Previous day | Jump to hour: 01 02 03 04 05 06 07 08 09 10 11 12 13 14 15 16 17 18 19 20 21 22 23 | Next day

Seconds: Show Hide | Joins: Show Hide | View raw
Font: Serif Sans-Serif Monospace | Size: Small Medium Large

Click in the nick column to highlight everything a person has said.
The Logo icon identifies that the person is a core developer (has commit access).

#rockbox log for 2006-01-21

00:00:47FebsIn any event, let me see what I can come up with. I am going to be doing a lot of traveling over the next few weeks/months, so I'll have time at night to kill.
00:01:10FebsThere's no harm in me taking a look at the issue and putting out a proposal for comment.
00:01:20 Quit webguest63 (Client Quit)
00:02:32FebsOr maybe I'll just create my own "H300 Optimized" keymapping and post it only on Mistic River.
00:02:34*Febs ducks.
00:02:54preglowwill it be blinking and have fancy colours?
00:02:57preglowplease, pretty please :/
00:03:07FebsOf course. I'll install LEDs in the buttons!
00:03:08BHSPitLappy2a keymapping?
00:03:18peturrandom mapping at startup!
00:03:37 Join DocterD [0] (
00:07:29 Join webguest65 [0] (
00:07:32b0brlostlogic: seemlessness is unpleasant issue, but when trying to A-B wav the result is very inaccurate and the loop is usualy much shorter..
00:09:20LinusNamiconn: so you want archos recorder to be able to load color bmp's?
00:09:28preglowthat's because the reported position is inaccurate, i'd believe
00:09:30preglowdue to buffering
00:09:49lostlogichmm... yes, I may need to tweak where the markers are set as well.
00:10:01lostlogicb0br: the problem is worth with wav than other codecs?
00:10:49amiconnLinusN: Yes. Of course, the format parameter will be dontcare, as native format == mono format
00:16:20b0brlostlogic: I've tested only wave so far, but now I tried mp3 and it's the same (second mark is set wrong and still changing)..
00:16:32 Quit akaidiot (Read error: 110 (Connection timed out))
00:16:42lostlogicseems to me like the first mark is set wrong, not the second :-\
00:17:51preglowamiconn: btw, to fix that long_call attribute problem i had, i would have to move all the ICODE_ATTR stuff to the header files, yes=
00:18:10b0brlostlogic: :) may be, but the end is cut (about 2 sec)
00:18:11 Quit Thus0 (Read error: 104 (Connection reset by peer))
00:18:11 Quit DocterD ("Leaving")
00:18:18amiconnI think so, but I don't know for sure
00:18:19 Join Thus0 [0] (
00:19:31 Quit XavierGr ("Trillian (")
00:20:35lostlogicb0br: right, the end is cut so the B marker is broken...?
00:21:17preglowamiconn: 'cause we really need to get rid of this bloody long-call problem
00:21:21preglowit's turning up everywhere
00:22:24 Quit linuxstb ("Leaving")
00:22:27 Quit Kaggen ("Download Gaim:")
00:24:19b0brI don't know, you are the programmer.. I gave up C five years ago, because it was too complicate for me :)
00:25:01 Join dave___ [0] (
00:25:07 Nick dave___ is now known as linuxstb (
00:25:08preglowit's dave___!
00:28:18preglowdid anyone try to fix that dircache alignment bug, btw?
00:31:52 Quit ender` (" I was reading the dictionary. I thought it was a poem about everything.")
00:35:08 Part b0br
00:35:25preglowlinuxstb: new button driver seems to be heaps better
00:35:29preglowi can press a button at any time now
00:42:08eEyeThe iFP port has been commited to Rockbox CVS and new versions will be available only there.
00:42:18eEyethat is one Tomek's homepage, how do i access that now?
00:44:36stevebi hate rockbox.
00:45:11*petur takes large hammer
00:45:12stevebever since i found the sudoku and other game plugins i have been paying less and less attention in lectures
00:45:31stevebits evil i tell you!
00:45:37*muesli__ nodes
00:46:06*petur hands large hammer to steveb and points at his DAP
00:46:20stevebi love it too much :(
00:46:34peturrockbox addiction... hehe
00:46:55stevebsomeone should start a support group
00:47:55preglowlinuxstb: it's somewhat better, at least
00:47:57 Join Kyl3 [0] (
00:48:04preglowit's still got some crazy ass glitches that's hard to explain
00:48:05lostlogicSlasheri: why does codec_seek_complete_callback call pcmbuf_play_stop()?
00:48:07Kyl3Hey is there anyway to have the H300 boot Rockbox when you charge it? I hate having to set my clock every time
00:48:27stevebKyl3: ?
00:48:30lostlogicKyl3: probably next bootloader version.
00:48:47Kyl3oh ok
00:48:52preglowsteveb: bejeweled is particularily dangerous to me
00:48:54steveboh. i think i have only plugged the charger in while rockbox is running
00:49:20stevebpreglow: sudoku has got me. i was addicted to it before rockbox but limited by only having paper copies.
00:49:36preglowi've never even tried sudoku
00:49:37stevebthe only annoying thing is that i cant do any of the ones i have on my iRiver... they are all too difficult
00:49:50stevebalso the blocks game is good
00:49:57Kyl3paper sudoku usually has numbers filled in right?
00:51:52stevebthe rockbox version does as well
00:51:58stevebyou just have to find them on the net first ;)
00:52:21stevebif you run it from the plugin browser it allows you to create new games (ie, from a paper version)
00:52:24Kyl3im bad at sudoku.. idk how to play it
00:52:32stevebits not that hard
00:53:36Kyl3whats that thing on the side for
00:56:10preglowlinuxstb: would you like to test a bootloader using this new code?
00:56:56 Join ModernExecutive [0] (
00:57:25 Join Jungti1234 [0] (n=jungti12@
00:58:54***Saving seen data "./dancer.seen"
01:00:22Rob2222Hi ModernExecutive ;)
01:00:26ModernExecutivehi rob
01:00:41Rob2222How much players now? 5 :)
01:00:56ModernExecutivestill just 1. ups didn't make a drop today
01:02:35Rob2222hehe ;)
01:02:57Rob2222LinusN: Have you took a look at the US FW again?
01:03:07preglow i love code that just wont behave
01:05:51preglowSlasheri: playback still isn't stopped when i load an unbuffered track
01:07:57BHSPitMonkeylinuxstb: hey-lo?
01:10:48 Quit Moos ("Glory to Rockbox")
01:20:33 Join YouCeyE [0] (
01:20:37 Join ashridah [0] (
01:21:53*preglow out
01:23:24 Quit San||Away ()
01:25:32 Quit Tejik (Read error: 110 (Connection timed out))
01:26:18 Quit ModernExecutive (Excess Flood)
01:26:57Rob2222lol ^^
01:29:05 Join ModernExecutive [0] (
01:31:23Rob2222^^ <= twitch eyebrows
01:31:47Jungti1234I know. :)
01:31:51Rob2222ah ok
01:32:46Rob2222Jungti1234: The display that changes color when looking from different angles up and down, is that the new or the old`
01:33:50Jungti1234ah yes
01:34:03Rob2222i got the 2 pics of the 2 different displays
01:34:16Rob2222but i couldnt see, how to recognize
01:34:16Jungti1234Send me.
01:34:34Rob2222no, i mean i saw the 2 pics from you.
01:34:46Rob2222my girlfriend killed my new 280 euro digi cam :(
01:34:54Rob2222cant take a shoot
01:35:31Rob2222At your display, does the color sightly chanes when lookinf fom up and down?
01:35:36 Nick BHSPitLappy2 is now known as BHSPitLappy (
01:35:44ashridahRob2222: killed it? how?
01:36:10Rob2222the iriver fw blue ... its totally blue and from up/down and almost white from the other side
01:36:15Jungti1234what is 'lookinf fom'
01:36:38ashridahJungti1234: a misspelling. he meant "looking from"
01:36:40Rob2222ashridah: Sony DSC P200 fallen 50cm .. exact landed on the optics
01:36:46Rob2222"looking from"
01:36:50ashridahRob2222: ouch :(
01:37:25Jungti1234Rob2222: yes
01:38:06Rob2222Jungti1234: OK then I have the same display.
01:38:06Jungti1234My LCD looks like white when see left and right.
01:38:29Jungti1234Test picture
01:38:30Rob2222and no color changes from up and down?
01:38:40Rob2222then we have different LCDs
01:39:18muesli__Jungti1234 what are those pix symbolize?
01:39:32Rob2222at my unit the color changes at up and down moves, not left and right
01:39:38Jungti1234Do you have horizontal LCD?
01:40:08Jungti1234muesli__: It's picture to test LCD.
01:40:13Rob2222i dont know how decide ^^
01:40:19muesli__which lcd?
01:40:47muesli__did rbx produce it?
01:40:49Jungti1234New LCD(horizontal LCD), Old LCD(vertical LCD)
01:41:24Jungti1234It's hardware problem.
01:41:48Jungti1234Rob2222: hmm..
01:42:04muesli__Jungti1234 did rbx produce it?
01:42:12Rob2222what means horizontal and vertical? the viewing angle?
01:42:39Rob2222i think im to stupid to understand
01:42:42Jungti1234muesli__: I don't know it's what meaning.
01:42:45Rob2222or its to late
01:43:07muesli__how did you make it?
01:43:29muesli__just curios
01:43:34Jungti1234Downloaded in Internet.
01:45:44 Quit linuxstb ("Chatzilla [Firefox 1.0.2/20050317]")
01:46:30 Join linuxstb [0] (
01:46:33Jungti1234Iriver's fans hate that I talk LCD's problem.
01:46:46Jungti1234They attack me.
01:46:53muesli__wish lcd problem?
01:48:09Rob2222the 4 pics at this thread
01:48:17Rob2222are first generation units
01:48:25Rob2222you see it at the cover afaik
01:49:03Jungti1234Pictures prove LCD's problem!
01:49:27Rob2222i had both units
01:49:34Rob2222that with the cover at the thread
01:50:14Rob2222was in one of the first full retrail boxes with cradle, nonlcd remote and external battery pack
01:50:37Rob2222then they switch to the H340SE package without these 3 things
01:51:22Rob2222they have different covers
01:51:33Rob2222at least here in germany
01:51:41Jungti1234I don't know it.
01:51:48Jungti1234Do you have picture?
01:51:59Rob2222H340SE = H340 package without: NONLCD-Remote, Cradle, external battery pack
01:52:28Rob2222 <= new scheme
01:53:35Jungti1234that is h340se?
01:53:52Rob2222 <= old scheme
01:54:18Rob2222at least my SE package had the new cover
01:54:19 Quit Febs (Read error: 110 (Connection timed out))
01:54:30Rob2222i cant guarentee thats always so
01:57:12Jungti1234I have it.
01:58:10Jungti1234I have Fractal Black
01:59:10 Join akaidiot [0] (
02:01:56Rob2222that fractal units were in the first packages
02:02:07BHSPitMonkeyhey akaidiot...
02:02:31Rob2222seems like iriver got the problem and changed the polarisation at later units
02:03:03Jungti1234Pinspot came out first.
02:03:18Jungti1234And fractal came out.
02:03:48Jungti1234And at case, it's two.
02:04:06Rob2222i searched for a unit with full accesoires. at the time it only gaves the SE with less accesoires
02:04:23Rob2222and the "old" unit i got was fractal
02:04:46Rob2222then i got a new SE package from iriver for my bricked H140 and this was pinspot
02:05:04Rob2222yes, that case was in my SE package
02:05:27Jungti1234Pinspot isn't made now.
02:05:34Rob2222 <= an this in the first
02:05:52Rob2222all units i see at local stores are pinspot
02:06:05Rob2222but well, ok, i dont know which units are manufactured.
02:06:12Jungti1234pinspot + old case + old LCD
02:06:32Rob2222so my posted cas is the new?
02:06:45Jungti1234old case
02:07:28Jungti1234 <- old, first
02:07:33Rob2222 <= This was my first H340. Fractal, 1-part case
02:07:36Jungti1234 <- new
02:07:57Rob2222yes and in my package was a fractal player with old case
02:08:12Rob2222the pinspot player came with the new case
02:08:17Rob22221 year later
02:08:22Rob2222for me
02:08:24 Quit petur ("here today, gone tomorrow")
02:08:50Jungti1234new packages
02:09:40Rob2222we have no ext. battery pack here at the latest packages and pinspor players ;)
02:09:50Rob2222ok, so it differs from country to country
02:10:16Rob2222i thought the SE special Edition packages had all pinspot.
02:10:17Jungti1234I didn't hear about H340 SE.
02:10:36 Quit birdfish ()
02:10:52Jungti12341. pinspot + old case + old lcd
02:11:01Jungti12342. fractal + old case + old lcd
02:11:01Rob2222SE here in germany only means thats a lower price package without remote, cradle, battery pack
02:11:11Jungti12343. fractal + new case + old lcd
02:11:19Jungti12344. fractal + new case + old lcd
02:11:34Jungti12344. fractal + new case + new lcd
02:11:41Jungti1234I have 3
02:11:42Rob2222but SE replaced the normal package.
02:11:56Rob2222I had Fractal + old case + unknown display
02:12:05Jungti1234why unknown?
02:12:14Rob2222have pinspot + new case + new display
02:12:27Jungti1234Try check.
02:12:37Rob2222i didnt knew anything about the display issues as i sold the player
02:12:44Jungti1234pinspot + new case + new lcd?
02:12:52Rob2222but i knew they looked difference
02:12:57Rob2222the displays
02:13:12Rob2222i liked the display of the fractal player
02:13:20Rob2222it was brighter
02:13:22Jungti1234The Korea doesn't have a person who buy like the package.
02:13:31 Join DJDD__ [0] (
02:13:43Rob2222pinspot + new case + new lcd my current player
02:14:57Rob2222i remember as i sold one player and had to decide which
02:15:19Rob2222i liked more the pinspot than the fractal
02:15:32Jungti1234me too
02:15:40Rob2222but the fractal display was whiter when looking direct from front
02:16:46Rob2222do you know if there are fractal H320, too?
02:16:57Jungti1234I have it.
02:17:15Rob2222ah ok
02:17:46Jungti1234different interface..
02:19:08Rob2222that was a prototype, afaik
02:19:18Jungti1234It planed at begin.
02:19:32Rob2222would like to have such case :)
02:21:48Rob2222are units of those sold?
02:22:33Jungti1234I don't understand
02:22:50Rob2222are units of those build/sold?
02:22:57Rob2222withe silver buttons
02:23:12Rob2222but not sold official?
02:23:13Jungti1234Wasn't sold.
02:23:16Rob2222ah ok
02:24:07Jungti1234I'm going to make out voice file.
02:24:27Jungti1234Do you know whether you make it how?
02:24:47Rob2222how to make a voice file? no, i dont know.
02:25:56 Quit ModernExecutive ()
02:25:58Jungti1234I want that people who use H300 know LCD problem.
02:32:08 Quit matsl (Remote closed the connection)
02:34:07Rob2222Have you tried to RMA your player?
02:34:39Rob2222changed by iriver
02:34:43Rob2222new unit
02:35:22Rob2222Have you called iriver and said that you want a new player?
02:35:39Jungti1234I want new lcd
02:36:19Rob2222maybe they change your player to a player with a new lcd.
02:36:39Jungti1234But, iriver didn't recognize LCD problem.
02:36:58Rob2222I would like to know, if all problem-LCDs are in the fractal case
02:37:25Rob2222then simply sell yours and buy a new(used) one.
02:37:30Jungti1234He attacks me.
02:37:51Rob2222i cant read that
02:38:01Jungti1234you can read 'iOceon'
02:38:39Rob2222no i cant
02:38:52Rob2222maybe i have the wrong font
02:39:08Jungti1234Iriver doesn't recognize.
02:39:33Rob2222 <= is that your case?
02:39:56Jungti1234I don't manage it.
02:41:10Jungti1234He is iriver's follower.
02:41:37Jungti1234Fiery fan]
02:56:56muesli__could somebody test my ftp?
02:58:56***Saving seen data "./dancer.seen"
03:03:19 Quit akaidiot (Read error: 110 (Connection timed out))
03:11:18 Part LinusN
03:18:57 Join Tejik [0] (
03:19:09 Quit Tejik (Client Quit)
03:27:35 Quit muesli__ (Read error: 110 (Connection timed out))
03:36:30 Join birdfish [0] (
03:49:53eEyethat thing is cool looking
03:50:40eEyewish icould read that tho
03:51:52Jungti1234There is an opportunity that use it into free of charge.
03:53:49 Join CrookedRain [0] (
03:56:30 Join Maxime` [0] (n=flemmard@
03:56:50 Join Membrillo [0] (
04:05:08 Quit einhirn (Read error: 104 (Connection reset by peer))
04:11:00 Quit webguest65 ("CGI:IRC (EOF)")
04:11:20 Join kernel_sensei [0] (
04:13:54 Quit kernelsensei (Read error: 110 (Connection timed out))
04:14:01 Nick kernel_sensei is now known as kernelsensei (
04:14:01 Quit Maxime (Connection timed out)
04:16:35 Quit darkless (Read error: 110 (Connection timed out))
04:18:48 Quit TCK (Read error: 104 (Connection reset by peer))
04:19:32 Join coob [0] (
04:19:39Jungti1234Rockboy fullscreen looks like strange.
04:20:10Membrilloyeah, i think it also runs slower
04:20:18Membrilloi dont use full screen
04:21:12Jungti1234Ratio doesn't fit.
04:21:17BHSPitLappythat device you pasted before
04:21:18Membrilloyes i know
04:21:27BHSPitLappythe record buttin is the same as the A-B repeat button
04:21:28Membrilloit looks strange and stretched
04:21:41BHSPitLappywasn't somebody complaining about that button assignment in rb earlier?
04:22:40ashridahBHSPitLappy: half of that stems from rockbox not having the ability to have per-arch labels, so what's treated as "play" on rockbox ends up being 'navi'
04:22:52ashridahbut they were kinda thrown together really early on :)
04:23:02ashridahnot much thought went into it
04:28:10Jungti1234I wonder whether I make out voice file how.
04:32:37 Quit Membrillo ()
04:33:26 Join Bluechip [0] (
04:36:57 Quit Jungti1234 ()
04:39:56 Part Bluechip
04:41:20 Quit Kyl3 ("( :: NoNameScript 3.81 :: )")
04:43:36 Join akaidiot [0] (
04:43:46BHSPitLappyhola again, akaidiot ...
04:49:39 Join darkless [0] (
04:58:59***Saving seen data "./dancer.seen"
05:02:56 Join Rob2222_ [0] (
05:05:36 Quit akaidiot (Read error: 110 (Connection timed out))
05:09:23 Quit tvelocity ("Leaving")
05:19:56 Quit Rob2222 (Read error: 113 (No route to host))
05:26:05BHSPitMonkeylinuxstb: ping
05:28:11 Part Sando
05:28:27*BHSPitMonkey shoos away aegray
05:28:37aegrayi'll shoo you away
05:28:43aegrayjust try me
05:28:46BHSPitMonkeyi was here first
05:29:31*aegray types /kickban somewhere esle
05:30:26aegraykeep talkin
05:30:37BHSPitMonkeyyou started it
05:30:44aegrayprobably did
05:30:45aegrayas usual
05:31:04BHSPitMonkeyrestrain your bodily noises next time.
05:31:16aegraymy bodily noises don't tranfer to irc
05:31:20aegrayi wish they di9d
05:31:22aegraythatd be funn
05:31:45BHSPitMonkeywhat was 'blech' them
05:33:02BHSPitMonkeyi knew how to spell that too.
05:33:24aegraybet the job offers are coming in like crazy
05:33:35BHSPitMonkeyoh they are
05:33:45BHSPitMonkeytoo bad i have to turn them down to continue playing with my ipod
05:34:00aegrayyea....thats the excuse I make for myself too
05:34:03aegraywhen i fail an inteview
05:34:18*BHSPitMonkey has never failed an interview.
05:34:18aegraywoooooo no job for liefe!
05:36:06BHSPitMonkeynever apply at Target
05:36:11BHSPitMonkeyjust FYI
05:38:13 Quit mikearthur (Read error: 104 (Connection reset by peer))
05:40:45aegrayi only go for computer jobs now
05:40:51aegrayi bagged groceries for 2 years
05:40:56aegraydon't need to do that again
05:43:35BHSPitMonkeydon't you think the video player should keep the backlight on?
05:44:41BHSPitMonkeyhmm, that's actually a tricky question...on the greyscale models, the video looks better with the backlight off; maybe it should be in settings?
05:51:11aegraysend me a patch
05:51:19aegrayi'm completely rewriting
05:51:25aegrayeverything wil be in kernel for now
05:51:30aegraybecause i need somewhat real time stuff
05:51:32aegraywith interrupts
05:53:00aegrayhopefully new stuff on the way fd;sa?
05:53:20BHSPitMonkeyyou gonna make a pz2 module?
05:53:28aegraygod no
05:53:42BHSPitMonkeywhat's the disdain?
05:53:42aegrayits more direct to hardware based
05:53:56aegrayi dunno - just making a modules means noting
05:54:00aegrayits justa dding somefucntiosn
05:54:06aegraygod that was brokesn
05:54:06BHSPitMonkeywell obviously
05:54:22 Quit perplexity (Read error: 110 (Connection timed out))
05:54:34BHSPitMonkeyactually, it would probably just be filebrowser still, right?
05:54:45BHSPitMonkeyor maybe a Videos menu :D
05:54:46aegrayit don't atter
05:54:54BHSPitMonkeythat would require video indexing
05:54:59aegraythe way you find videso acn be changed easilly
05:55:03aegrayi'm talking tha player itself
05:55:06BHSPitMonkeycould mpd index videos?
05:55:10aegrayi dunno
05:55:14BHSPitMonkeyaegray: slow down
05:55:23aegraymy eyes
05:55:25aegraydot work
05:55:44aegraydrugs are bad
05:55:56aegraywhat esle does he say that with
05:59:32BHSPitMonkeywhy aren't you in #ipodlinux?
06:00:13birdfishDoes Rockbox allow for multiple genres (per track) through id3v2?
06:01:06eEyewill rockbox cook me dinner? ;)
06:01:19*birdfish wonders
06:01:43birdfishIt's too bad that sort of stuff isn't included :(
06:02:17eEyeyea.. i wouldnt mind a steak right about now
06:16:46 Join perplexity [0] (n=joust@
06:40:21 Join Paul_The_Nerd [0] (
06:45:20 Join akaidiot [0] (
06:51:23Paul_The_NerdIt is so very silent in here.
06:51:29 Join Sando [0] (
06:53:10 Part Paul_The_Nerd
06:59:00***Saving seen data "./dancer.seen"
07:06:31 Quit akaidiot (Read error: 110 (Connection timed out))
07:18:49 Join Lost-ash [0] (
07:22:11 Quit RotAtoR ("zzzzzzzzz")
07:24:45 Quit ashridah (Nick collision from services.)
07:24:50 Nick Lost-ash is now known as ashridah (
07:51:38 Join ST [0] (
08:03:48 Join Paul_The_Nerd [0] (
08:12:28*ST np: All Eyez On Me RnB Mix [02:50m/175Kbps/44KHz]
08:24:06 Join Bger [0] (n=Bager@
08:25:29 Quit Paul_The_Nerd ("Leaving.")
08:28:30 Quit ST ()
08:37:08 Join BHSPitLappy2 [0] (
08:37:11 Quit BHSPitMonkey (Read error: 104 (Connection reset by peer))
08:37:22 Quit BHSPitLappy (Nick collision from services.)
08:39:04 Join Musicmad [0] (
08:41:11 Nick BHSPitLappy2 is now known as BHSPitLappy (
08:45:00 Join Matze [0] (
08:46:42 Join akaidiot [0] (
08:52:20 Join EvilDude [0] (i=EvilDude@
08:59:03***Saving seen data "./dancer.seen"
08:59:55 Join jborn_ [0] (
09:00:56 Quit JoeBorn (Nick collision from services.)
09:00:59 Join ST [0] (
09:01:09 Nick jborn_ is now known as JoeBorn (
09:01:28SThey guys what does the vbrfix.rock do?
09:01:40 Join jborn_ [0] (
09:01:42 Quit BHSPitLappy (Read error: 110 (Connection timed out))
09:03:54Midgey31its more my lips
09:03:57 Quit Musicmad ("Trillian (")
09:04:03Midgey31sorry about that
09:04:07Midgey31wrong window
09:08:20 Join Jungti1234 [0] (n=jungti12@
09:09:49 Quit akaidiot (Read error: 110 (Connection timed out))
09:11:29 Quit Jungti1234 (Client Quit)
09:13:49 Join Jungti1234 [0] (n=jungti12@
09:25:23 Part Midgey31
09:29:34BgerST it's for fixing VBR files ...
09:32:19 Join Cari [0] (n=zeroirc@
09:33:02 Quit Cari (Client Quit)
09:34:54 Nick ashridah is now known as Lost-ash (
09:55:48 Join DangerousDan [0] (
10:01:39 Join BHSPitLappy [0] (
10:02:08 Join akaidiot [0] (
10:02:53 Join SteL [0] (
10:05:21 Quit EvilDude ()
10:07:28 Join EvilDude [0] (i=EvilDude@
10:07:35 Part EvilDude
10:13:06Papricawow thw color bmp viewer work good!!
10:14:50STpaprica: link please
10:16:35STnice, do u think it will be on the optimized build soon, btw can we talk about that build here? because i got abused for doing so one time
10:17:10 Join amiconn_ [0] (
10:17:35 Join matsl [0] (
10:17:46Papricamm dont know
10:18:30Papricai think that no probmlem to talk about the build
10:19:36 Join BHSPitLappy2 [0] (
10:19:46 Quit BHSPitLappy (Nick collision from services.)
10:20:40 Nick BHSPitLappy2 is now known as BHSPitLappy (
10:21:00BgerST? abused for talking about the optimized build ?
10:22:28Bgeri really doubt it
10:22:45perplexityI mean, it's not really optimised is it.. ? It just has a collection of patches and 3rd party plugins added and collected in one place..
10:23:04Bgerah, yes
10:23:04perplexitybut then that's just me being a language pedant I guess
10:24:11perplexityiirc Bger someone came in with a problem with the "optimised build" and someone else mentioned it was unsupported by this particular channel.. was fair I thought..
10:24:40Bgerperplexity ok, but i won't call this "abusing" :)
10:25:29perplexityI'd have to check the logs for "abuse", but I agree that my particular example was not abuse :)
10:29:27SThmm, well someone called the optimized build "rubbish", when i asked about something, instead saying it a different way..and they went on to say other things, then i left.
10:29:39perplexityJan 19 16:15:12 <B4gder> ST: the "optimized" build is not our baby
10:29:40perplexityJan 19 16:15:30 <preglow> the "optimized" build looks like a bag of shit
10:29:54perplexityhave logs, will travel
10:31:28perplexitywell, you just must have caught people on a bad day.. it happens.. but as I recall you wanted to edit the colour field in a wps..
10:31:52STi mean they could have said they didnt know.. instead of their opinions on it..ok yeh
10:32:12perplexityyou left too early :)
10:32:14perplexityJan 19 16:17:00 <ST> B4gder: ok calm down, thx for your help
10:32:14perplexityJan 19 16:17:03 * ST has quit ()
10:32:14perplexityJan 19 16:17:23 * B4gder is as calm as a... herring
10:32:14DBUGEnqueued KICK perplexity
10:32:14perplexityJan 19 16:17:31 <preglow> and by god, they're calm
10:32:14perplexityJan 19 16:17:34 <preglow> cool too
10:33:19STok, dont worry about it anymore
10:34:52 Quit akaidiot (Client Quit)
10:35:07 Quit amiconn (Read error: 110 (Connection timed out))
10:35:08 Nick amiconn_ is now known as amiconn (
10:36:27Lost-ashgeek has opinion, news at eleven.
10:36:51Lost-ashST: the biggest problem is that the optimised build starts to smell a bit like a fork after a while.
10:37:06Lost-ashand the last major third party fork resulted in some bad blood, and the death of some good ideas
10:37:39Lost-ashultimately, it might be easier to do the work, and get most of the patches the so callde 'optimised' build inserted into rockbox
10:38:04STok i see
10:38:21 Nick Lost-ash is now known as ashridah (
10:39:13ashridahbesides, i'm not sure why it's called "optimised". it's not like they had the ability to do -mcpu=athlon or whatever :)
10:40:40Bgerashridah maybe -O3 :)
10:41:19Bgerashridah i think it's more like "more usable
10:41:31ashridahvalue added
10:41:49Bgerfor example the remotes
10:42:07 Quit ST ()
10:45:25ashridahi had thought that those had already been added in
10:45:36ashridahor haven't we got a way of distinguishing them yet?
10:47:52Bgerwhere's HZ defined ?
10:48:07Bgerashridah that's not the problem
10:48:32Bgerthey can be added with an option as an start
10:48:55Bgerthe problem is that h100 and h300 remotes are very different button-situation-wise
10:50:09Bgeri started working on logical mapping (that is to map button -> action), but i don't have time ...
10:59:04***Saving seen data "./dancer.seen"
11:07:41Jungti1234Iriver followers such as beggar
11:07:44BgerJungti1234 ?
11:08:00Jungti1234They attack me
11:09:06Jungti1234Iriver's fans in Korea iriver community attack me.
11:09:45Jungti1234They hate that I inform LCD's problem to people.
11:10:13Bgeru mean the different lcds ? (the new and the old ones) ?
11:10:42Jungti1234I get angry to them.
11:11:07Bgeri don't see any reason
11:11:08Jungti1234They don't recognize LCD's problem like iriver.
11:12:46Jungti1234What reason?
11:12:55Bgerto attack you
11:13:22Jungti1234I don't know reason too.
11:13:32Jungti1234They attack me without reason.
11:13:47Jungti1234So, I'm angry.
11:15:26Bgermaybe they are iriver employees
11:16:29Jungti1234Among them, employee is.
11:18:01Jungti1234But, there is a person who I attack in writing that I write.
11:19:21Jungti1234In writing that I write only.
11:20:06BHSPitLappyJungti1234: what's your native language?
11:20:18Jungti1234hahaha um korean
11:20:32Jungti1234I don't well English.
11:20:41BHSPitLappyI think my friend speaks korean :P
11:21:07BHSPitLappyno wait
11:21:10BHSPitLappyhe doesn't :P
11:21:14Jungti1234hehe What country is he?
11:21:36BHSPitLappyi was wrong, he's vietnamese
11:21:51Jungti1234vietnamese? .....
11:22:12Jungti1234Korean and Vietnamese are perfectly different.
11:24:46Bgerhah i like every time when you say "i don't well english":)
11:24:46BHSPitLappyI know
11:25:17BHSPitLappyJungti1234: I could not remember what country he was from
11:25:26BHSPitLappyI know they are different
11:26:10Jungti1234I want to speak well English.
11:27:17Jungti1234ak!!!!!!!!!! Korean education way mistook!!
11:27:46BgerJungti1234 what i like is that this sentence ("i don't well english) is missing the main verb
11:28:03 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
11:28:24Bgerfor example :)
11:29:42Jungti1234korea isn't liveable.....
11:32:12 Join muesli__ [0] (n=muesli_t@
11:32:28Jungti1234hi muesli__
11:33:03muesli__re Jungti1234 ;)
11:33:24muesli__have u been 8h online while i was sleeping?
11:33:38 Quit dwihno (Read error: 110 (Connection timed out))
11:34:31Jungti1234I don't know '8h'
11:35:01t0mas8 hours
11:35:07BHSPitLappyJungti1234: typing error :)
11:35:08BHSPitLappyah lol
11:35:10BHSPitLappynever mind
11:35:46 Join JdGordon [0] (
11:35:46 Join Moos [0] (
11:36:30muesli__Jungti1234 motivated for another ftp test? ;)
11:36:46muesli__fine ;)
11:38:35 Quit JdGordon (Client Quit)
11:38:41 Join JdGordon [0] (
11:51:27 Join BHSPitLappy2 [0] (
11:51:46 Quit Bger ("BitchX: born to raise hell")
11:54:19 Quit BHSPitLappy (Connection timed out)
11:58:39 Quit Jungti1234 ()
12:06:50 Quit kernelsensei (Read error: 110 (Connection timed out))
12:18:41 Join mozetti [0] (
12:30:25 Part SteL
12:32:33 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
12:42:37 Join TCK- [0] (
12:43:33 Join Philip_0729 [0] (
12:55:26 Join Thus0__ [0] (
12:56:42 Quit Thus0 (Read error: 113 (No route to host))
12:58:57 Quit perplexity (Read error: 110 (Connection timed out))
12:59:08***Saving seen data "./dancer.seen"
13:00:15 Join tvelocity [0] (
13:00:24 Quit tvelocity (Client Quit)
13:04:15 Quit TCK- (Read error: 110 (Connection timed out))
13:06:29 Join tvelocity [0] (
13:08:17 Quit kernelsensei (Read error: 104 (Connection reset by peer))
13:08:32 Join kernel_sensei [0] (
13:08:41 Nick kernel_sensei is now known as kernelsensei (
13:25:40 Join perplexity [0] (i=heh26907@
13:26:51 Join einhirn [0] (
13:40:35 Join SteL [0] (
13:44:23 Part SteL
13:45:34 Join NicoFR [0] (
13:49:58Philip_0729how do i apply a .patch file to a build???
13:51:26perplexityyou don't.. you need to apply it to the source tree and then build it
13:51:29t0maslinux or windows?
13:51:31ashridahpatches can only be applied to the source, you use the 'patch' utility
13:51:53ashridahyou'd need the devkit for windows, or a cross compiler on linux or something to that effect
13:51:56Philip_0729in windows..
13:52:08t0masthen you need the devkit
13:52:10Philip_0729i have the devkit and can build the
13:52:22t0masthen use the patch command to apply the patch
13:52:32 Join webguest16 [0] (
13:55:49 Quit uwe_ (Read error: 104 (Connection reset by peer))
13:58:40Philip_0729damn thats so complicated :(
13:59:37ashridahaah, usually it's just a matter of going into the rockbox source toplevel directory, and running 'patch -p1 < /path/to/file.patch'
13:59:47ashridahif that complains that it can't find the file to patch, try -p0 instead
14:00:05Philip_0729thanks i'll try that
14:00:10ashridahthe -p option just tells patch where you are relative to where the patch is expecting to be
14:00:34ashridahso -p1 is one level deep, -p0 is the same directory
14:00:34ashridahit just depends on how the patch was created
14:02:14 Join kernel_sensei [0] (
14:05:02Philip_0729the patch is how do i know what i'm meant to patch?
14:05:56ashridahis it asking you which file to patch?
14:06:09t0masin /rockbox/apps
14:06:21 Quit kernelsensei (Connection timed out)
14:06:22 Nick kernel_sensei is now known as kernelsensei (
14:06:41t0mas1 file outsize apps
14:07:10t0mashave you tried doing it in the rockbox dir?
14:07:18t0masrockbox-devel or rockbox
14:07:20t0maswith -p0 ?
14:07:44t0masbecause this:
14:07:50t0mas−−- apps/screen_access.c24 Nov 2005 00:10:10 -00001.16
14:07:50t0mas+++ apps/screen_access.c21 Jan 2006 02:09:14 -0000
14:07:57t0maslooks like it's done from the rockbox main source dir
14:08:35t0masif that's where your source is...
14:14:14Philip_0729I have unzipped build from CVS into C:\Rockbox\home\guest\ as you woud for a normal build...
14:14:27Philip_0729*source not build sorry
14:15:46Philip_0729i need to know where to place the colorbmp.patch file and what syntax? to execute
14:17:37 Quit ashridah ("Leaving")
14:17:39webguest16put the pachfile to \home\guest\rockbox
14:18:47webguest16and then execute in \home\guest\rockbox\ the comand "patch −−binary -p0 < colorbmp.patch"
14:19:03Philip_0729I don't have that directory... :(
14:19:26Philip_0729there is a .rockbox directory uder build??
14:19:56 Join dionoea [0] (
14:20:25dionoeai was wondering if anyone here knew of a projet proting linux to Gmini 402 ?
14:20:47dionoea(or any other archos handheld video players)
14:20:56dionoea-proting +porting
14:21:11webguest16what folder do you have in \home\guest\ ?
14:21:55Philip_0729apps, bootloader, build, docs, firmware, flash, fonts, gdb, tools, uisimulator
14:22:46webguest16ok, then put the patch to \home\guest
14:23:29webguest16and execute the comand in this folder
14:24:23Philip_0729thankyou sooo much... now to try it out
14:25:03webguest16btw, the colorbmp patch is great
14:25:27Philip_0729yey time to start making some cool WPS's
14:26:32 Join needleboy [0] (
14:26:49needleboyhey everyone
14:27:10needleboyanyone knows anything regarding Linus
14:27:24needleboyLinus' color BMP loader?
14:27:35Philip_0729no but i think you need to add the colour bmp patch to your optimized build
14:27:55needleboythat depends
14:28:07needleboyhe said the RWPS crashes if it tries to load a color BMP
14:28:31Philip_0729i'll have a test and see what this means...
14:28:41needleboy\i want to know if it'll happen whenever, or only when the RWPS is set to load the color BMP
14:28:53Philip_0729yeh i'll test that
14:29:03needleboyi don't have an LCD remote, so I can't check it...
14:29:07needleboylet me know?
14:29:09linuxstbThe RWPS probably expects mono bitmaps - so if the code doesn't handle that....
14:29:19 Part dionoea
14:29:23Philip_0729building at present
14:29:48needleboybut the RWPS has it's own settings file
14:30:07Philip_0729thats what i thourght it wouldn't be a problem
14:30:15needleboyif i load a color WPS does it change the RWPS as well?
14:30:48needleboyanyway, if this works, it renders the ColorWPS patch useless...
14:31:05needleboywhich is a bummer cause it took me a whole day to fix, and it was only 3 days ago
14:31:07 Join actionshrimp [0] (
14:31:16needleboybut it's a better option so... whatever
14:31:50Philip_0729i'm going to assume i have to heavily modify a wps to test this it may take a while
14:32:13needleboyjust load one color BMP and see if it crashes the remote
14:32:36needleboyjust take a WPS with the ROCKbox logo, and replace the logo itself with a color one
14:34:14linuxstbBagder: A week or two ago, you mentioned you might look at adding bmp2rb to the build system. Are you still intending to do that?
14:35:04 Quit DreamTactix291 (Read error: 110 (Connection timed out))
14:36:41 Quit NicoFR ()
14:38:48 Join Kohlrabi [0] (
14:41:53Philip_0729colour bmps on wps work like a charm
14:42:13needleboywhat about the remote?
14:42:17Philip_0729plug in remote and no crash
14:42:32needleboydoes the remote still load the old RWPS?
14:42:48needleboyso the problem is only if the RWPS file itself loads color BMPs?
14:42:52Philip_0729crash when try to load colour bmps onto remote
14:42:55needleboynot a problem then...
14:42:59needleboycool, it's in
14:43:00Philip_0729no not realy
14:43:09needleboya new version in 30 minutes
14:43:16Philip_0729i'll start skinning
14:48:58Philip_0729is there a transparency colour?? cos i load a white background it covers all the text....
14:50:35Philip_0729i'll try pink, black... any other suggestions??
14:50:37needleboytry the pink default
14:50:43needleboythat's usually it
14:50:51needleboybut why a background?
14:52:13Philip_0729i was going to essentially copy over a skin i made for iriver firmware and they use background images
14:53:02Philip_0729hmm. nope and nope
14:54:27 Join SteL [0] (
14:55:29JdGordonhey all
14:55:38JdGordonis there any calendar plugins for rb yet?
14:55:47JdGordonactually, weekly timetable like plugins
14:56:30needleboyPhilip, what's the RGB value for the transparent pink?
14:57:17Philip_0729its usually 255 0 255 but i tried that and it was ...........
14:58:31needleboymaybe there's no transperancy in 24bit BMPs?
14:58:34needleboyanyone knows?
14:59:10***Saving seen data "./dancer.seen"
15:04:09needleboywell, i think there's no transperancy yet
15:04:18needleboyi tried everything
15:04:58JdGordonall 16 billion combos?
15:05:21needleboyis there support for it in BMPs at all???
15:05:28needleboydoesn't seem like it
15:05:44JdGordonisnt there something in the metainfo at the start of a bmp about which colour is transparent?
15:05:49needleboyif i save a transaprent image to BMP, then reload it in Photoshop, it just shows BG as white
15:06:50linuxstbNo, BMP files don't support tranparency. You would have to choose a random colour and hard-code that colour to mean "transparent".
15:06:55CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
15:06:55*preglow reappears
15:07:11linuxstbgood afternoon.
15:08:01preglowi suppose it is
15:08:07preglowat least there's no storm anymore
15:08:55linuxstbI managed to "dd" an ipod firmware partition to my h140's fat32 partition last night...
15:10:26linuxstbI was surprised that the iriver firmware was happy to enter USB mode, even though the filesystem was corrupt.
15:10:39preglowoh really?
15:10:43preglownow that _is_ a surprise
15:10:44needleboylinuxstb: hardcode in the patch itself?
15:10:53preglowi would have lost quite a bit of money in that bet
15:11:16linuxstbneedleboy: Yes. Or more accurately, in the LCD driver(s).
15:11:27JdGordonhas any1 tried the h300 optimeized build?
15:11:33linuxstbpreglow: Maybe it's because the partition table was still intact.
15:12:20needleboyok, i'll ask linus about this
15:13:31needleboyJdGordon: i'm compiling that version
15:13:57JdGordonit keeps crashing when it tries to load one of the new wps's
15:13:57preglowlinuxstb: well, the impression i have is that iriverfw isn't very forgiving if it comes to a broken partition
15:14:27linuxstbSo I was probably lucky because my partition was _very_ broken. If it was half-working, I guess it would have choked.
15:14:44linuxstbBut I of course still had the rockbox usb bootloader mode as well.
15:14:56preglowwhich is feature of the year
15:15:47linuxstbYep. Apple kindly already provided that for us.
15:16:07JdGordonneedleboy: i have the nonunicode rockbox fw... could that be why?
15:16:13preglowprobably wouldn't even have a port if that wasn't so
15:17:07needleboyJdGordon: Could that be why WHAT?
15:17:11needleboyoh, i see
15:17:14needleboyyes it is
15:17:19JdGordonhmm... ok
15:17:23needleboythere's new color codes in the WPS
15:17:34linuxstbJdGordon: Is there a reason you've stuck with a non-unicode rockbox?
15:17:59preglownonunicode rockbox fw?
15:18:07JdGordonnot really... i speak english so dont need it
15:18:08preglowyou've stuck with a build that's older than a month?
15:18:18JdGordonforce of habit or something
15:18:33linuxstbI only speak English, but there's no advantage to not upgrading. It doesn't hurt anything.
15:18:51preglowi just upgrade because of all the other features
15:19:03preglowi use english on all daps/computers/whatever anyway
15:19:40linuxstbUnicode has been in Rockbox CVS since 6 December 2005 - 6 weeks is a long time in Rockbox development...
15:20:15JdGordonhappens to be the same length of tiume uve been out of the country
15:20:21needleboycould anyone else but Linus code in the transperency issue?
15:21:34*JdGordon cant find a damn reseter thingy
15:21:45preglowwhat do you need transparency for?
15:22:50needleboyLinus made a patch that allows 24Bit BMPS to be used in the WPS
15:23:03preglowneedleboy: coding transparency is easy, just find a crazy ass shade of pink, call that transparant, and have the drawing routine NOT draw all pixels that are crazy ass pink
15:23:19needleboyBut the BMPs loaded in the WPS now cover the text
15:23:22 Join Jungti1234 [0] (n=jungti12@
15:23:35needleboypreglow: sure thing, but i'm not a programmer, so no idea how to do this
15:23:49Jungti1234Big match!!
15:23:55preglowtoo bad i'm not the slightest bit interested, then
15:24:09Jungti1234Korea vs Greece
15:24:26needleboypreglow: you're not, but every color LCD user is...
15:24:26preglowwhat sport?
15:24:33preglowneedleboy: oh, i use a colour lcd
15:24:43preglowdon't much care for soccer anymore
15:24:49Jungti1234Korea 1 : 1 Greece
15:24:55needleboyand you don't see this as a huge step up?!?!?
15:25:17preglowwell, yeah, i'm not just interested enough in coding it myself
15:25:21preglowi don't care very much about looks
15:25:31JdGordonneedleboy: does the wps flash with the bmps like they did 5 weeks ago when i last tryed a wps with bmps?
15:25:37needleboyi'm asking for help... not interest...
15:25:57needleboyJdGordon: What?!?
15:26:14preglowneedleboy: perfect situation to learn coding!
15:26:22needleboyi've started actually
15:26:24preglowthink of the glamour and the babes
15:26:33preglowthen hooray
15:26:42JdGordonthe playing screen blinks when it refreshes the screen if it had any bmps on it... does it still do that?
15:26:58needleboynot that i'm aware of
15:27:01JdGordonoh good :D
15:27:17needleboyJdGordon: take a look at this:
15:27:20needleboyyou have all the info there
15:27:27needleboyjust clean reinstall
15:28:16JdGordoncrap... iriv is in korean for some reason.. can some1 let me know which menu option is the language?
15:29:08Philip_0729lol hold navi press navi go down one press navi iirc
15:32:50 Quit needleboy ("CGI:IRC (EOF)")
15:33:54JdGordonwtf?? imcompatible model on bubbles and brickmania?
15:35:46perplexitythat would be old 3rd party plugins with the new plugin loader I guess
15:35:57perplexitythey need to be modified slightly and recompiled
15:39:49Philip_0729what is the RGB value of rockbox blue??
15:40:45linuxstb#define LCD_DEFAULT_BG LCD_RGBPACK(182, 198, 229) /* rockbox blue */
15:40:57lostlogicSlasheri: ping?
15:41:08 Join ghode|afk [0] (
15:42:02 Quit webguest16 ("CGI:IRC (Ping timeout)")
15:42:04 Quit Rob- (Read error: 110 (Connection timed out))
15:43:53Slasherilostlogic: pong
15:45:41lostlogichey, I'm trying to figure out a way to do make a seamless seek functionality on software codecs, but with clearing the pcmbuf on seek complete, I can't see a way to do it. Thoughts?
15:48:03Slasheriah, that is necessary to prevent old audio fragemnts playing when the seek is complete (audio is paused while seeking)
15:49:11lostlogicAre we explicitly against playing those audio fragments as we pass them?
15:50:12SlasheriHmm, it would sound like weird when you release the seek button and you would hear a fragment from the previous position and right after that the playback continues at the right position
15:50:56lostlogicSlasheri: I mean more like seek-peeking like some DAPs or CD players do so you hear glimpses of audio as you pass them
15:52:27Slasheriah, then it would be an entirely different thing to do
15:53:01Slasheriand then we shouldn't of course clear the pcm buffer
15:53:14 Join jackinabox^ [0] (
15:53:24jackinabox^hi all
15:54:02jackinabox^!ping pls
15:54:06lostlogicbasically it seems like we're putting in a fair amount of effor to ensure dead silence during seeking by pausing playback and clearing the buffer on both sides of it, but with some tweaking we might get a useful feature by just not doing that.
15:55:14Slasherihmm, true. that would be a nice thing to try at least
15:55:45 Quit jackinabox^ (Client Quit)
15:55:56lostlogicwill play with it. otherwise I'll see if there's some other way to keep seeking clean and get the seamless that ab-repeat needs.
15:55:57 Join jackinabox^ [0] (
15:56:23jackinabox^back, think i got rid of my lag
15:59:05 Part jackinabox^
15:59:09 Quit SteL (Connection timed out)
16:01:06 Quit matsl (Remote closed the connection)
16:02:02preglowlostlogic: what, you're looking at the playback code and understanding it?
16:02:13 Quit BHSPitLappy2 (Connection timed out)
16:02:20lostlogicpreglow: somewhat :-P
16:02:22preglowwe have a messiah!
16:02:41 Join SteL [0] (
16:03:34 Join BHSPitLappy [0] (
16:09:24 Quit JdGordon (Read error: 110 (Connection timed out))
16:13:59linuxstbSlasheri: I have A fundamental question about the rockbox playback code - how many different threads are involved in audio playback, and what they each do?
16:15:20 Join BHSPitMonkey [0] (
16:16:53Slasherilinuxstb: there are three different threads at the moment. 1) the playback core that controls buffering, 2) codec thread where audio codecs run, and 3) a separate codec thread for the voice codec
16:18:39lostlogicSlasheri: pcmbuf and codecbuf run on the same thread?
16:19:10lostlogicI guess that makes sense since there has to be load balancing between the two buffers kinda.
16:19:23linuxstbSo does the main thread (running the UI) only communicate with your 1) ?
16:19:25preglowSlasheri: you have to memcpy the entire codec address space, plus iram, yes?
16:19:28preglowSlasheri: when using voice ui
16:20:05Slasherilostlogic: hmm, no. codec thread keeps the pcm buffer full and audio core thread makes sure that codec buffer keeps filled
16:20:21Slasheripreglow: yes, true
16:20:47preglowgood reason for me to make an elf loader :>
16:20:50lostlogicSlasheri: ok, so the callbacks run on the codec thread technically.
16:21:06Slasherilostlogic: yep, main thread only communicates with the audio core thread
16:21:16preglowdsp also runs in codec thread
16:21:18Slasherilostlogic: yes
16:24:48lostlogicpcmbuf_play_stop clears the buffer, but does not pause playback (if I'm reading this right)?
16:25:34 Join mikearthur [0] (
16:26:12 Join uwe_ [0] (
16:27:45lostlogicwell I've not gotten my desired effect yet, but it's kinda amusing to have the audio keep playing where it was during seeking.
16:27:53Slasherilostlogic: correct.. it doesn't pause the playback but it stops it :D so pcm dma needs to be started again
16:36:32 Join DocterD [0] (
16:37:35lostlogicgettin' there.
16:39:22 Quit tvelocity ("Leaving")
16:40:14 Join goa [0] (
16:43:27lostlogicproblems arise with synching the GUI with the seeking. that's going to take much more work :(
16:44:18lostlogicSlasheri: how would you reproduce the bug that you put the pcmbuf_play_stop in seek_complete_callback to fix so I can make sure I don't resurrect that bug?
16:44:51preglowis there some way to sync wps to where the pcm buffer is currently at?
16:44:57preglowi'm tired of it being several seconds wrong
16:45:30lostlogicit attempts to by calculating the pcmbuffer latency to give the WPS the place to display
16:45:38Slasherilostlogic: if you are going to remove it, you should also remove the pcmbuf_play_stop call from AUDIO_FF_REWIND queue event in the audio core thread
16:45:51Slasherithen it will have effect
16:47:00lostlogicSlasheri: actually in trying to get the desired seek peeking effect I removed the front one but not the end, and took out the pauses in gwps-common, and moved the audio_ff_rewind call into the calculation loop, but it's going to take more than I thought to keep the GUI from getting confused, and the codec from exploding from the many many ff_rewind calls.
16:47:40lostlogicplus it becomes much harder to keep from overrunning the end of the song and stuff, so I'm nixing that particular project for now.
16:47:51lostlogicand going back to adding a separate seamless seek audio function.
16:48:33Slasherihehe, true. but that sounds good :) now going to eat something
16:48:44 Join Paul_The_Nerd [0] (
16:50:15 Join ender` [0] (i=ychat@
16:51:00 Quit BHSPitMonkey (Read error: 110 (Connection timed out))
16:52:54lostlogicpreglow: your WPS is several seconds wrong from song position?
16:56:04 Join kernel_sensei [0] (
16:58:01 Quit kernelsensei (Connection timed out)
16:59:02 Nick kernel_sensei is now known as kernelsensei (
16:59:12***Saving seen data "./dancer.seen"
17:01:56preglowlostlogic: i assume so, since it track position indicator ends several seconds before the track is at the end
17:02:32lostlogichmm, it probably stops moving too soon, because there are no more decode events, and decode events trigger the actual updates.
17:04:24preglowstill, doesn't look too pretty
17:04:36preglowdoes it stop even longer before when crossfading is enabled, then?
17:04:55 Quit Jungti1234 (Read error: 104 (Connection reset by peer))
17:05:05lostlogicpreglow: not sure −− haven't used crossfade yet. I'm gonna see if I can make it update whenever the latency changes as well
17:05:54 Join G-G-G-Gesus [0] (
17:10:30 Join needleboy [0] (
17:11:17 Quit needleboy (Client Quit)
17:13:25lostlogicseems like there are several variables that belong in IRAM in pcm related code... the interrupt handler is in IRAM, so should the variables and functions that it calls in the usual case.
17:19:54 Join kernel_sensei [0] (
17:19:59 Join vger_ [0] (
17:20:12 Quit kernelsensei (Nick collision from services.)
17:20:47 Quit thegeek_ (Read error: 110 (Connection timed out))
17:21:02Paul_The_NerdOkay, I can't seem to do Makes any more.
17:22:31 Join BHSPitLappy2 [0] (
17:23:33Paul_The_NerdWith a fresh CVS update, in a clean build directory, I get "Missing Separator".
17:24:56lostlogichmm, might be able to register a watermark callback that would update as the DMA empties teh remaining buffer.
17:26:11linuxstbPaul_The_Nerd: Do "make V=1" to see exactly where the error is.
17:29:38lostlogicpreliminary results say that IRAMing the DMA callback related code gives a statistically significant performance improvement.
17:29:46Paul_The_NerdMakefile:1: *** missing separator. Stop.
17:29:47Paul_The_Nerdmake[1]: *** [rocks] Error 2
17:29:50 Join jborn__ [0] (
17:29:53Paul_The_NerdAh, gotcha
17:30:19Paul_The_NerdThanks, just after pasting it I realized where my problem is. Forgot I had that extra mpeg2 stuff still there.
17:30:49 Quit JoeBorn (Nick collision from services.)
17:30:55 Nick jborn__ is now known as JoeBorn (
17:32:45Paul_The_NerdOr, not.
17:33:14*Paul_The_Nerd wishes he were more knowledgeable about GCC / Make / etc.
17:34:34G-G-G-Gesussorry to bother you guys, but i was wondering if there's some screenshots of the rockbox/the ui
17:34:59Paul_The_NerdThe UI is pretty much just text based.
17:35:36G-G-G-Gesusoh ok
17:35:55Paul_The_NerdLinuxstb: CC xlcd.c AR+RANLIB /home/DarkkOne/rockbox-devel/h120/libplugin.a Makefile:1: *** missing separator. Stop.
17:36:10 Quit BHSPitLappy (Connection timed out)
17:37:00 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
17:38:22 Quit kernel_sensei (Connection timed out)
17:38:25linuxstbPaul_The_Nerd: I'm guessing it's your apps/plugins/Makefile that's somehow bad.
17:39:01linuxstbTry doing "cvs update" in your apps/ directory, and see if it tells you if there were any conflicts.
17:39:45Paul_The_NerdI would assume a C is a conflict? I swear that in the past it was actually somewhat verbose about what conflicted..
17:39:50 Join Rob- [0] (
17:43:47preglowc is a conflict
17:43:59preglowit's never been very verbose
17:44:23Paul_The_NerdI swear I'd had one in the past that told me a little more about it, but maybe I'm confusing CVS and patch.
17:44:27Paul_The_NerdAh well, all resolved now
17:44:51lostlogicPaul_The_Nerd: the first time you cvs up with a conflict it talks about the merge failing, but after that, it's not making new conflicts so it just shows the file as 'C' status.
17:45:11Paul_The_NerdAaah. Gotcha
17:46:01lostlogichow does 3% boost improvement for ~250 bytes of IRAM in pcm_playback.c and pcmbuf.c sound?
17:48:29 Quit DocterD (Remote closed the connection)
17:50:10 Quit ender` (" The Web is a procrastination apparatus: It can absorb as much time as is required to ensure that you won't get any real work)
17:52:40 Join BHSPitLappy [0] (
17:55:06 Join mirak [0] (
17:58:52 Quit G-G-G-Gesus ()
18:01:16 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
18:04:42 Part Paul_The_Nerd
18:06:43 Join DocterD [0] (
18:06:56 Join DreamTactix291 [0] (
18:08:18 Join webguest66 [0] (
18:08:57linuxstblostlogic: I don't think anyone could complain about that :)
18:09:52lostlogic:) still on the hunt in this area of code for similar "low hanging fruit"
18:13:51Slasherilostlogic: Hmm, have you checked how much there are currently iram left in the main code?
18:14:09 Quit BHSPitLappy2 (Success)
18:14:23 Join b0br [0] (
18:15:14lostlogicSlasheri: iend in is at 8188 here, which says there's still 16k left right?
18:16:02Slasherihmm, interesting.. just a moment
18:16:02lostlogicor have to count stack too so that's only 7k?
18:16:11Slasheriyes, stacks also
18:16:14lostlogic7k left.
18:16:25Slasheriok, so still plenty of iram left..
18:16:48 Join Benacool [0] (
18:17:36Slasherithat's good then
18:18:20lostlogicSlasheri: in order to keep the playback updating as tracks end, I'd like to add another watermark that won't always be set (the current one is always set, just set in different places), this watermark will be used to bridge the gap between when codec decoding of a song ends, and when the playing of the next song begins
18:18:37b0branybody knows, how difficult it would be to write instrument tuner for RB, something like this: ?
18:18:44 Join BHSPitLappy2 [0] (
18:19:05Slasherihmm, nope.. i just checked by creating a large static char array and it seems to be there is only little over 2k of iram left
18:19:49lostlogicweird... why does .stackend show a188 here?
18:19:58Slasherilostlogic: hmm, interesting. please do so if you can make it work :)
18:20:09lostlogicSlasheri: cool, I'm workin' on it.
18:21:17lostlogicstatic char filler[7800] IDATA_ATTR;
18:21:28lostlogicjust fit into IRAM here no complaints, are our compilers that different?
18:21:59Slasheriweird.. i just tryid static char filler[3000] IBSS_ATTR; and that wouldn't fit
18:22:01 Join gtkspert [0] (
18:22:51Slasheri(that in playback.c)
18:22:56lostlogicI wonder which of these situations is more similar to the 'official' build.
18:23:15 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
18:25:28 Quit DocterD ("Leaving")
18:25:48linuxstblostlogic: Do you have an ARM compiler installed? It's very easy to break ipod builds by changing iram usage.
18:26:01Slasheriinteresting.. i have .stackend at 0xb49c.. so little less than 3k of iram left..
18:26:28 Join Paul_The_Nerd [0] (
18:27:48lostlogiclinuxstb: yeah, lemme check.
18:28:17 Quit BHSPitLappy (Connection timed out)
18:28:22 Part kro
18:30:38lostlogiclinuxstb: I should probably get a 4.0 arm compiler setup
18:30:58lostlogicstackend on ipod build here is add0
18:31:28lostlogic_even with_ my filler array in.
18:31:51 Join _FireFly_ [0] (
18:32:02lostlogic8f58 otherwise, so it looks pretty clear with gcc-3
18:35:57 Quit Paul_The_Nerd ("Leaving.")
18:43:59miraklostlogic: what I don't get is that the stack is supposed to be 8k
18:44:10mirakbut I have seen bigger arrays
18:44:24mirakin a struct declared localy to a method
18:44:27mirakand it doesn't crash
18:44:36mirakI don't know how this work
18:45:16lostlogicmirak: if they aren't in IRAM it doesn't matter
18:45:23lostlogicmirak: they are statically allocated from system ram
18:46:21mirakwhat do you mean staticaly ?
18:46:52lostlogicI mean that whenever that plugin or piece of code is loaded, that array's memory is allocated
18:47:32mirakin wich memory arrea is it allocated ?
18:47:49 Join linuxstb_ [0] (
18:48:01 Quit Benacool (Read error: 104 (Connection reset by peer))
18:48:02lostlogicfor plugisn in the plugin buffer in SDRAM, for the system just out of the general SDRAM pool, afaik
18:48:12linuxstb_mirak: Do you have an example in the current Rockbox CVS that does that?
18:49:27miraklinuxstb_: well in the mpeg2 decoding stuff I did there is a struct wich contains an array of size 16k
18:49:58mirakand hem
18:50:19 Join ender` [0] (i=ychat@
18:51:13 Quit ghode|afk (Read error: 104 (Connection reset by peer))
18:51:40 Join BHSPitLappy [0] (
18:52:13miraklinuxstb: ok my bad it's not
18:52:27mirakit uses the fake malloc
18:52:46mirakthough this part needs to be put in iram
18:52:59mirakI don't know how, but xvid was better designed for that
18:53:19mirakwith xvid a little buffer was allocated on the stac and the idct was processed on it
18:53:39mirakactually with mpeg2 it's done in the normal ram
18:54:31mirakthat's certainly why the result are disapointing for mpeg2
18:55:06 Quit ender` (Client Quit)
18:55:54 Quit _FireFly_ ("Leaving")
18:56:04 Quit Philip_0729 ("bye everybody")
18:57:29 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
18:59:14***Saving seen data "./dancer.seen"
19:00:42 Join Mmmm [0] (
19:09:50 Join RotAtoR [0] (
19:10:19 Quit BHSPitLappy2 (Read error: 110 (Connection timed out))
19:20:59 Join BHSPitLappy2 [0] (
19:21:11 Quit BHSPitLappy (Nick collision from services.)
19:21:14lostlogicSlasheri: Ok, so I've got the better UI update whatzit basically working but I don't see where in the code to deregister my callback to make sure that it is always deregistered when the next track starts playing
19:21:19 Nick BHSPitLappy2 is now known as BHSPitLappy (
19:29:03 Join linuxstb_ [0] (
19:50:11 Quit linuxstb_ ("CGI:IRC")
19:50:14 Quit linuxstb (Remote closed the connection)
19:50:34 Join linuxstb [0] (
19:53:09 Quit BHSPitLappy (Connection timed out)
19:56:03 Join DangerousDan [0] (
19:56:53 Join Paul_The_Nerd [0] (
19:58:33 Join ghode|afk [0] (
20:07:26 Quit gtkspert ()
20:15:49 Part SteL
20:17:21 Part Paul_The_Nerd
20:19:03 Quit Mmmm ()
20:22:27 Join thegeek [0] (
20:30:17 Join einhirn_ [0] (
20:39:33preglowb0br: not too hard, but it will require quite a bit of dsp knowledge
20:39:43preglowand people with that aren't too common around here
20:47:58 Join BHSPitLappy [0] (
20:48:55 Quit einhirn (Read error: 110 (Connection timed out))
20:49:40 Join alphakiller [0] (n=bohas@
20:59:17***Saving seen data "./dancer.seen"
21:01:08 Quit CrookedRain ()
21:02:09 Join linuxstb_ [0] (
21:02:42linuxstb_I've been working on adding bmp2rb a little more, and would like to commit something to CVS to keep the momentum going. I've limited the bmp2rb usage to creating native bitmaps for the main lcd - so this means that it doesn't support mono bitmaps for the main LCD (i.e. the icons), or bitmaps for the LCD remote.
21:04:39linuxstb_So there is an apps/bitmaps directory which contains native-format bitmaps (as .bmp files), and also an apps/plugins/bitmaps/ directory for native bitmaps used by plugins.
21:06:04lostlogicgah, the problem is that the audio threads completely forget about the outgoing track's state once the codec is done decoding it. So either I have to hold onto that state for longer, or I have to have the WPS do the updates for the tail-end of the song.
21:06:05 Nick jelle-k|offline is now known as jelle-k (
21:07:48 Join thegeek_ [0] (
21:11:47 Join BHSPitLappy2 [0] (
21:12:01 Quit BHSPitLappy (Nick collision from services.)
21:12:10 Quit alphakiller ()
21:12:21 Nick BHSPitLappy2 is now known as BHSPitLappy (
21:13:18 Quit JoeBorn ("Leaving")
21:15:10linuxstb_Patch is here:
21:15:23linuxstb_Plus the new bmp files and Makefiles:
21:16:40Papricaplugin bitmap loader?
21:18:09 Quit linuxstb (Read error: 110 (Connection timed out))
21:18:19 Join FingerSoup [0] (
21:18:28BHSPitLappylinuxstb_: sup
21:18:35linuxstb_No. It's a patch to the Rockbox build system that uses bmp2rb at compile-time to automatically convert .bmp files to .c files - in the appropriate format for that target.
21:18:37 Nick jelle-k is now known as jelle-k|offline (
21:19:25FingerSouperp!!! new plugin code broke my connect 4 plugin - I get an error with TEST_PLUGIN_API
21:19:34FingerSoupwhen compiling for sim...
21:19:41FingerSoupany ideas on fix?
21:19:49 Join BHSPitLappy2 [0] (
21:20:14 Quit BHSPitLappy (Nick collision from services.)
21:20:24 Nick BHSPitLappy2 is now known as BHSPitLappy (
21:22:20 Join BHSPitLappy2 [0] (
21:22:50 Quit BHSPitLappy (Nick collision from services.)
21:22:51PapricaMAKE in bitmaps
21:22:51PapricaBMP2RB rockboxlogo.220x68x16.bmp
21:22:51DBUGEnqueued KICK Paprica
21:22:51Paprica./rockboxlogo.220x68x16.bmp: 1: Syntax error: word unexpected (expecting ")")
21:22:51Papricamake[2]: *** [/home/guest/rockbox-bleeding/simu/apps/bitmaps/rockboxlogo.220x68x
21:22:52***Alert Mode level 1
21:22:52Paprica16.c] Error 2
21:22:54Papricamake[1]: *** [libbitmaps] Error 2
21:22:55Papricamake: *** [all] Error 2
21:23:07 Quit thegeek (Read error: 110 (Connection timed out))
21:23:17 Quit linuxstb_ (Read error: 104 (Connection reset by peer))
21:23:22 Join linuxstb__ [0] (
21:23:26 Nick linuxstb__ is now known as linuxstb (
21:23:29CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
21:23:29*amiconn appears
21:23:55amiconnFingerSoup: Remove TEST_PLUGIN_API - it doesn't exist anymore
21:24:46preglowwith all these new games coming, we might want to move all game plugins to a separate menu :>
21:25:09*amiconn wonders about Slasheri's codec lader fix
21:25:41amiconnThe old code should behave *exactly* the same
21:25:42Slasheriamiconn: the problem was that size was reduced to 0.. so the header check failed
21:26:09FingerSoupok, thanks
21:26:12Slasherithat header check (including the struct size check) was new
21:26:20*amiconn silly
21:26:28 Join BHSPitLappy [0] (
21:28:33amiconnSlasheri: There was s slight chance that it worked before - if the codec did wrap in the buffer
21:29:07 Quit einhirn_ ("Miranda IM! Smaller, Faster, Easier.")
21:29:17amiconnSomehow I overlooked that size was manipulated...
21:29:24Slasheriamiconn: hmm, it worked before fine
21:29:36amiconnNo, I mean before your fix
21:29:44Slasheriah, hmm
21:30:19Slasheriyes, i very slight chance :)
21:30:21 Join einhirn [0] (
21:31:55lostlogicI think I finally have updates->end of track working.
21:32:10Slasherisounds nice :)
21:32:22 Join linuxstb_ [0] (
21:32:23lostlogicI'll want you to review this before I commit if you are willing...
21:32:24Slasherido you have a patch?
21:32:30Slasherisounds good
21:32:31lostlogichehe, yeah, willbe posting soon.
21:32:35linuxstb_Paprica: Did you re-run ../tools/configure?
21:32:46Papricaah no
21:32:53***Alert Mode OFF
21:33:10 Join Paul_The_Nerd [0] (
21:33:20lostlogicSlasheri: btw, I pulled out that pcmbuf_play_stop in the seek_complete_callback and it doesn't seem to let any 'old' audio sneak through.
21:33:20 Join linuxstb__ [0] (
21:34:12Slasherilostlogic: ok, probably that isn't issue as long as the buffer will be clear at some other place
21:34:13BHSPitLappyhey linuxstb__ , linuxstb_ , and linuxstb
21:34:25 Quit linuxstb__ (Client Quit)
21:34:53lostlogicSlasheri: yeah, wps pauses, and clears at the beginning of a seek, and during pause, no data is allowed into the buffer, so I'm fairlu sure that is OK (plus it lets me make AB repeat seamless work :))
21:35:51preglowSlasheri: but yeah, wasn't some of your fixes supposed to make playback stop if you skip to a non-buffered track?
21:38:29Slasheripreglow: yes, at least while skipping. But still, there can be about half a second delay
21:38:38 Quit linuxstb (Nick collision from services.)
21:39:04 Nick linuxstb_ is now known as linuxstb (
21:39:42preglowit doesn't stop at all here
21:39:45preglowwith yesterdays build
21:39:50preglowbehaves just like before
21:40:02preglowi skip to another track, and it continues playing until the new ones has been loaded
21:41:08lostlogic @ Slasheri , preglow
21:42:20preglowwhat's it for?
21:42:23Slasherilostlogic: thanks, i will look that soon (maybe tomorrow)
21:42:38lostlogicpreglow: makes the track status keep updating to the end of track
21:42:43lostlogicinstead of stopping with a few seconds left
21:42:54preglowi have no idea how playback and buffering works, so don't think i can be of much use reviewing it, though
21:43:07lostlogicpreglow: just wanted you to test since you complained ;)
21:44:43 Quit BHSPitLappy2 (Success)
21:46:29 Quit DangerousDan (Read error: 104 (Connection reset by peer))
21:46:48 Join BHSPitLappy2 [0] (
21:51:32 Join BHSPitMonkey [0] (
21:55:08 Quit CoCoLUS (Read error: 104 (Connection reset by peer))
21:56:30 Join CoCoLUS [0] (
21:59:37 Quit BHSPitLappy (Connection timed out)
22:00:33 Join thegeek [0] (
22:02:07 Join solexx [0] (
22:03:31 Join DangerousDan [0] (
22:03:51 Join NicoFR [0] (
22:06:01 Quit thegeek_ (Read error: 104 (Connection reset by peer))
22:06:33 Quit b0br ("CGI:IRC")
22:06:49 Join thegeek_ [0] (
22:08:23 Join SteL [0] (
22:09:22 Quit Matze (Read error: 104 (Connection reset by peer))
22:13:56 Quit solexx_ (Read error: 110 (Connection timed out))
22:19:12 Quit jborn_ ("")
22:19:39 Quit vger_ ("BitchX: No windows left!")
22:24:34 Quit thegeek (Read error: 110 (Connection timed out))
22:25:55 Quit thegeek_ (Read error: 104 (Connection reset by peer))
22:38:08 Quit SteL ("Chatzilla 0.9.67+ [SeaMonkey 1.0b/2005121900]")
22:44:17Slasherilostlogic: the patch looks good, however it doesn't work always with some hard special situations, like seeking when the playback is at the very end (try just seeking to the beginning, the elapsed time goes wrong as the track changes)
22:44:27Slasheribut i think you should commit it :)
22:44:58 Join Pi___ [0] (
22:45:25lostlogicok, I will take a few minutes to see if I can improve that brokenness, and commit it. thanks
22:46:23lostlogicwhy are pcmbuf_read_index and pcmbuf_write_index declared volatile? Do multiple threads read and write them to make that declaration necessary? It makes some of the compiled code slower than it would otherwise be.
22:46:24amiconn'Improve that brokenness' sounds ambiguous ;)
22:46:36lostlogicamiconn: yep, I'm going to make sure the elapsed time is wrong at all times.
22:46:51Slasherithat can be hard to fix if we want to allow seeking of the previous track.. or you could just ignore the seeking if we are too end of the track
22:47:03Slasheriamiconn: hehe :D
22:47:13Paul_The_NerdIt sounds like you're trying to make it a better form of broken.
22:47:21lostlogicSlasheri: not sure I understand your last line
22:47:49Slasherilostlogic: hmm, in fact i can try to improve that in future
22:47:52Pi___silly question, what does << do?
22:47:59lostlogicPi___: logical left shift
22:48:09preglowor arithmetic
22:48:16preglowin practice, at least
22:48:27lostlogic:) left shift, in short.
22:48:54Pi___thanks though i dont know what it means, i can google that
22:49:35Slasherilostlogic: it might be good to keep the indexes as volatile because dma interrupt modifies the read index at least.. but i am not sure if there is any real reason for that
22:49:41preglowit just shifts the bit pattern of a variable left
22:49:57Pi___got it thanks
22:50:13Slasherilostlogic: yep, seems to be unnecessary
22:50:29Slasherii just wanted to be sure ;)
22:50:48lostlogicSlasheri: yeah, in concept they _are_ volatile... but inpractice the volatile declaration only makes the interrupt handler code slower and doesn't seem to impact the other uses of the variables.
22:51:11Slasheriah.. then just go ahead and remove the volatile declarations
22:51:25Slasherithose in pcmbuf.c seems to be useless at least
22:51:55lostlogicI wonder if it's enough to make a statistically significant difference :)
22:52:02Slasherihehe :)
22:57:03lostlogicthe problem with this kind of testing is that I end up making myself sick of whatever song I choose as a test track for the day :(]
22:57:51preglowgood excuse to look up new music
22:58:48lostlogicSlasheri: is most of this audio engine all your code?
22:59:12Slasherilostlogic: playback.c and pcmbuf.c mostly at least
22:59:13lostlogicthere's some really impressive stuff in here.
22:59:18***Saving seen data "./dancer.seen"
22:59:19Slasherihmm :D
22:59:56Slasherijust found a way to remove very little tiny pop when seeking
23:10:30linuxstblostlogic: Thanks for removing my accidental Makefile commit.
23:10:50Paul_The_Nerdlinuxstb: Consequently, that's the one that caused my Makefile to conflict. :-P
23:11:14*linuxstb bows head in shame
23:11:47linuxstbGood job the build system doesn't test "make clean".
23:11:50Paul_The_NerdOddly enough, it's because I'd added mpeg2 as well.
23:11:58lostlogiclinuxstb: yeah, I was surprised it didn't
23:12:54 Join akaidiot [0] (
23:13:38Bagdersimply doing rm -rf * is way faster
23:14:11*Paul_The_Nerd agrees.
23:14:15lostlogicrm -rf [b-z]* apps if you don't want to kill autoconf.h and Makefile
23:14:30linuxstbBut it's not good for sims if you've set up a nice archos directory.
23:14:30Bagderthe build system always do configure
23:15:22 Quit DangerousDan (Read error: 110 (Connection timed out))
23:15:27linuxstbBagder: Can I interest you in my bmp2rb build-system patch?
23:16:14linuxstbCheck the logs from a couple of hours ago.
23:16:18 Nick Thus0__ is now known as Thus0 (
23:17:34linuxstbIt's not complete (it only deals with native bitmaps for the main LCD), but if my approach seems sensible, then I would like to commit it as it is now.
23:18:08linuxstbSo bitmaps for the remote, and mono bitmaps used by all targets are still in the .c files.
23:19:21Bagderstill a first step!
23:19:49linuxstbYep, and it deals with the biggest bitmaps. The ones remaining in the source are relatively tiny.
23:20:47linuxstbThe only thing we need to get right is the directory structure. I'm thinking that we need extra directories (or subdirectories in the bitmaps/ directories) to deal with the different formats.
23:22:40Bagderyou mean if a single bmp is converted to more than one format?
23:23:34linuxstbThat's the problem. Not only could it be a different format, but it could need a different name. The example is the logo - we have a mono logo for the archos devices, and the same mono logo for the iriver remote.
23:24:07linuxstbWhen that bitmap is used on the archos, we want it called "rockboxlogo", but when it is used for the remote, it should be something like "rockboxremotelogo"
23:24:31Paul_The_NerdYou could create a prefix setup.
23:24:54Paul_The_NerdWidth_Height_Depth_Name, so they aren't named for their platform, but what screen they fit?
23:25:01linuxstbI'm starting to think that it would be cleaner to just have multiple copies of the .bmp files, rather than making the build system too complex.
23:25:23BagderI agree
23:25:28linuxstbPaul_The_Nerd: Yes, that's how the .bmp files themselves are named. But they need to be converted to a common name for use in Rockbox.
23:26:13linuxstbi.e. we have rockboxlogo.160x57x2.bmp, rockboxlogo.112x37x1.bmp and they are both converted to rockboxlogo[] in the .c file
23:27:19Paul_The_NerdThe problem is that you'd use the 112x37x1 logo for the remote?
23:27:19BagderI think your current effort is a good approach and should be committed
23:27:54Bagderthe irivers could use a somewhat bigger logo for the remote anyway ;-)
23:28:25 Quit eEye (Read error: 104 (Connection reset by peer))
23:29:02linuxstbSo we should extend it to have three bitmap directories - mono, native and remote - and not worry if the same .bmp files are duplicated?
23:30:00Paul_The_NerdI kinda like that way.
23:30:04Bagderisn't it just remote and main?
23:30:48Paul_The_Nerdlinuxstb: What about, say, color remotes or other things?
23:31:10linuxstbNo - we have mono bitmaps (i.e. the icons) which should always be mono, to allow for transparency and change of fg colour.
23:31:18Bagderah, right
23:32:08Bagderso mono native and remote it is
23:32:09linuxstbPaul_The_Nerd: Yes, colour remotes will simply be the addition of a fourth directory when that time comes.
23:32:23Bagderor having the remote ones be in color
23:33:16linuxstbAny preferences for directory structure? Should it be bitmaps_mono, bitmaps_native and bitmaps_remote or bitmaps/mono, bitmaps/native and bitmaps/remote ?
23:33:31BagderI like bitmaps/mono
23:33:39Paul_The_NerdI like bitmaps/blah too
23:33:55linuxstbSo do I (even though it means adjusting my patch).
23:34:31linuxstbAnd I guess the same for the plugin bitmaps...
23:37:14linuxstbOK, I'll extend the patch to do that.
23:41:09 Quit linuxstb ("CGI:IRC")
23:42:08 Join linuxstb [0] (
23:43:30 Join BHSPitLappy [0] (
23:43:32amiconnBagder: There are potentially 4 different bitmap formats - the remote may have a different mono format than the main
23:44:15amiconnThis will very probably happen for H300
23:44:53linuxstbYes, we can have remote_mono and remote_native directories.
23:45:13linuxstbAre you happy with the approach though?
23:46:04*Bagder goes to bed
23:46:05amiconnHowever, in case of the H300 remote, native==mono
23:46:07 Quit Bagder ("Off to search for that connect-resetting peer guy!")
23:47:22Moosamiconn: and potentially the future iaudio remote 128*96
23:50:04amiconnlinuxstb: How does your bmp patch work?
23:50:56amiconnI think that we might be able to go without separate source bitmaps for different depths (mono vs. native), we'll probably only need different sizes
23:51:28linuxstbDownload this file:
23:51:28amiconnThe difference will be the switch used when invoking bmp2rb
23:52:02linuxstbThis is the apps/bitmaps directory (and Makefile and SOURCES) and also the apps/plugins/bitmaps/ directory. The actual changes to the build system probably aren't what you're interested in.
23:52:31preglowautomatically converting from colour to mono wont always be pretty
23:52:33linuxstbI am planning to extend that to have subdirectories for the different types of bitmaps - mono, native, remote_mono and remote_native
23:52:50linuxstbEach of those four subdirectories will be processed with their own bmp2rb command.
23:55:21amiconnlinuxstb: Where is that ?
23:55:43linuxstbJust switching computers - brb
23:55:46 Quit linuxstb ("Leaving")
23:56:46 Join linuxstb [0] (

Previous day | Next day