get clouds too?
19:42:19sharpei'm just not myself until i know how large the current song is...
19:42:50sharpeTerinjokes: not unless you have some way of getting cloud cover information, moon phase can be calculated. :-D
19:43:34Terinjokesi'm still wondering if there is a page on the wiki on how to make a theme
19:44:04petursharpe: for targets with RTC
19:44:32sharpebut for the other ones, we can pick a random phase.
19:44:39preglowSlasheri: doesn't buffer_alloc automatically align by 4 now?
19:45:04sharpeand then just blame it on poor syncronization. and then just keep taking it further and further until they have no idea what we're talking about.
19:45:13*amiconn wonders why user-selected values need to be displayed in wps at all
19:45:30sharpeamiconn: what values...?
19:45:32Slasheripreglow: yep
19:45:33preglowamiconn: no idea
19:45:55amiconnI mean, why display whether crossfade is enabled? I know what features I have enabled...
19:46:01preglowSlasheri: amen
19:46:28Mikachudo you think repeat mode should also not be displayed?
19:46:46sharpewouldn't it be pretty evident if it was on? :)
19:47:11Mikachunot if you have 20 more songs in your playlist and want to know if it will turn off or drain the battery after you fall asleep
19:47:47amiconnRepeat mode is probably an exception, as is volume
19:48:00preglowSlasheri: looks like you do it manually a place in dircache.c still
19:48:35amiconnRepeat mode is not easily noticeable (before it actually repeats), and volume can be set directly from within the wps
19:48:47Slasheripreglow: yes, i do (and in dircache also). Replacing that needs more designing as we would have to guess some default size
19:48:58SlasheriAnd freeing once buffered data wouldn't be possible
19:49:02Mikachu- * Copyright (C) 2002 Bjrn Stenberg
19:49:03Mikachu+ * Copyright (C) 2002 Bj�n Stenberg
19:49:08amiconnTaking it to the extremes, we would also have tags for treble/bass/balance, EQ, etc etc
19:49:11Mikachuthat hunk doesn't look so good
19:49:23 Join obo [0] (
19:49:29SlasheriThat way there would be no more any delays on boot up, but audiobuffer would remain reduced in size when initial builds happen
19:49:33Mikachu1.57 in gwps-common.c
19:49:48peturMmmm <−−-
19:50:30Slasheripreglow: anyway, i am now thinking if i should start designing the playback buffer api. That way most of the dynamic allocation problems would be solved and playback engine hopefully would become much more stable after it has been ported to use the buffering engine api
19:51:41Terinjokesif i set album art in itunes, and copy the itunes tracks to my rockpod, well it display the album art
19:52:03preglowSlasheri: sounds good to me
19:52:28Terinjokesso, how does rockbox display the album art?
19:52:45Mikachuit doesn't
19:52:51sharpetechnically, rockbox doesn't display album art. a patch is needed before it has the functionality to.
19:53:18*Terinjokes meant with that patch
19:53:59sharpeas far as i know, it reads an image in the current directory.
19:54:00amiconnSlasheri: Will it also make the playback engine more compact and easier understandable?
19:54:31preglowonly lobotomy will do that
19:54:41Slasheriamiconn: should be, at least all buffer manipulations and pointer handlings would be handled with the buffering api
19:55:16Slasheriso playback engine would just need to store and retrieve data from the buffering engine
19:56:00sharpewoo... no longer real-time encoding...
19:56:19sharpeer. transcoding.
19:57:01Slasheriamiconn: so no more ring buffers or sturctures in the playback engine itself (even metadata could be handled by the buffering engine)
19:58:53 Quit ehour ()
19:59:53mirakas the rockbox api changed a lot recently for plugins ?
20:00:36mirakSlasheri: you want to memory handling ?
20:01:45amiconnSounds rather complex instead of a simplification...
20:02:09 Join joe2 [0] (
20:02:22Slasheriamiconn: that should simplify the playback engine itself very much and then there would be defined api how buffering happens
20:02:46Slasherithen buffering could be either static or dynamic, no matter how the playback engine works
20:03:33 Quit jn (Read error: 110 (Connection timed out))
20:04:12 Quit YouCeyE (Read error: 60 (Operation timed out))
20:05:03*Bagder detects //-comments
20:05:16Slasheriamiconn: in future even archos could use the same buffering code.. And there could be ifdefs for selecting whether to support static or dynamic buffering
20:05:43SlasheriBagder: hehe, those are my special comments for temporary debugging code, that probably needs to be removed in future
20:06:03SlasheriBagder: so it's intended those //-comments would catch eye
20:06:14BagderI would suggest you instead make a fat_logf() function that is ifdef'ed at the top
20:06:25Slasherihmm, that sounds better
20:06:45Bagderoften used approach in the linux kernel
20:06:49Slasheribut without commenting out those lines i found out it would be impossible to use the player at all :D
20:07:56amiconnDo you run a logf build for normal use??
20:08:17preglownot silly at all
20:08:18Slasheriyes, always
20:08:27preglowhelps debugging by heaps for those obscure bugs
20:08:34Slasherithat way it's easy to find out when something goes wrong
20:09:09Slasherifor example those rare playback engine desyncs.. when it happens, i just run logfdump and analyze the results later
20:09:32*amiconn never runs logf builds for extended times
20:10:55*sharpe has never run a logf build.
20:12:55amiconnI only used logf() 2 times for my own debugging so far. One was archos recording (logf didn't help me finding the problem), the other was iriver recording (logf was helpful but caused instability)
20:13:44Slasherihmm, interesting.. i have used logf almost always to find a bug
20:14:06Slasherionly exceptions are crash bugs when it's possible to reproduce on simulator
20:14:07preglowwell, the playback code is very well logfed
20:14:09preglowso it's not very strange
20:14:18Slasheriand tagcachce code as well :)
20:14:24preglowdefinitely a good thing
20:17:49 Join _Veseliq_ [0] (n=veseliq@
20:19:49 Join freqmod [0] (
20:24:17 Quit filR ()
20:25:58mirakmarkun: hi
20:26:10markunhi mirak
20:26:20markundid you play around with your video plugins?
20:26:22mirakdoes any dev have manifested an interest for video ?
20:26:24mirakmarkun: no
20:26:36mirakactually after a day of work I really don't want to code
20:26:37markunlinuxstb has
20:26:52mirakin fact I can't code, that's physically impossible
20:26:58markunand after the gigabeat port I also want to look into video playback
20:27:00mirakand mentaly
20:27:38 Quit tvelocity ("Ex-Chat")
20:27:39mirakmost of the work would be optimisation
20:27:45mirakas usual
20:28:03mirakI tried to have something non optimised that would work so people could try to improve it
20:28:26mirakmaybe including a codec or plugin into cvs could help
20:28:28*preglow off
20:28:45mirakI mean interest some people to try to improve
20:29:11 Join filR [0] (
20:29:33mirakmarkun: the problem is I think the codecs would not compile now
20:30:06markunyou can checkout an old version of rockbox, apply the patch and then cvs update
20:30:46 Join falconfox [0] (
20:31:06falconfoxis there any way to change the speed of the scroll-wheel for the color ipod
20:31:22 Join hannesd [0] (
20:33:04mirakmarkun: I think I could never do something that look finished enough to be included in cvs tree
20:37:20 Quit _Veseliq_ (Read error: 104 (Connection reset by peer))
20:37:44 Join pixelma [0] (i=pixelma@
20:39:04 Quit spiorf (Remote closed the connection)
20:39:36 Join _Veseliq_ [0] (n=veseliq@
20:45:46 Join tvelocity [0] (
20:52:49 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
20:58:03 Join bluey- [0] (
20:58:55 Join Kohlrabi [0] (
21:04:30***Saving seen data "./dancer.seen"
21:24:53*petur swears a bit at real life
21:25:32Terinjokesanyone know how i can disable itunes?
21:25:33dionoeaas long as you don't swear on IRC ... :)
21:25:47dionoeaTerinjokes: uninstalling it might work :p
21:25:59linuxstbThat's probably the only way.
21:26:10Terinjokesevery time i plug the rockpod to add songs, itunes opens up and tells me it can't read the ipod
21:26:58 Join rconan [0] (n=richard@
21:27:54oboyou can try stopping the ipod service ("ipod helper service" or something similar)
21:28:19Terinjokeson a mac
21:28:36obohmmm, maybe not then :)
21:29:30dionoeaTerinjokes: if it's a mac intel you can install bootcamp, install windows and then do as obo just suggested :D
21:29:36dionoeathat might be a bit overkill though
21:30:00 Join nudelyn [0] (
21:30:04Terinjokesok, found an "iTunesHelper" loading at login, thats probably it
21:31:10 Join hannesd_ [0] (
21:31:42 Join BigMac [0] (n=dummeska@
21:31:53Terinjokesok, i kill the process
21:32:16Terinjokesgrr, itunes opened
21:34:02Terinjokesok, kill another ituens process, let me see
21:34:36markunpetur: trouble?
21:35:11peturnah... just time leaking away :/
21:35:55filRyou can set, in the itunes options, that itunes should not be started when connecting an ipod. (apple+, -> ipod -> start itunes when this ipod is connected)
21:42:13Terinjokesyes, but that tab is disabled because itunes doesn't recognise my ipod
21:43:54 Quit lostnihilist ("Leaving")
21:44:10 Join TeaSea [0] (
21:45:17 Quit bluey- (Read error: 110 (Connection timed out))
21:49:15Terinjokesits not funy
21:51:01 Quit rconan (Client Quit)
21:53:41Terinjokeshow can i tell what was changed in daily builds
21:55:38BigMacbrb you guys
21:56:01 Quit TeaSeaLancs (Read error: 110 (Connection timed out))
21:57:52 Join lds_ [0] (
21:57:58 Quit TeaSea ("Leaving")
21:58:40 Quit hannesd (Read error: 131 (Connection reset by peer))
21:58:56 Quit hannesd_ (Read error: 54 (Connection reset by peer))
21:59:03 Quit lodesi (Read error: 110 (Connection timed out))
21:59:27 Join leftright [0] (
21:59:42 Join hannesd [0] (
22:01:19leftrightwith the replaygain tags off|track|album|shuffle surely it isnt a good idea to have another 'shuffle' indicator, wouldn't it be more appropiate to rename that to tack mode, for that is what it is, track mode if shuffled
22:01:52 Join hannesd_ [0] (
22:04:50 Quit leftright (" HydraIRC -> <- IRC with a difference")
22:05:06Terinjokesi feel stupid, just can't find rockboy
22:05:57bluebrotherTerinjokes, it's a viewer. You need to "play" a rom.
22:06:38Terinjokesyes, i know what it is, just i need to install it, the "viewer" isn't installed on my device
22:07:00bluebrotherhow do you know that? Is the rom file listed?
22:08:51 Join _Lucretia [0] (
22:10:21Terinjokesok, i just set view files to all, and clicked on MARIO.GB and nothing happened
22:11:03 Join TeaSea [0] (
22:11:57 Quit Poka64 ("XChat 2.6.6 -")
22:12:00 Quit freqmod (Remote closed the connection)
22:12:04bluebrothercheck for /.rockbox/viewers/rockboy.rock
22:12:20 Join dan_a [0] (n=dan-mirc@
22:12:28bluebrotherif this file isn't present then your installation is missing it.
22:12:29 Quit _Lucretia_ (Connection timed out)
22:12:58Terinjokesnope, not listed, where can i download it?
22:13:10bluebrotherjust reinstall rockbox.
22:13:20bluebrotherit's part of the daily / cvs builds.
22:17:40 Join lostnihilist [0] (
22:19:26Terinjokesok, i just did that, nothing
22:20:05linuxstbWhat ipod do you have?
22:20:07peturwhat target?
22:20:35peturlet's bet rockboy isn't supported on it ;)
22:20:58Terinjokeshow much you betting?
22:21:12petur10 bytes
22:21:25amiconnNo leffe? ;)
22:21:40*Terinjokes sends petur 10 bytes
22:21:55peturamiconn: not anymore - boycotting InBev
22:22:32Terinjokesso, ah, i see rockboy is video only...mean people
22:22:41amiconnNo it's not
22:22:42 Join Mahesh [0] (
22:23:28linuxstbTerinjokes: There's only a small number of players Rockboy doesn't work on - simply because no-one has adapted the display code for the LCD yet.
22:23:39MaheshHI all i just installed rockbox
22:23:44linuxstbI think the Nano is one of them.
22:23:48Terinjokeshey mahesh!
22:23:54Maheshworked well for a while, and
22:23:58Maheshhi teri
22:24:06Maheshmy ipod mini is hanged
22:24:08Terinjokesyes, i might have to borrow my friend's video to play a little bit of mario...
22:24:09Maheshits not responding
22:24:20Terinjokesmy name is not Teri!
22:24:21linuxstbcut it down then.
22:24:32Maheshany body help me
22:24:37Maheshits not even shutting down
22:24:40Terinjokesis it hanging by the headphones?
22:24:41Maheshit is hung
22:24:46Maheshnot responding for any key
22:24:51Terinjokeshave you tried holding menu + select
22:24:58Terinjokesfor like 5 seconds?
22:25:04Maheshi selected a mp3 file and it is hung
22:25:18Terinjokestry holding menu+select
22:25:23Maheshi got it rebooted
22:25:32Maheshthank god
22:25:50TerinjokesMahesh, a hammer would have turned it off ;-)
22:25:56Maheshso for force rebooting what to do? is it menu + select?
22:26:26Maheshthanx teri
22:26:34TerinjokesMahesh, its terin
22:26:44Maheshterin :)
22:27:55Maheshdeleting a folder is taking a long time and it hangs some times, any suggestions
22:28:43Terinjokesdeleting a folder on the ipod from the computer?
22:29:05Maheshnope, using file explorer in rockbox
22:29:21linuxstbAre you playing music at the same time?
22:30:22Terinjokesyou can delete a folder from withen rockbox?
22:31:23Maheshi didnt get you
22:31:30linuxstbMahesh: That will probably help explain the slowness. Rockbox doesn't have much spare CPU time when decoding music at the moment. Are you 100% sure it hangs, rather than just being extremely slow?
22:32:03amiconnDeleting a dir is always slow and can even appear to hang, on all targets, and also without playing music
22:32:17amiconnI don't know why that is, though
22:32:26Maheshsee i have a folder of size 1GB i tried delteing that folder using rockbox file explorer
22:32:36linuxstbNot a good idea...
22:32:45Maheshthen it says folder movies is deleted and not responds for a while
22:32:50amiconnIt depends on the number of files and subdirs, not the size of the folder
22:33:03Maheshok thanx a lot guys
22:33:33Maheshi started liking rockbox a lot, really amazing
22:33:56*linuxstb had to reformat his hard drive after attempting to delete the 32,000 file HVSC on his H140...
22:33:57 Part jaczehack
22:34:27linuxstbThe battery ran out, and there was no way to abort it. So the disk ended up corrupted.
22:35:07linuxstbReverse the order of my first sentence - I couldn't abort, and then the battery ran out...
22:35:25amiconnPlugging the charger would have been an option..
22:35:48 Quit lostnihilist ("Leaving")
22:36:03Maheshguys, another doubt
22:36:18linuxstbI didn't have it with me. But it wasn't a problem - I have all my music on my computer anyway, so just recopied.
22:36:28*amiconn is curious how long deleting a 32000-file folder would take
22:36:29Maheshcan i copy itunes files as normal files to ipod root folder and can browse?
22:36:52*amiconn should perform a series of tests
22:36:56linuxstbI think it took about 80 minutes before the battery died.
22:37:17amiconnI'd like to know why it takes so extremely long
22:37:29linuxstbAlthough thinking about it, it may have been an ipod, rather than my h140. Not that it would make much difference.
22:37:42 Nick Terinjokes is now known as BugsBunny (
22:37:50Mikachumaybe you update the whole fat on disk after every file
22:38:09amiconnSurely not
22:38:24amiconnThe FAT is way too large to be cached as a whole
22:38:25Mikachui know you can make linux do that by mounting with -o sync, and it is indeed very slow
22:38:38amiconnOnly a few sectors are cached
22:39:05 Nick BugsBunny is now known as Daffy (
22:39:18 Join freqmod [0] (
22:39:57 Quit akaias (Read error: 110 (Connection timed out))
22:46:55 Nick Daffy is now known as SeanAUS (
22:48:15 Part BigMac
22:49:48 Join akaias [0] (
22:52:45 Join _FireFly_ [0] (
22:54:31 Quit mirak (Connection timed out)
22:55:55 Quit TeaSea ("Leaving")
22:59:36SeanAUS::Spader Out::
23:01:35 Quit SeanAUS ()
23:01:58 Quit _FireFly_ ("Leaving")
23:04:33***Saving seen data "./dancer.seen"
23:09:51 Join Vasu [0] (
23:10:12 Quit rp- ("change means nothing, when nothing wants to change")
23:10:13Vasuhi all
23:10:24Vasu I have an urgent question
23:11:05Vasuhow can I boot the original apple firmware on an Ipod video (5g) after installing rockbox?
23:11:09Maheshyes vasu
23:11:27Maheshyou can use bootloader or uninstall rockbox
23:12:07Bagderthe rockbox bootloader can load apple OS too
23:12:56Vasuwhere to get it?
23:12:56Bagderits just a matter of pressing the right key at start
23:13:04Vasuthat is?
23:13:11Bagderyou already have rockbox installed, right?
23:13:22Bagderso you have it installed
23:13:32Vasuand how to use it?
23:13:39BagderI don't know what key it is
23:13:41Vasuthere's nothing about it in the manual
23:13:56Mikachubefore you see the apple
23:14:09Vasuthere istn such a menue
23:14:15Mikachuthe button MENU
23:14:48 Join nytroe [0] (
23:15:26Vasuthanks alot
23:15:38Vasuis there any colourvideo player out by now?
23:15:47linuxstbI believe Apple have one...
23:16:01Bagdercolourvideo player?
23:16:16Vasuyeah I know theres one in black and white for rockbox
23:16:22Nibblernew ipod, it roxx
23:16:25*amiconn feels somewhat silly for committing several single-line changes in sequence :/
23:16:25Nibblerfor video
23:16:33Vasuit is a new ipod
23:16:48*petur checks the sillyness
23:16:49Vasubut the its limited to the apple file version
23:16:55Vasuwhen it comes to videos
23:16:59Vasubut thanks alot
23:17:05Vasunow Ill isntall counterstrike on it hehe
23:23:32Bagdermips assembler is weirder than most
23:23:55Bagderjust in case anyone was wondering ;-)
23:26:45*amiconn wonders how any asm can be weirder than x86 - judging from the little bit I've seen
23:27:31Galoishave you tried IA-64 ???
23:27:32Bagderto tell the truth, I've not really used x86 asm since the 80s
23:27:56Bagderbut I'm having a mips right here and it is weird
23:28:37amiconnGalois: Nope. In fact I don't know too many asm dialects
23:29:53amiconnJust SH1, m68k, z80 and very little x86 (reading only, never wrote sth myself)
23:29:57Galoiswell, I think IA-64 is the most warped assembly language
23:29:59amiconnAnd I need to learn arm
23:31:54 Quit mikearthur (Read error: 54 (Connection reset by peer))
23:34:29 Quit lds_ (Read error: 110 (Connection timed out))
23:35:22 Join lds_ [0] (
23:36:48 Join mikearthur [0] (
23:40:49dionoeado bitmaps need to have a multiple of 8 width ? (bitmaps converted at compile time)
23:41:46Vasudoes anyone have experience with counterstrike through doom?
23:42:06Bagdercounterstrike is not doom-based
23:42:16Bagderdionoea: I don't think so
23:42:43dionoeaBagder: yeah ... i had the stride wrong when using the part bitmap thing which messed it up :)
23:43:06linuxstbIs there any way to undo the chaos caused by a "cvs update" with conflicts?
23:43:25 Quit hannesd_ (Read error: 60 (Operation timed out))
23:43:25dionoeayou should have .mine files
23:43:33dionoeawith your old version
23:43:41Bagder... you you use svn
23:43:43Bagderif you
23:43:53Mikachuhow would you do it with svn?
23:44:37linuxstbhow would you do it with cvs?
23:44:39dionoeagrep "<<<<" and fix what's wrong :)
23:44:48Bagderlinuxstb: I don't think you can do it without manual editing
23:45:03Bagderyou can of course script it
23:45:03Mikachuthere is surely some editors that let you jump to the conflict points easily
23:45:10Mikachu+ grammar
23:45:22dionoeatortoisesvn on windows does that
23:45:26dionoeai don't know on linux
23:45:33dionoeamaybe you can tweak vim/emacs to do that
23:45:33Bagder<<<< [keep this] ==== [cut this] <<<<<
23:46:08Mikachuthe second <<<<< is >>>>>
23:46:19linuxstbBagder: There is nothing where you've typed [keep this]
23:46:36Mikachuthe first part is what you had before, and the second part is what's in cvs
23:55:47sharpeit's very humid outside...
23:56:00linuxstbThe British summer has returned - cold and wet :)
23:56:10*linuxstb is happy again
23:56:17sharpei would prefer that over hot and humid.
23:56:42 Quit lightyear (Remote closed the connection)
23:57:09sharpethe funny thing is, it's apparently so dangerous here for hot and humid weather there has been a warning issued.
23:57:13*petur joins linuxstb
23:57:13 Quit akaias (Read error: 110 (Connection timed out))
23:58:33 Quit Mahesh ("CGI:IRC")

Previous day | Next day