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

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

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

#rockbox log for 2005-12-22

00:01:15 Join gromit` [0] (n=gromit`
00:03:08 Join jaebird [0] (
00:03:12Mongey|was wondering if this plugin could be optimised for the h300
00:06:32Jungti1234Why may not I become connection to sourceforge?
00:08:23 Quit Paul_The_Nerd ("Leaving.")
00:14:43Jungti1234oh alarm?
00:15:03preglowlinuxstb_: so, you got some sleep code coming up, or?
00:16:34mirakthe plugin is now 450ko and fit with the standart plugin ram size preglow
00:17:05mirakmaybe having it display something would attract some people to improve it
00:17:12preglowoh yes
00:17:14preglowthat's for sure
00:18:19Mongey|so...... can it be done
00:19:01linuxstb_preglow: Yep, just about to commit now. It's just a simple write to one register in the pcf50605.
00:19:15preglowlinuxstb_: btw, am i the only one getting the odd glitch with the clichwheel
00:19:32preglowsometimes when i scroll, the cursor will skip back one notch
00:19:37preglowbefore continuing as usual
00:19:38linuxstb_Yes, I've seen that.
00:19:42preglowhmm, ok
00:19:45preglowi wonder why that is
00:19:48 Quit Jungti1234 ("bye")
00:19:57preglowi did remove some code from the ipl driver that i just didn't see the need for
00:20:09nathanhpreglow: how close is the ipod nano port... you got reliable audio?
00:20:18preglownathanh: no, but audio
00:20:27preglowlinuxstb_: i see it now, it's a crossover issue at the top of the wheel
00:20:44preglowbah, i'm too stupid
00:21:47peturand another patch: H300 brightness - see
00:22:08preglownathanh: linuxstb has hacked in preliminary audio
00:22:15linuxstb_We don't seem to have a button combination mapped to shutdown though. Guess I need to enable the HAVE_SW_POWEROFF define.
00:22:16preglownathanh: so we get skippy audio
00:24:03mirakcrap I think I removed some usefull things
00:26:20 Join JoeBorn [0] (
00:28:56lostlogicxMy latest Tremor optimizations: If anyone wants to test boost levels between CVS and that patch, it'd be helpful to verify that profiling results match real world results.
00:30:16Moosprobably Lear
00:30:49Mongey|lostlogicx;would you be able to alter that plugin
00:32:35linuxstb_preglow: Just committed.
00:33:15preglowi'll do a fix for the wheel problem now
00:33:17preglowthen i've got to go
00:33:23 Join bazz [0] (
00:34:24linuxstb_It's nice to finally be able to switch the ipod off.
00:34:41bazzso, i'd like to check in/submit what i have of the sdl sim. it works on linux and mostly on windows, although there needs to be some hackery to configure and the makefiles to make it build all the way on windows. it doesn't step on the win32 or x11 sim, so no functionality is lost. how would/should i go about doing this?
00:35:00linuxstb_Submit a patch to the patch tracker.
00:35:24peturOT: any tip for capturing and converting a RealAudio stream under windows?
00:35:27bazzlinuxstb_: does that allow for all the new files i need to add?
00:35:33 Quit RotAtoR (Read error: 110 (Connection timed out))
00:36:49linuxstb_Yes, patches can include complete new files. I think you need the -N option to diff
00:37:13bazzlinuxstb_: nod, i'll figure it all out
00:37:19linuxstb_I'm not an expert, but I'm sure someone here can help with the necessary diff command if you have problems.
00:37:57linuxstb_So we'll have three sims with your patch? X11, win32 and sdl?
00:38:04bazzlinuxstb_: yes
00:39:24nathanhive already got an sdl sim patch up on sourceforge
00:39:45preglowthen now we've got two!
00:40:10bazznathanh: ohh, no one told me :) does it support all the players and sound and stuff?
00:40:17nathanhnot sound, but all the players yes
00:40:29 Quit ender` (" Theorem: a cat has nine tails. Proof: No cat has eight tails. A cat has one tail more than no cat. Therefore, a cat has n")
00:40:30nathanhi dont have a soundcard so i wrote the code, but i cant test it, apparently sound didnt work
00:41:12bazzhrmm, okay, i wonder if it's worth me sumbitting the patch then
00:41:16nathanhit sure is
00:41:22preglowdo so
00:41:24nathanhi'll nick all the best bits of yours, and you'll nick all the best bits of mine
00:42:17bazzokay, so did you just cvs diff for the stuff you changed in existing code and then zip up all the new files?
00:42:36nathanhoriginally yes, but then i used "cvs diff -u" followed by "diff -uN >>"
00:42:41nathanhto get a single patch file
00:42:42 Quit perplexity (Read error: 104 (Connection reset by peer))
00:43:11nathanhi would like to see how you handled the surface zooming problem
00:43:15nathanhmy solution is horribly inefficient
00:43:38bazzmine doesn't do surface zooming :(
00:43:56nathanhwell thats ok, because my zooming is awfully slow
00:44:06nathanhi think we need to butt heads and work out how to do it better
00:44:36bazzmine is like the windows sim in that is draws the image of the player, so zooming is tricky
00:44:46nathanhooo, i dont have the images of the players
00:44:52nathanhbut i wanted that feature
00:45:09nathanhi thought one way to do it would be to use an offscreen surface which is zoom=1
00:45:15bazzthey are in the win32 sim dir
00:45:22nathanhthen use one of the bitmap drawing sdl functions to zoom for us
00:46:01bazznod, i didn't realize this was a desired feature, i'll start thinking about it
00:46:21*nathanh keen to see bazz's patch :-)
00:46:36bazzit'll be submitted soon
00:48:45 Join perplexity [0] (
00:50:17preglowlinuxstb_: so how do i switch it off now?
00:50:44linuxstb_You can't yet. But the idle poweroff works.
00:51:03linuxstb_I'm just testing it now on the 5g.....
00:51:07preglowhow does retailos go to sleep, again?
00:51:10miraknathanh: what's that zoom thing ?
00:51:17linuxstb_preglow: Long press on play
00:51:30nathanhmirak: ./rockboxui −−zoom 3
00:51:51linuxstb_I've mapped that to stop in the wps, so I'm thinking the same combination for poweroff
00:52:09 Join dropandho [0] (n=dropandh@
00:52:12miraknathanh: what do you use to zoom ?
00:52:21dropandhosup all!
00:52:34nathanhmirak: what do you mean
00:52:41miraknathanh: the algorithm
00:52:51mirakthe UI is borken on my computer
00:52:56nathanhread the patch :-)
00:53:46 Join YouCeyE [0] (
00:53:50 Quit mirak (Remote closed the connection)
00:56:52bazznathanh: so, you checked out the cvs code somewhere and then diff -uN mycodedir cvscodedir?
00:57:20nathanhbazz: thats right
00:58:39bazzif you cvs diff and then diff -uN doesn't that create entries twice for changed existing files?
00:59:48nathanhmy second diff was only on uisimulator/sdl
00:59:58bazznathanh: clever :)
00:59:58nathanhbecause i didnt really have a full cvs checkout
01:01:04linuxstb_Mmm. Seems the ipod is the only target with software power-off, but no OFF button...
01:01:10 Part LinusN
01:01:44preglowlinuxstb_: testing it out now
01:02:05linuxstb_I'm relieved to report it works fine on the 5g as well.
01:02:21preglowyeah, saw
01:02:41preglowhow do you open these bloody buggers anyway?
01:02:45preglowthere are no screws, no nothing
01:02:54nathanhyou can fix everything with a hammer
01:03:02preglowi can destroy anything with one
01:03:13linuxstb_Guitar picks do the job, or so I've read.
01:03:22*preglow tries
01:04:12preglowseems jazz 3 is a bit too thich, heh
01:04:58preglowlinuxstb_: poweroff works, lcd dims gradully over about thirty secs
01:05:11preglowlinuxstb_: where it first gets corrupted, just like if i had reset it
01:06:21 Join NicoFR [0] (
01:06:23linuxstb_But does it power back on?
01:07:35preglowapple logo and all
01:07:46preglowwhich reminds me, that doesn't happen for retailos, does it?
01:08:15linuxstb_Not immediately. It goes into a "light sleep" first, and then a deeper sleep after a long time in light sleep.
01:08:23preglowahhh, alright
01:08:33preglowwhich resets it completely, just like this, yes?
01:09:04linuxstb_The light sleep probably just powers off the peripherals and slows the CPU down to the minimum.
01:09:29preglowguess who's going to entertain himself with bejeweled during the flight
01:09:57linuxstb_Sorry I couldn't give you music to go with your bejeweled...
01:10:06preglowi'll be alright
01:10:38preglowwith some luck, i might even survive the ordeal
01:11:50preglowi wish i knew what the cache issue was
01:12:08linuxstb_When you disconnect your ipod from Windows, does it reboot?
01:12:25linuxstb_In Linux, it stays in disk mode.
01:12:31preglowno, doesn't
01:12:41preglowit stays in disk mode, unless i rewrote the boot part
01:13:07linuxstb_Yep, I read someone say on #ipl that it will reboot if you write to the first 32MB of the disk.
01:14:02preglowman, these things really do scratch a bit too easily for my tastes
01:14:13preglowi've been careful with the bugger, and the backlight doesn't look like shit
01:15:19*petur tries to find a scratch on his H340
01:15:48preglowmy h120 looks realy nice
01:18:39 Part PaulJ
01:20:28preglowbut i've got to start thinking about sleep
01:20:56nathanhmy h340 is made out of scratchidium, the most scratchable element in the universe
01:21:06nathanhi reckon the only holding it together is the crisscross pattern of scratches
01:21:27linuxstb_preglow: Any objections to a long press on play for the power-off?
01:21:35preglowlinuxstb_: none
01:23:00 Join HypnoticMonkey [0] (
01:23:15preglowbut ok
01:23:19preglowi've gotta go
01:23:25preglowbye all
01:23:49Rob-How far off is recording from line-in / external mic on the h340?
01:24:52nathanhi heard it was working already ?!
01:24:52peturis available
01:28:41 Quit Moos ("Rockbox?... The best ;)")
01:29:28peturWTF? did a rebuild after CVS update and recording no longer works - could be my patch :( - testing CVS build now
01:29:47NicoFRis it normal the peak meter doesn't seem to work anymore in recording mode on h300 ?
01:30:30peturno - I'm seeing the same thing
01:31:23NicoFRand also it crashes when I stop a recording
01:31:46peturone moment, testing bleeding-edge build to rule out my stuff
01:33:00peturAAARGH... something has broken recording :(
01:33:14peturmaybe it's the LCD DMA thing?
01:33:51NicoFRor the fix that enabled recording from line in ?
01:34:18peturnah, that just toggles the mux pin
01:34:36peturie selects the correct analog input
01:34:50peturinternal mic doesn't work as well
01:35:38 Join Nepbaland [0] (
01:37:05Rob-I'm currently using daily 20051220. I can record with the internal mic, but not line in.
01:37:47peturLinus fixed that on that date in the evening
01:38:49peturI'm downloading the previous LCD driver to check if that caused it...
01:40:42dropandho"it wasn't me"
01:43:21peturyep that's it
01:43:34peturworks just fine with the old LCD driver
01:46:13dropandhothis is 320 or 120/140?
01:46:48peturH1xx should be fine
01:46:54NicoFRhow come the LCD driver can brake recording ?
01:47:33peturbecause it now does DMA, and so does the recording...
01:47:46***Saving seen data "./dancer.seen"
01:48:14peturhave sent it on the dev mailing list, I hope Linus picks it up soon
01:49:02 Join mirak [0] (
01:49:05 Quit YouCeyE (Read error: 54 (Connection reset by peer))
01:49:48NicoFRsalut mirak
01:50:07peturgot to get some sleep
01:50:21NicoFRgood night
01:50:25 Quit petur ("here today, gone tomorrow")
01:50:39 Join San [0] (n=test@
01:52:06NicoFRmirak c'est toi "ilaglisser" sur le forum HFR ?
01:53:01mirakNicoFR: shut
01:53:26NicoFRpourquoi ?
01:53:43mirakt'es moderateur sur HFR ?
01:54:01mirakje me suis fait ban d'un sous forum
01:54:41mirakenfin sur un topic je me suis pris le choux avec un modo
01:54:50mirakenfin c'etait en pv plutot
01:55:09mirakNicoFR: pkoi ?
01:55:24NicoFRj'ai vu que tu parlais de ton plugin qui "lit" les XviD
01:56:18mirakj'ai reussit a le reduite
01:56:38mirakmais y a des structures de donées plutot grosses defineies en static
01:56:49NicoFRcd il lit plus vite ?
01:56:50mirakgenre 250ko pour un tableau ça fait un peu big !
01:56:56miraknon du tout
01:56:56NicoFRah oui
01:57:03mirakc'est pour le faire tenir dans rockbox
01:57:06miraklà ça tient
01:57:25mirakc'est pas pres d'aller plus vite je pense
01:57:32mirakil faudrait de l'assembleur
01:57:55miraksi ça te tente ...
01:58:32NicoFRpas trop non... dj que je suis pas l'aise en C...
02:00:32mirakje pense pas qu'il puisse faire mieux que ce qu'on fait iriver ce plugin
02:00:39mirakc'est des malades
02:00:49miraklà c'est 2fps
02:00:52miraklaisse beton
02:01:07mirakje voulais voir bah on est fixé maintenant
02:01:32mirakl'idée d'attendre 25fps est vraiment tres éloignée
02:01:58mirakirréaliste en fait
02:01:59NicoFRton plugin il affiche les images dcodes l'cran ?
02:02:10miraknon que dalle
02:02:16mirakil peut sauver en fichier
02:02:35mirakça serait pas dur a faire mais 2fps c'est vraiment pas interessant
02:02:38dropandhowhat happened here?!
02:03:07NicoFRje me demandais s'il tait pas ralenti par le driver LCD qui est encore lent
02:03:13mirakenfin y a aucune optimisation en même temps
02:03:18mirakpas d'iram
02:03:23mirakpas d'assembleur
02:03:40mirakmais bon la vidéo fait juste 270x176 en plus
02:03:49lostlogicdropandho: french happened, apparently :-\
02:04:07mirakI will try to put the image on screen to attract some people
02:04:07dropandhooui oui!
02:04:54mirakit's him who started to speak french −−-> dropandho . That's not my fault
02:04:56NicoFRwe were talking about his xvid decoder
02:05:00mirakhem NicoFR
02:05:05NicoFRsorry about that
02:06:49NicoFRmaybe others who know ASM will be intersested and will look into it
02:07:04NicoFRand maybe one day it'll reach 10 fps :D
02:07:10dropandhomirak- it's ok...i was being dramatic!
02:07:21dropandhoso what needs to be looked into? i know someone who knows ASM
02:07:46mirakwell first, get xvidcode
02:07:49*lostlogic knows asm, but is not planning to leave Tremor-land for the moment ;)
02:07:49NicoFRthe xvid decoder needs optimisation
02:07:51NicoFRa lot
02:07:52mirakxvidcore source
02:08:22mirakyou will see where are localised the asm optimisation for x86
02:08:32NicoFRbtw what is tremor exactly ?
02:08:38NicoFRa decoder ?
02:08:38mirakthat's probably where we should start to convert to coldfire
02:08:42mirakNicoFR: vorbis
02:09:21 Quit miner49er ("using sirc version 2.211+KSIRC/1.3.11")
02:09:23dropandhoi saw we got a 2-3 % cpu gain
02:09:32dropandhoso that means more batt life on ogg- ya?
02:10:38dropandhowe dont have any complete batt stats these days- do we?
02:10:51dropandholike stuff comparing running it with the remote and all that
02:11:19NicoFRI don't think so
02:11:33NicoFRand is anyone working on a color JPEG viewer ?
02:13:48dropandhonot that i know of
02:14:05Mongey|for what?
02:14:22dropandhowe really should update the todo page
02:14:38dropandhocause i am pretty fuzzy as to why we are waiting on an offical iriver release
02:14:47lostlogicwe actually need a runtime battery stat collector to recreate battery runtime estimates.
02:15:00 Quit HypnoticMonkey ("CGI:IRC")
02:15:49bazznathanh: request id 1387679 is my patch
02:16:34NicoFRoh and what is sdl ?
02:16:46Mongey|will someone edit a plugin? the alarm one that linked earlier
02:17:40Mongey|? please
02:18:52Mongey|:( anyone?
02:18:54dropandholostlogic is that a known need or is someone workin on that?
02:19:03NicoFRtime to sleep
02:19:05NicoFRgood night
02:19:18 Quit NicoFR ()
02:19:18dropandhoi think it would be a great benchmark so we can see if any work people do helps or hinders the bat life
02:19:33dropandhoand can offer a nice comparison to iriver
02:20:29lostlogicdropandho: it's a known feature of the old firmware that we don't have... don't think it's something epople are working on though.
02:21:19dropandhoi feel like we really need someone to get in and update-
02:21:26dropandhothe release roadmap
02:21:36dropandhoso these things are out on the table
02:21:49lostlogicyou have a wiki account, right? :-P
02:21:52linuxstb_dropandho: Go ahead :)
02:21:59Mongey|dropandho; for what dap?
02:22:49lostlogicI'm pretty happy though, with my opts and today's patch by Linus for the LCD update speed, I'm down to about 41% boost playong my Q7 oggs
02:22:49dropandhoiriver 120/40 in specific
02:23:01lostlogicand 320 and 340!
02:23:03dropandhoi can setup a wiki account...i just dont know these low level things
02:23:07lostlogicmost of the stuff is common
02:23:32dropandhoi am not in the i cant adjust these items
02:24:34Mongey|lostlogic; hear what happened to misticjeff
02:24:41lostlogicMongey|: no, what happened???
02:24:52Mongey|#misticriver, his mam
02:25:18lostlogic*nod* sucky.
02:26:42linuxstb_Mongey|: I personally don't think that plugin is the way to go - I would prefer to implement a sleep/wakeup alarm in the core of Rockbox.
02:28:13Mongey|linuxstb_; yes that would be nice in the future but since that plugin is there and could be short term or w/e
02:28:42dropandhohow come the website doesn't flaunt how many people have downloaded various build or might help to build confidence
02:30:10miraklinuxstb_: I am trying to write to lcd, what I do is just :
02:30:25mirakthat should work, right ?
02:32:28linuxstb_No - BPP will be 16 and the number of bytes is XDIM*YDIM*2
02:32:49linuxstb_Unless you meant bytes-per-pixel
02:32:56linuxstb_In which case, yes, it should work.
02:33:04linuxstb_What's the problem?
02:33:40mirakno BPP is set to two
02:33:52mirakI test a bit more
02:34:15 Quit San (Read error: 110 (Connection timed out))
02:34:58linuxstb_Have you set xvid_dec_frame.output.csp to XVID_CSP_RGB565 ?
02:35:22mirakmy god it works
02:35:26miraklinuxstb_: yes
02:35:49mirakThat's awsome
02:35:52Mongey|mirak; what does it do
02:36:01mirakan awsome first step
02:36:11Mongey|for videos?
02:36:18mirakit reads an mpeg4 stream
02:36:25mirakand outputs it on screen
02:36:34mirakat the incredible frame rate of 2
02:36:47Mongey|nice one
02:40:15 Join ehntoo [0] (
02:45:47Bluechiphello, seems my irc connection still logs into rockbox by default - lol
02:46:15BluechipI see the ipod guys are in's it going?
02:49:05ehntooblech. the screen just cracked on my H120
02:49:06linuxstb_It's going well. We're quite close to getting audio playback working.
02:49:18ehntooanybody have some advice?
02:49:50Bluechiplinuxstb_: well done guys - what about the interface - or was that just simple code tweaking?
02:49:55ehntooI just put $140 into it a couple months ago for a new battery and a new harddrive, even
02:50:14lostlogictime to install my ebay battery!
02:50:18Bluechipehntoo: do newmp3technology do iriver parts?
02:50:40ehntooBluechip, I have no idea
02:50:41 Quit Nepbaland ("I-n-v-i-s-i-o-n 2.0 Build 3515")
02:51:09Bluechipehntoo: then that is my suggestion - check their site and if they do, buy one and replace the borken one :)
02:51:11mirakMongey|: ?
02:51:24linuxstb_Bluechip: If you're talking about the UI, then it was just a matter of adding ipod-specific button mappings everywhere. Plus of course the 16-bit colour LCD code, but amiconn did most of the work on that, and it's shared with the h300.
02:51:58Bluechiplinuxstb_: fantastic ...amiconn is a worthy alli in such matters
02:52:02ehntoocurse our ISP's dns servers. It's been taking 10-30 seconds for lookups
02:52:02mirakI am a bit disapointed by the cpu power, I though H300 had more gas than that. My powermac 120mhz can read that xvid stream on linux without a problem ;p
02:52:35Mongey|mirak; so it plays movies
02:52:57Mongey|very slowly but it does, right?
02:54:30ehntoohrm... no luck for replacement iriver stuff at newmp3technology.
02:54:36lostlogicmirak: so why is the 124mhz coldfire so much slower than your powermac 120?
02:56:18miraklostlogic: the cache maybe
02:56:30mirakthe H300 drive is way faster on the opposite
02:56:44Bluechippipeline? number of transistors used per op?
02:56:47lostlogichmph, nod
02:57:11mirakthe mac have a 3G scsi hard drice that's doing a hell of a noise, and slow as well
02:57:52mirakthe ppc is risc. It's supposed to be as fast a intel 200mhz probably
02:58:18mirakproblem also is maybe the slow ram and iram stuff
02:58:35mirakdon't know
03:01:09 Quit mirak ("Ex-Chat")
03:25:54 Quit Bluechip ()
03:26:54 Quit jaebird ("Leaving")
03:29:05 Join DJDD_ [0] (
03:38:28 Quit dropandho ()
03:41:22 Join YouCeyE [0] (
03:46:57 Join DrumRBoy32O [0] (
03:47:50***Saving seen data "./dancer.seen"
03:57:28 Quit DrumRBoy32O ()
03:57:54 Quit bazz (Read error: 113 (No route to host))
04:03:59 Join XavierGr [0] (n=XavierGr@
04:09:17 Quit ehntoo ("Leaving")
04:09:37 Join San [0] (
04:14:19 Join ehntoo [0] (
04:39:01 Join TCK [0] (n=tckocr@
04:46:12 Join RotAtoR [0] (
04:55:29 Quit ehntoo ("Leaving")
04:55:54 Quit TCK ("I shall not cease!")
05:05:36 Join Zak1392 [0] (
05:07:27Zak1392hey guys
05:10:04Zak1392what's everyone up to?
05:11:01Zak1392not much, just got out of the shower
05:12:28Zak1392just read about video playback
05:12:32Zak13922 fps
05:13:36Zak1392nice ;)
05:13:42Zak1392to bad mirak isn't here
05:16:44 Join grantashman [0] (
05:16:52grantashmanHey all.
05:17:32grantashmanWhat's happening?
05:18:02Zak1392not much
05:19:14grantashmanI read video is coming along?
05:19:27grantashmanBaby steps :)
05:20:17 Quit Rob2222 (Read error: 110 (Connection timed out))
05:23:19grantashmanIt's a start.
05:23:29grantashmanNot that video is the most important feature, still nice though.
05:30:52 Join Rob2222 [0] (
05:37:58 Quit San (Read error: 110 (Connection timed out))
05:38:16 Quit Mongey| ()
05:39:13 Quit grantashman ("Leaving")
05:47:55***Saving seen data "./dancer.seen"
05:53:51 Quit Zak1392 ()
06:26:02warewolfOh, anyone know if rockbox will support whatever sort of RTC is in the iRiver H300 series?
06:26:27warewolf(why is it that certian iRiver firmwares don't support the clock, and some do? *boggles*)
06:41:41 Join nathanh_ [0] (
06:56:46 Join pabs [0] (
07:02:36 Quit perplexity (Read error: 60 (Operation timed out))
07:03:20 Join Jungti1234 [0] (n=jungti12@
07:05:50bagawkJungti1234: hello :)
07:05:58Jungti1234hi bagawk
07:10:28Jungti1234new iriver
07:10:33 Join |u|smooth [0] (
07:12:49 Join San [0] (n=test@
07:14:05 Quit XavierGr (
07:14:05 Quit Jungti1234 (
07:14:05 Quit akaidiot (
07:14:05 Quit nathanh (
07:14:05 Quit solexx (
07:14:05 Quit mbr (
07:14:05 Quit Ismo (
07:14:05 Quit RotAtoR (
07:14:05 Quit YouCeyE (
07:14:05 Quit DJDD_ (
07:14:05 Quit JoeBorn (
07:14:05 Quit Slasheri (
07:14:05 Quit |u|smooth (
07:14:05 Quit nathanh_ (
07:14:05 Quit Rob2222 (
07:14:05 Quit gromit` (
07:14:05 Quit San (
07:14:05 Quit Rob- (
07:14:05 Quit Rick (
07:14:05 Quit ze (
07:14:05 Quit `3nergy (
07:14:05 Quit phaedrus961 (
07:14:05 Quit bagawk (
07:14:05 Quit ghode|afk (
07:14:05 Quit novimon (
07:14:05 Quit elinenbe (
07:14:05 Quit Nilisco (
07:14:05 Quit t0mas (
07:14:05 Quit dwihno (
07:14:05 Quit goa (
07:14:05 Quit merbanan (
07:14:05 Quit ashridah (
07:14:05 Quit lostlogicx (
07:14:06 Quit crash__ (
07:14:06 Quit linuxstb_ (
07:14:06 Quit amiconn (
07:14:06 Quit pill (
07:14:06 Quit actionshrimp (
07:14:06 Quit edx__ (
07:14:06 Quit thegeek (
07:14:06 Quit Vladoman (
07:14:06 Quit Nibbler (
07:14:06 Quit vHs_ (
07:14:06 Quit DreamTactix291 (
07:14:06 Quit Sando (
07:14:06 Quit Strath (
07:14:06 Quit Bger (
07:14:06 Quit Lynx_ (
07:14:06 Quit crwl (
07:14:06 Quit Seed (
07:14:06 Quit slowcoder (
07:14:06 Quit AciD (
07:14:06 Quit wubbla (
07:14:06 Quit CoCoLUS (
07:14:06 Quit Bagder (
07:14:06 Quit Kingstone (
07:14:22DEBUGEOF from server (Success) (snapshot: netstuff.c line 545)
07:14:22***Saving seen data "./dancer.seen"
07:14:22***Started Dancer V4.16
07:14:22DEBUGconnect(2) failed on socket 3 (Connection refused) (snapshot: netstuff.c line 150)
07:14:22***Connected to on port 6667
07:14:22***Logfile for #rockbox started
07:14:43***Server message 501: 'logbot :Unknown MODE flag'
07:14:43Mode"logbot :+i" by logbot
07:14:44 Join logbot [0] (n=bjst@
07:14:44 Join ashridah [0] (n=ashridah@
07:14:44 Join Bagder [0] (n=daniel@
07:14:44 Join linuxstb_ [0] (n=linuxstb@
07:14:44 Join Paprica [0] (
07:14:44 Join edx [0] (
07:14:48 Join Seed [0] (i=ben@
07:14:58 Join Bger [0] (n=Bager@
07:15:05 Join Lynx_ [0] (n=lynx@
07:15:06Mode"#rockbox +t-s " by
07:15:06 Join XavierGr [0] (n=XavierGr@
07:15:06Mode"#rockbox +bbb *!* *!* *!*user@82.67.68.* " by
07:15:06Mode"#rockbox +bbb *!*Yu-Gi-oH@* *!*jirc*@* *!* " by
07:15:06Mode"#rockbox +bbb *!*icechat*@* *!*ju_@* *!*@* " by
07:15:06DEBUGParse error in OnTopic(from = "", line = "#rockbox :Seriously, we DO NOT estimate dates. You'll find out when it happens.") (snapshot: server.c line 299)
07:15:07 Join Nibbler [0] (n=sven@
07:15:12 Join |u|smooth [0] (
07:15:12 Join nathanh_ [0] (
07:15:12 Join Rob2222 [0] (
07:15:12 Join gromit` [0] (n=gromit`
07:15:19 Join lostlogic [0] (n=lostlogi@
07:15:25 Join Slasheri [0] (
07:15:25 Join AciD [0] (n=gni@
07:15:27 Join merbanan [0] (i=banan@
07:15:46 Join San [0] (n=test@
07:15:46 Join Rob- [0] (
07:15:46 Join Rick [0] (i=rick@unaffiliated/Rick)
07:15:46 Join ze [0] (
07:15:46 Join `3nergy [0] (
07:15:46 Join phaedrus961 [0] (
07:15:46 Join bagawk [0] (n=lee@unaffiliated/bagawk)
07:15:46 Join ghode|afk [0] (
07:15:46 Join novimon [0] (
07:15:49 Quit merbanan (Nick collision)
07:15:49 Quit ashridah (Nick collision)
07:15:49 Join elinenbe [0] (
07:15:49 Join Nilisco [0] (
07:15:49 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
07:15:49 Join dwihno [0] (n=dw@
07:15:49 Join goa [0] (
07:15:49 Join merbanan [0] (
07:15:49 Join crash__ [0] (
07:15:49 Join lostlogicx [0] (
07:15:49 Join ashridah [0] (
07:15:57 Join Jungti1234 [0] (n=jungti12@
07:15:57 Join akaidiot [0] (
07:15:57 Join nathanh [0] (
07:15:57 Join solexx [0] (
07:15:57 Join mbr [0] (n=mb@
07:15:57 Join Ismo [0] (i=laitinei@
07:16:11 Quit ashridah (
07:16:11 Quit lostlogicx (
07:16:11 Quit crash__ (
07:16:11 Quit merbanan (
07:16:11 Quit dwihno (
07:16:11 Quit elinenbe (
07:16:11 Quit Nilisco (
07:16:11 Quit t0mas (
07:16:11 Quit goa (
07:16:17 Join Kingstone [0] (n=sg1@
07:16:25NJoinlostlogicx [0] (
07:16:25 Quit novimon (
07:16:25 Quit ghode|afk (
07:16:25 Quit bagawk (
07:16:25 Quit phaedrus961 (
07:16:25 Quit `3nergy (
07:16:25 Quit Rick (
07:16:25 Quit Rob- (
07:16:25 Quit San (
07:16:25 Quit ze (
07:16:27 Join CoCoLUS [0] (
07:16:31NJoinze [0] (
07:16:53 Join RotAtoR [0] (
07:16:53 Join YouCeyE [0] (n=YouCeyE@unaffiliated/youceye)
07:16:53 Join DJDD_ [0] (
07:16:53 Join JoeBorn [0] (
07:17:05NJoinNilisco [0] (
07:17:05 Join phaedrus961 [0] (n=bob@
07:17:06 Join warewolf [0] (i=warewolf@
07:17:07 Join amiconn [0] (
07:17:07***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! ( )'
07:17:09 Join Strath [0] (
07:17:11 Join thegeek [0] (
07:17:11NJoin`3nergy [0] (
07:17:15 Join pill [0] (
07:17:15NJoinSan [0] (n=test@
07:17:17warewolfcan freenode please stop breaking their !@#$ network
07:17:21NJoinelinenbe [0] (
07:17:26 Join slowcoder [0] (
07:17:29 Join bagawk [0] (
07:17:29NJoindwihno [0] (n=dw@
07:17:30 Join DreamTactix291 [0] (
07:17:32***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! ( )'
07:17:34NJoinghode|afk [0] (
07:17:39 Join actionshrimp [0] (
07:17:40 Join Rick [0] (
07:17:42***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! ( )'
07:17:47NJoint0mas [0] (n=Tomas@unaffiliated/t0mas)
07:17:49NJoingoa [0] (
07:17:52 Join ashridah_ [0] (
07:17:55 Join banan_ [0] (
07:18:44 Join Vladoman [0] (
07:19:27 Join preglow [0] (
07:19:31 Join pabs [0] (
07:19:31 Join Naked [0] (
07:19:34 Join crash [0] (
07:19:36NJoinnovimon [0] (
07:19:44 Nick Naked is now known as Hadaka (
07:19:45 Quit |u|smooth ("( :: NoNameScript 3.81 :: )")
07:19:48 Join vHs [0] (
07:19:50 Join crwl [0] (
07:19:55 Join wubbla [0] (
07:20:37bagawkHadaka: why do you change eng-jap?
07:22:26 Join |u|smooth [0] (
07:24:07 Quit Rob2222 ()
07:28:14 Quit RotAtoR ("Leaving")
07:32:03 Nick ashridah_ is now known as ashridah (
07:32:03 Quit nathanh_ ("CGI:IRC (EOF)")
07:45:10Bgermorning :)
07:51:21Jungti1234morning, Bger :)
07:52:35Jungti1234How are you?
07:53:26Bgersleepy :P
07:54:02Jungti1234I had gone to the Supreme Court today.
07:54:20 Join perplexity [0] (
07:54:23Bgerwow, why :)
07:54:27Bgerperplexity :) morning
07:54:36 Quit San (Read error: 110 (Connection timed out))
07:55:06Jungti1234To watch.
07:55:48Bgerto watch a suit ?
07:57:10Jungti1234Learned by observation.
07:58:02Bgeraah, okay
08:04:00 Join Sando [0] (
08:05:17perplexityO_O morning Bger.. my client appears to have connected unbeknownst to me :) It's always a suprise
08:06:24 Quit goa (
08:06:24 Quit Slasheri (
08:09:25NJoingoa [0] (
08:11:27NJoinSlasheri [0] (
08:18:42 Join ender` [0] (i=ychat@
08:29:25 Quit nathanh ("Leaving")
08:29:41 Quit akaidiot (Read error: 110 (Connection timed out))
08:39:56 Join Membrillo [0] (
08:40:14MembrilloHey kids
08:40:35 Join Zak1392 [0] (
08:42:26 Quit Zak1392 (Client Quit)
08:48:33Membrilloits 648pm here
08:48:47Membrillowhy are you even awake in the morning *yawn*
08:48:55amiconnhehe :) 849am in germany
08:49:04Membrillogees... you must be eager
08:49:13Membrilloi got up at 130pm today haha
08:49:31Membrillooh, the joys of being a teenager in school holidays
08:49:58Bgerhehe amiconn :) mooning
08:50:31Bgeramiconn : regarding iriver remotes
08:51:17Bgerdid you read my talk yesterday with _FireFly_ ?
08:51:18ashridahMembrillo: just wait till you're unemployed, you'll have all year to get up then then :)
08:51:41Membrillothats the life for me
08:52:00Membrilloits like looking into a crystal ball into my future
08:52:32 Quit Membrillo ()
08:55:29 Join nathanh [0] (
09:14:23***Saving seen data "./dancer.seen"
09:17:15 Part Sando
09:28:30 Join San [0] (
09:46:22 Quit nathanh ("Quit")
09:51:01 Join petur [0] (
09:59:49Bgerpetur ;)
10:00:33peturfinally done my patches ;)
10:00:41Bgerah, really ? :)
10:00:56*petur points to tracker
10:01:23Bgerbrightness :)
10:01:47peturand pre-recording...
10:02:10peturand also found out that the LCD optimization broke recording on H3xx
10:02:18 Join Vlad0man [0] (
10:02:52Bgerhow's that
10:03:10*petur points to forum, mailing list and IRC log between 1 and 2 am
10:03:10Bgeryou mean the DMA opt ?
10:03:25 Quit Vladoman (Read error: 110 (Connection timed out))
10:03:40peturI'm afraid this is for LinusN :(
10:05:35Bgeroh, you've commented heavily the code ;)
10:06:49peturdid I?
10:07:20Bgerit's ok:)
10:07:38peturwhat patch are you talking about?
10:08:21Bgerthe only thing i wanted to say is that the last patches have more comments than previous ones
10:09:20warewolfoh right
10:09:29warewolfI just downloaded the lastest h300 rockbox
10:09:39warewolfwhere the heck is the backlight brightness stuff?
10:09:41warewolfI couldn't find it
10:10:19Bgerno, it isn't
10:10:50Bgerit's not in the cvs builds yet
10:14:38 Join HCl [0] (
10:18:18Bgerpetur : do you set the brightness explicitly somewhere ?
10:18:29Bgerapps/settings.c: settings_apply()
10:19:06peturyes, why?
10:19:28peturdefault setting should be 9 (50%)
10:19:48Bgerah, yes, i saw it :)
10:20:05peturthat was missing from the previous patch
10:20:32Bgeri had problem with this yesterday, that's why i'm asking you
10:20:50peturI found out it didn't apply it after loading the settings at startup...
10:21:03peturshould work now
10:21:41Bgeryes, i had the same problem with my work yesterday ..
10:30:43Bgerpetur did you edit the .patch file (brightness) manually ?
10:31:02peturcould be... not ok?
10:31:32 Nick linuxstb_ is now known as linuxstb (n=linuxstb@
10:32:01Bgerpatch: **** malformed patch at line 129: @@ -538,3 +544,12 @@
10:32:20Bgeryes :(
10:32:38Bgerpetur : PM
10:35:33 Quit petur ("CGI:IRC 0.5.7 (2005/06/19)")
10:44:31 Quit perplexity (Read error: 110 (Connection timed out))
11:06:36Bgerwb, Jungti1234
11:07:05Jungti1234IRC is silent.
11:07:25*Bagder drops a needle
11:08:14 Join _FireFly_ [0] (
11:10:35 Quit linuxstb ("Leaving")
11:11:03 Join Kohlrabi [0] (
11:14:26***Saving seen data "./dancer.seen"
11:16:50 Join ratpack91 [0] (
11:19:21 Join PaulJ [0] (
11:25:41solexx(wow, either that needle has fallen from very far above or gravity in IRC is exceptionally low!)
11:28:54amiconn5 km
11:29:02amiconnErm, wrong :)
11:35:56warewolf5:07 to 5:24, 17 minutes.
11:36:14warewolf17 minutes is 1020 seconds
11:36:15Bgerany objections in commiting the h300's LCD brightness patch ?
11:36:38warewolfif gravity accelerates objects at a rate of 32.2 feet per second per second, then that needle fell from a distance of ...
11:36:57warewolffuck if I know.
11:37:01warewolfmy math isn't taht good.
11:37:16amiconn~5000 km, but gravity isn't constant about such long distances
11:37:35Bgeramiconn ?
11:39:14Slasheribut the needle was light and air resistance more limiting factor :)
11:39:39solexxSlasheri: do you think that's air you're breathing now? :)
11:42:43amiconnwarewolf: s = 0.5 * a * t^2 with a = g = 9.81 m / s^2 and t = 1020 s
11:43:03warewolfamiconn: I was just looking at
11:43:37warewolfyou probally took physics in school. I didn't.
11:43:57warewolfI decided I need to go into the IT industry and actually learn a paycheck, and dropped out of college.
11:45:49warewolf5000 km is 3106 miles. Damn that's far.
11:46:25warewolfthat's almost a third of what I drive my car in a year.
11:48:00*warewolf considers blackholing 60.224/13 from his mail server
11:48:26solexxconclusion: you could have saved a year minus 17 minutes if you had gravity propulsion :)
11:48:58 Join nathanh [0] (
11:52:16 Join igor47 [0] (
11:53:15 Quit _FireFly_ (Read error: 110 (Connection timed out))
12:00:30 Join Moos [0] (
12:00:49 Join root [0] (
12:05:01 Quit root (Client Quit)
12:06:21 Join root [0] (
12:06:21 Join JazzBone [0] (
12:07:06warewolfoh man
12:07:08warewolfplease don't irc as root
12:07:13warewolfthat's just asking for trouble
12:09:21 Part root
12:10:26 Join root [0] (
12:12:57 Part root
12:17:24 Join _FireFly_ [0] (
12:19:08 Join miner49er [0] (
12:21:51miner49erhey there guys, can someone answer an IRC related question for me please?
12:23:27miner49erokay, I registerd my nick last night but, naughty naughty, I was logged into my machine as root. I've logged in sensibly as a user, but I need to re-register my nick...i think, as i can't send private messages
12:24:19Slasheriyou should never log to your machine as root (at least not from any window manager). Instead you should use su -
12:25:14miner49erI know, I know. but what can I do now with regards to this IRC thingy
12:25:38ashridahminer49er: registering doesn
12:25:50ashridah't record the ident your machine is using
12:26:16ashridahso just identifying again should work (/msg nickserv identify <nick> <pass>)
12:26:48miner49erashridah: Okay, well why when I opened up my IRC client, as user miner49er did it say nick already registered?
12:27:00Kohlrabibecazse you did last night
12:27:00miner49erthat didn't do anything
12:27:08Kohlrabienter your password there
12:27:15miner49erokay, do I need to log in with my passoword somehow then?
12:27:31miner49erKohirabi: where dude?
12:27:39Kohlrabi/ns identify YOURNICK YOURPASSWORD
12:27:48_FireFly_"/msg nickserv identify <nick> <pass>"
12:28:14KohlrabiJust password I think?
12:28:18KohlrabiAs parameter
12:28:24Kohlrabi/ns identify YOURPASSWORD
12:28:30_FireFly_"/msg nickserv identify <pass>"
12:28:56 Quit Jungti1234 ("bye")
12:29:20Kohlrabiworks, miner?
12:31:09miner49erit works now, i just want a nice quick way of doing it now...automatically
12:31:17miner49ercheers everyone :-)
12:31:41Bgerw00t, my first (ever) commit didn't make red builds :)
12:32:19Bgerand now brightness setting for H300 in cvs
12:32:19 Part miner49er
12:32:52 Join miner49er [0] (
12:33:03_FireFly_miner49er it depends on you irc-client
12:33:18miner49erKSirc is what i'm using
12:33:25miner49erI think!
12:33:30_FireFly_modtly you can define an command or a list of commands which gets executed when you are connected to an irc-server
12:33:35Bagdercongrats on the commit Bger!
12:34:09_FireFly_Bger was you yesterday with the new button-assignment for the remote of h1xx/h3xx
12:35:13Bgeryes, i am :)
12:35:18Bgeri'm working on it
12:35:34Bger10x, Bagder
12:35:43_FireFly_maybe i was a bit fast with may no against the new one
12:35:58Bgerwhy ?
12:36:02Bgerdid you tried it ?
12:36:30_FireFly_after i had played with it a bit longer it isn't so bad as i thought
12:37:01Bger_FireFly_ i don't have such remote and i can't tell anything ...
12:37:11_FireFly_only the vol-up/down should be switched
12:37:17_FireFly_for the h1xx-remote
12:37:42_FireFly_so when you press vol-down on this remote the cursor moves up
12:38:13Bgerisn't this just a matter of habbit ?
12:38:41_FireFly_true :)
12:38:45Bgeri mean
12:38:57Bgerif this is the only problem ...
12:40:00Bgerhm, i should post a question @ misticriver
12:40:14Bgerto get more opinions
12:40:40 Join Denzil [0] (
12:42:08Denzilhi everyone
12:42:13Bgerhi, Denzil
12:42:20DenzilI'm webguest80
12:42:30Denzilhere I have the link
12:42:38DenzilI was talking about yesterday
12:42:48Bgerah, yes :)
12:43:02Denzildunno which nationality the site is
12:43:03 Join mirak [0] (
12:43:16Denzilbut it looks like a university course or such
12:43:49BagderI'm quite sure we have all that info already
12:44:40Denzilso you got the instruction set...
12:44:47BagderI think we do, yes
12:44:56Bgerthis is czech or poland or such
12:45:01Denzilwell, ok, nothing new then
12:45:15Bgerslavic for sure :)
12:45:34Denzilok, that's all
12:45:36Denzilgotta go
12:45:40Denzilsee you!
12:45:45BgerDenzil bye & 10x
12:45:48Bagder"University of Belgrade"
12:45:57Bgerhaha ok :)
12:45:59Bagderowns the IP at least
12:46:19 Quit Denzil ("CGI:IRC (EOF)")
12:46:24Bgerok, serbian :)
12:46:45*Bger didn't guess the language of his neighbours ...
12:47:12Bgerin fact, i can read this pretty easy
12:47:35 Quit miner49er (Remote closed the connection)
12:48:42BgerBagder so the following is known already:
12:49:02BagderI think so. I'm not personally that involved in the MAS hacking
12:49:28Bgerthere are even examples ...
12:49:49Bger(serbian: "Primeri")
12:49:50Bagderwell, we even have code from Archos I believe
12:49:56Bagderfor the wav stuff
12:50:05Bgernot the code, only binary ...
12:50:18Bagderhm, yes that might be right
12:50:20Bgerthe code is in hex in one array
12:50:24nathanhhrm, i think when theres blood on the guitar strings, it might be time to take a break
12:50:34Bgerok, maybe more arrays :)
12:52:26 Join Zak1392 [0] (
12:52:41Zak1392hey guys
12:54:44 Join miner49er [0] (
12:54:54 Part miner49er
12:59:27 Quit edx (Read error: 110 (Connection timed out))
12:59:34 Quit nathanh ("Quit")
13:01:45 Join Weazel_ [0] (
13:01:48 Join perplexity [0] (
13:01:59 Join petur [0] (
13:02:28peturBger: congratulations on the commit!
13:03:06Zak1392what commit?
13:03:19peturbrrrrightnesssss ;)
13:03:27Zak1392oh, that
13:03:30Zak1392pretty cool
13:03:55Zak1392i have mine on 5
13:04:32peturBger: shall I remove the patch from the tracker?
13:07:19peturBger: nevermind, I did already...
13:08:38peturBger: btw, you need to remove yours (1379581)
13:09:26 Join linuxstb [0] (
13:10:35linuxstbpetur: Have you looked at any other features of the pcf50606 apart from the backlight power? I was wondering if the standby mode and wakeup-on-RTC-alarm features would would on the h300.
13:10:43linuxstbs/would would/would work/
13:10:57peturhaven't, but I could ;)
13:11:31Bgerpetur ah yeah :)
13:11:40Bgeryes, remove it, yes, i'll remove (close) mine
13:11:51peturlinuxstb: but it will be for next year...
13:12:11Bgerlinuxstb i can look at them next week :P
13:12:15Bgerbut i don't promise
13:12:41linuxstbI don't own a h300, but I'll be implementing something similar on the ipod's pcf50605. Which is why I mentioned it.
13:12:59Bgerlinuxstb yeah, i know
13:13:16peturI was also looking at the gain setting for recording because I'm not convinced we're using all available gain and mixer settings
13:14:07 Join amiconn_ [0] (
13:14:27***Saving seen data "./dancer.seen"
13:15:34preglowipod nano with cache was in confusingly working state for me today
13:16:17 Quit lostlogic (Read error: 110 (Connection timed out))
13:17:11linuxstbpreglow: I got that sometimes. Reminds me of the bootloader LCD corruption.
13:18:57preglowand that was caused by the cop
13:19:00preglowat least for me
13:19:54 Quit ratpack91 (Read error: 113 (No route to host))
13:20:01preglowbut cop shouldn't affect things here, since it's sleeping, and has got a cache of its own in any case
13:23:10 Nick ashridah is now known as Lost-ash (
13:27:04 Join VagueRant [0] (i=vaguery@wikipedia/vague-rant)
13:27:29VagueRanthowdy all
13:28:27VagueRantquick q, for the life of me i can't work out why my rockboxed h300 volume goes from -100 to 0
13:28:53VagueRantthere was a thread on mistic about it but i can't find it
13:29:49BagderI bet there are several threads
13:29:58Bagderthe scale is now dB
13:30:00peturthere are :)
13:30:50 Quit amiconn (Read error: 110 (Connection timed out))
13:30:50 Nick amiconn_ is now known as amiconn (
13:31:22 Join Membrillo [0] (
13:31:30 Join ratpack91 [0] (
13:31:40MembrilloHiya Kids
13:31:52Bagderhi dad
13:32:30Membrillothats the sort of response i like after coming home from a hard day at the office
13:32:37Membrillonow wheres my wife
13:32:47Membrilloget in the kitchen and make me some dinner
13:33:05VagueRanti been searchin but i can't find it im sorry
13:34:12VagueRantsorry, my bad -84 to 0
13:34:17Membrilloyo linuxstb, you managed to get the ipod bootloader to patch on windows?
13:34:19amiconnThe volume scale should go -84..0 on units with UDA1380 (i.e. iriver H1x0 and H3x0)
13:34:31 Quit Zak1392 ()
13:34:36VagueRantas reported on
13:34:43 Join miner49er [0] (
13:35:30linuxstbMembrillo: No, I haven't looked at that at all. I'm hoping that feature will be added to the new Rockbox installer being developed.
13:35:55Membrillogood good. ill keep patient
13:35:58linuxstbBut I also need to fix the bootloader so that it will load the Apple firmware again.
13:36:06Membrilloof course
13:36:28Membrilloyou traced down the cause of that?
13:37:10 Join Jungti1234 [0] (n=jungti12@
13:37:12Membrillookey doke
13:39:34VagueRantalrighty, is there a way to do percent for volume?
13:41:25BgerJungti1234 yes ?
13:41:38VagueRantanybody got a good graphic volume?
13:42:22VagueRantalso looking for a nice battery symbol (animated?
13:42:31 Quit JazzBone ("Miranda IM! Smaller, Faster, Easier.")
13:45:10amiconnBger: Why isn't the backlight brightness just handled as a numeric setting in settings.c ?
13:45:28peturbecause I didn't know how?
13:45:58Bgerhm, i thinked about this but when it's 0 or 1 the screen is not readable ...
13:46:00amiconnHmm? Numeric settings are even simpler that multi-value settings, and there's a lot of exaples
13:46:31peturalright, hit me on the head ;)
13:46:54amiconnBger: So the valid range is 2..15. This doesn't stop it from being a numeric setting. I know that the current settings system does allow out-of-bounds values,
13:46:59*Bagder hands out the trouts ;-)
13:47:36amiconnWhile this shouldn't be a big problem, it'll be sorted when we switch to file-based settings at some point in teh future
13:48:58amiconnBger: If you want to make sure out-of-bouds values aren't possible, backlight_set_brightness could catch it
13:49:14peturit does...
13:50:12peturbut it will probably accept 0 and 1 as they are valid pwm values :(
13:50:29amiconnYes, that's what I mean
13:50:57Bgerso, the user could select 0 & 1, but they'll be in fact 2 ?
13:51:29amiconnNo, I mean settings.c should handle it as a plain numeric setting
13:51:52amiconnThe settings menu doesn't allow setting 0 and 1. If someone writes 0 or 1 into a .cfg file by hand, he should know what he is doing
13:52:33peturbacklight_set_brightness() should do proper limit checking... my bad
13:52:59Bgerpetur no worries
13:53:00 Quit thegeek (Read error: 104 (Connection reset by peer))
13:53:03amiconnImho it's not necessary, but it could do that
13:53:28*petur remembers you need a strong light to see the display if you set it to 0 or 1 ;)
13:53:32amiconnIt's not really dangerous to set 0 or 1 (dangerous in the sense that it could crash)
13:56:01BgerJungti1234 put it in the incoming/rockbox directory
13:56:16Jungti1234Permission denied
13:57:37Bgeramiconn so the line should be {4, S_O(brightness), 9, "brightness", NULL }
14:01:11miner49erhow can I force the backlight to stay on for a plugin (for archos)
14:01:46Bgerminer49er with calling backlight_on(1) periodically
14:03:07amiconnYou can set the backlight to always on while the plugin is running
14:03:10miner49eroooh, that sounds a bit lame. Can't the configuration settings be temporarily changed?
14:03:19amiconnJust don't forget to set it back before retuirning
14:03:22miner49eramiconn: yeah, that's what I want to do!
14:03:30amiconnThere are several plugins which already do that
14:03:45amiconnvideo.rock is one example
14:03:49miner49erokay, which ones, i'll have a look
14:04:00Bgeramiconn there are some other examples doing the backlight_on(1) thing ...
14:04:56miner49erokay, cheers amiconn
14:05:48amiconnBger: Afaik there is _no_ example that call this periodically
14:07:08amiconnrb->backlight_set_timeout(1); // keep the light on
14:07:45amiconn..and at the end: rb->backlight_set_timeout(rb->global_settings->backlight_timeout);
14:08:32miner49erexcellent, cheers amiconn :-)
14:09:03Bgerso, rb->backlight_set_timeout(1) sets it ON ?
14:09:07Bgerok, sorry, i'm dumb
14:09:51Jungti1234Bger, What do you want?
14:09:56amiconnThe timeout values are idices
14:10:01amiconn0 is always off
14:10:07amiconn1 is always on
14:10:32Bgeryes, just realised that ...
14:13:22 Join webguest95 [0] (
14:14:04 Join lostlogi1x [0] (
14:15:34 Quit perplexity (Read error: 113 (No route to host))
14:16:29Bagderdaily build meta info:
14:17:23 Join perplexity [0] (
14:18:24 Part VagueRant
14:22:28 Join thegeek [0] (
14:23:11 Join hd [0] (
14:23:53 Quit goa (
14:24:05Jungti1234Happy Holidays! :)
14:24:36NJoingoa [0] (
14:25:05 Quit goa (Connection reset by peer)
14:27:20*Bger cryes from the number of wps bmps ...
14:27:28preglowyes, that needs fixing
14:27:32preglowin a way that's not too annoying
14:29:51 Quit Membrillo ()
14:30:53 Quit lostlogicx (Read error: 110 (Connection timed out))
14:31:07*amiconn whispers: Unified bitmap patch by _FireFly_
14:33:19Bgeramiconn please, see my first "OOps" commit :)
14:34:03Bgeryes, Jungti1234
14:34:26Jungti1234Did you do it commit?
14:34:43Jungti1234-> Did you commit it?
14:35:01Bgerno, i corrected my previous commit
14:35:45Bgersomeone here with korean fonts ?
14:35:45Jungti1234I must sleep. :(
14:35:51Jungti1234I have.
14:35:54BgerJungti1234 oki, nite
14:36:10Bgeryes, but i'd like one of the devs to look at it ...
14:36:19 Join MrShlee [0] (
14:36:36Bgerhm, Jungti1234, i don't have time no, gotta go
14:36:44Bgerbut i'll look at it
14:36:47Bgertoday or tomorrow
14:36:58Jungti1234good night :)
14:37:03 Quit Jungti1234 ("bye")
14:37:46MrShleeWeird, I've never had the H3 booting problem until an hour ago :S
14:37:57MrShleeWonder if its got anything to do with the battery :|
14:38:00MrShleeas mine was flat
14:40:50_FireFly_amiconn not unified it si combined-bitmap ;)
14:40:53 Join edx [0] (
14:42:01 Quit San (Read error: 110 (Connection timed out))
14:42:10 Part PaulJ
14:42:57 Quit petur ("CGI:IRC 0.5.7 (2005/06/19)")
14:42:58 Quit JoeBorn (Connection reset by peer)
14:43:00 Join jborn_ [0] ( and
15:04:34 Quit lostlogi1x (Read error: 110 (Connection timed out))
15:05:48 Quit mirak (Remote closed the connection)
15:09:27 Quit Kohlrabi ("Leaving")
15:10:02 Quit MrShlee ("Changed by popular Request - "I am the heart of your darkness, you stinking oompa-loompa morons!"")
15:12:22 Quit _FireFly_ ("An error? Impossible! My modem is error correcting.")
15:12:23 Quit miner49er ("using sirc version 2.211+KSIRC/1.3.11")
15:13:26 Join miner49er [0] (
15:14:04 Join JazzBone [0] (
15:14:28***Saving seen data "./dancer.seen"
15:15:12 Join San [0] (n=test@
15:16:05preglowamiconn: how does that patch work? does it require people to work on a big bitmap, or does it come with conversion tools?
15:18:39BagderI think it simply works on a bit bmp
15:20:50Sanbadger, you relise you posted the H300 remote buttons twice on MR?
15:21:10BagderI didn't
15:21:15Sanyou did
15:21:20Sanit is sorted now tought
15:21:40Bagderbut perhaps Bger did
15:21:54Sanah, bager?
15:22:08Sanread the name wrong
15:22:22Bagderwell you addressed badger up there and there's no one here with such a nick ;-)
15:23:27Sanyou are badger
15:24:02Bagderwell, that only proofs I'm not the only dyslexic here
15:24:06Bagderproves even
15:26:09preglowhow i hate carrying things
15:28:11preglowlater, all
15:28:31 Quit San (Read error: 104 (Connection reset by peer))
15:28:40 Join lostlogic [0] (
15:30:08 Quit DJDD_ ("Trillian (")
15:30:36 Quit linuxstb ("Client Exiting")
15:30:49 Join lostlogicx [0] (
15:31:52 Nick Nibbler is now known as xNibbler (n=sven@
15:31:54 Nick xNibbler is now known as Nibbler (n=sven@
15:32:13 Nick thegeek is now known as thegeek_ (
15:37:38miner49erI've submitted a test version of my asteroids game to the patch tracker, it's just the .c file at the moment but please give me feedback as tonight is the lat night I can work on it for 2 weeks!
15:38:02miner49er...oh, and it probably only runs on the JBR at the moment...oops.
15:45:42 Join dropandho [0] (
15:53:31 Join RotAtoR [0] (
15:56:47 Quit jborn_ ("")
16:01:29 Join Rob2222 [0] (
16:04:25 Join Mongey| [0] (
16:06:18Bgeruf, these MR members :(
16:06:46Bgeri posted it 2 times: one in H300's RB forum and one in H100's RB ...
16:07:45Mongey|posted what?
16:09:48Mongey|ill test
16:10:41Mongey|im luke im 5, my dads bruce lee, he drives me round in his jcb
16:11:55Mongey|do u like that sog?
16:12:15Bgerin fact i don't know it
16:13:28Mongey|remote is brill
16:14:07 Quit Mongey| ()
16:15:17Bgerbril ?
16:15:45 Quit lostlogic ("Going to the moon")
16:16:08 Join webguest41 [0] (
16:16:36 Quit Nibbler (Read error: 104 (Connection reset by peer))
16:23:26 Quit XavierGr ("Trillian (")
16:30:59 Join tvelocity [0] (
16:34:08 Join Nibbler [0] (
16:41:30 Quit webguest41 ("CGI:IRC (EOF)")
16:42:12 Join Sando [0] (
16:44:54 Join _FireFly_ [0] (
16:57:28 Quit perplexity ("*crack*")
17:07:00 Join DiveineIn [0] (
17:07:13DiveineInhello, ne1 here?
17:08:07DiveineInid like some guidance pls...
17:08:17DiveineIni want to start building plugins for the h300
17:08:25DiveineInand i cant find any info on it...
17:08:33miner49erokay, well i've built some for archos
17:08:39DiveineInany chance you (or anyone else) can point me in the right direction?
17:08:57DiveineIni understand they are build in C, right?
17:08:59miner49ertake a look at the helloworld.c plugin for a start
17:09:04miner49eryeah, all in c
17:09:13miner49er...or go there!
17:09:28DiveineIndamn, i forgot c, moved on to .net, itll take a while to remember :(
17:09:37Bagderfirst make sure you can build everything from source fine
17:09:43miner49erI doubt you completely forgot it!
17:10:20DiveineInnot completely, but im sure ill confuse C# with C
17:10:21miner49erThe instructions on the site are very good. What operating system are you on?
17:10:57DiveineInalready d/l cygwin
17:11:18 Join akaidiot [0] (
17:14:06DiveineInwhats SH1 and COLDFIRE?
17:14:29***Saving seen data "./dancer.seen"
17:15:13miner49erThere is a very good development kit put together by Bluechip that works well on XP
17:15:13miner49eraah, well the devkit has it all wrapped up nicely. i'd recommend it
17:15:13miner49eryou just install and go really
17:15:15miner49erget the latest source tarball as well though, full instructions come with the devkit
17:15:21lostlogicxprocessor types
17:15:33miner49erthere the CPU's I believe, Archos uses SH1 and I think the iRiver uses Coldfire, though I'm not 100%
17:15:35miner49ermust dash now anyway, byee!
17:15:43lostlogicxthat's correct.
17:25:45 Join San||Away [0] (n=test@
17:36:44 Join pinkutank [0] (i=pinkutan@
17:37:14pinkutankis linus on, oh no, not on
17:37:24pinkutankanyone palying around with rockbox for h300
17:38:33pinkutankI'd like to know if flashing rockbox with us firmware kills drm
17:42:42lostlogicxpinkutank: afaik yes it does.
17:43:40pinkutankbah, would flash right away but I'm gonna sell mine in a month
17:43:46pinkutankI really dont like it
17:44:09lostlogicxhmph, I love it with rockbox, liked it a lot w/ original firmware too.
17:44:22pinkutankI dont like the key layout
17:44:28pinkutankI dont like the screen
17:44:37pinkutankmy fingers are too big for the keys
17:44:49pinkutankand I hate seeing a black screen when the backlight is off
17:44:50lostlogicxwhat are you going to replace it with?
17:45:00lostlogicxany color LCD will have that most likely.
17:45:06pinkutankI have a ihp 120 that I use
17:45:16pinkutankand not necessarily any color lcd has that
17:45:19pinkutankfor example
17:45:23pinkutankipod nano
17:45:25pinkutankipod 5g
17:45:30pinkutankcreative zen vision m
17:45:35pinkutankand archos gmini 40*
17:45:43lostlogicxthe backlight goes off and the screen is still readable!?
17:45:54pinkutankin all of those players
17:46:08pinkutankthere are different kinds of lcds out there
17:46:15lostlogicxI'll have to look at one of them −− I've never seen a color LCD like that.
17:46:24pinkutankthe only advantage of having one like the h320 is you can see it better from sides
17:46:32Slasherimy mobile phone color screens are too very readable when backlight is off
17:46:38lostlogicxwell that's a pretty big advantage, eh?
17:46:55pinkutanklol, I think it isnt even an advantage
17:47:04lostlogicxSlasheri: that's true, my cell is readable without backlight, but it's so dull compared to the H300
17:47:15pinkutankI dont put my player on a desk and try to watch a movie with 4 friends
17:47:27Slasherilostlogicx: hehe, might be true.. i have never seen the h300 screen
17:47:30lostlogicxhaha, I put it on my desk to th side and glance at it so I need the angles.
17:47:37pinkutankh300s screen is not that good
17:47:42pinkutankyou can still see it
17:47:45lostlogicxI guess this is why there are so many different types of DAPs ;)
17:47:55pinkutankbut it will be a little washed out at extreme angles
17:48:26SlasheriHmm, but i think oled displays would be the coolest ones :)
17:48:44lostlogicxyarr, lower power consumption too, right?
17:48:47pinkutankipod nano, ipod 5g, and zen visionm all have better screens than h300
17:48:53pinkutankI never seen an oled yet
17:49:18Slasherioleds are something like a very thin led matrix.. they have extremely wide angle and clear screen
17:49:34Slasheribut they draw some power constantly.. but it should be easy to dim those thought
17:58:38DiveineInummm, stupid question
17:58:48DiveineIni set up everything for starting to make plugins
17:58:57DiveineInbut i never worked with unix/linux before
17:59:08DiveineInis there a dev enviornment?
17:59:14DiveineInhow do i use it?
17:59:29DiveineIneven a simple text editor would be fine :)
18:02:52Slasherimake is everything you need to compile the code :) if you are looking a console editor, i could suggest jed (but you should modify the .jedrc file). kate is great for use in kde
18:03:18DiveineInSlasheri: was that in english?
18:05:40ender`type make, then type jed :)
18:06:58pinkutankin the console that is
18:07:09ender`DiveineIn: what OS are you used to?
18:08:22_FireFly_i think Windows
18:08:42 Quit pinkutank ()
18:11:23miner49erjust use notepad! or download crimson editor for free :-)
18:11:34_FireFly_or jedit ;)
18:29:53amiconnHmm, still no LinusN :(
18:30:10_FireFly_tja then you must wait ;)
18:30:20_FireFly_or mail him if you now the address
18:30:27_FireFly_an email ;)
18:36:27 Join mirak [0] (
18:38:15 Join dpassen1 [0] (
18:50:30dropandhohey firefly!
18:50:36dropandhoare you mr. remote?
18:50:48|u|smoothanything new for the h300?
18:50:59dropandhoyou do a bunch of work on the remote for iriver...dont cha?
18:51:16dropandhothank you sooo much!
18:51:20dropandhoit is reallly appreciated
18:51:24_FireFly_:) thanks
18:51:25dropandhoand sooo helpful
18:51:47dropandhowas wondering if you were still working at it? porting the other screens etc?
18:51:52_FireFly_TiMiD has created a good base to do it the right way
18:52:05dropandhoyes, definately!
18:52:16mirak|u|smooth: xvid
18:52:38|u|smoothvideo works?
18:53:04_FireFly_currently not much i have made only the sleep-timer-setting menu be shown and adjustable from the remote
18:53:08mirakthere is an unoptimised plugin
18:53:39dropandhogot it
18:53:45 Join petur [0] (
18:53:49dropandhoany plans to do any further?
18:53:52 Quit Rob2222 (Read error: 104 (Connection reset by peer))
18:54:30 Join Rob2222 [0] (
18:54:31_FireFly_at least the vkeyboard and bookmark select-screen are missing
18:54:44dropandhoyeah- i saw that
18:54:50dropandhoyou leaving that up to someone else?
18:54:56amiconnrecording screen...
18:55:14_FireFly_amiconn: if you want you can do it ;)
18:55:28amiconnIf I had enough time I would,
18:55:36amiconnbut remote is low-pri for me, you know...
18:55:37|u|smoothyea why when i boot up its automatically on recording
18:57:09 Join Lear [0] (
18:57:23_FireFly_a big task would be to made the vkeyboard multi-screen capable if the different keyboard-code for the charcell-lcd-targets should be also covered with the new one
18:57:34mirakcan I delete files from rockbox ?
18:57:59_FireFly_mirak: do you mean when running rockbox ??
18:58:09mirakdelete files from the file system
18:58:16mirakwith rockbox
18:58:41_FireFly_long press select then you can select delete in the showing menu
18:58:51amiconn_FireFly_: I don't think so. Going multiscreen requires more flexible vkeyboard code anyway, and integrating charcell support should become relatively easy
18:59:34_FireFly_amiconn: could be but i have currently no idea how to do it
18:59:41amiconn...including morse support
18:59:48amiconnI do
19:00:01amiconn(have an idea)
19:00:07_FireFly_tell me
19:00:11_FireFly_please :)
19:00:18_FireFly_you you have time
19:00:25_FireFly_if you
19:00:32amiconnSee yesterday's log
19:01:26_FireFly_i will look
19:01:38mirakcan anyone tell me how the iram is used actually in the codecs ?
19:02:05mirakare the data copied hover iram for each important function ?
19:02:18mirakthen passed back to ram ?
19:03:13dropandhothat would be sooo great to work out the remote stuff!
19:03:42Learmirak: soon after start each codec copies whatever is needed. Just look in the beginning of a codec.
19:04:20LearLook for the USE_IRAM ifdef.
19:04:53_FireFly_amiconn: i think this is a nice idea
19:06:18_FireFly_we could use the current one-page-array and define an rect which defines the viewable chars
19:06:36_FireFly_depending on the screen-size
19:07:00dropandhowooo- so its not soooo bad?!
19:07:50_FireFly_then we would have in the first stage only two realy pages the char-map and the morse-code-list
19:07:56 Join Febs [0] (
19:09:55_FireFly_also to have the non ascii/latin-1 eventually latin9 char-map for the keyboard to be part of the language file is great
19:10:05LearHrm... Glyph preloading crashes the simulator... So ... keep the preloading and fix it (and make it more complete?), or remove it? Any opinions?
19:13:38mirakLear: I see this line, but I don't understand how it can be enough
19:13:41mirak #ifdef USE_IRAM
19:13:41mirak rb->memcpy(iramstart, iramcopy, iramend - iramstart);
19:13:41mirak #endif
19:14:30***Saving seen data "./dancer.seen"
19:15:43LearIt copies all data from a copy to the beginning of the iram section. The data is either constant or zeroed out (bss), so there's no need to copy anything back (other than when switching to the voice playback, but the codecs don't need to worry about that).
19:16:14_FireFly_amiconn: i will try to create a small "design-paper" for this
19:16:35mirakLear: I don't understand
19:16:35dropandhothat is sooo great firefly
19:16:38dropandhothanks ones again
19:16:48dropandhoit is really appreciated that you are picking this up and running with it
19:16:55dropandhoi would love to have a complete remote
19:17:03mirakLear: the copy happen only once ?
19:17:29LearOnly once, during codec initialization, yes. Codecs are loaded on demand, after all.
19:17:46mirakso what is copied then ?
19:18:06amiconnLear: If glyph preloading crashes the sim, I'd say that should be fixed. The sim usually doesn't crash without reason...
19:18:08mirakI though music data where copied into iram then processed into it
19:18:15mirakLear: I don't get the picture
19:19:18LearNo, generally the file data isn't copied there (unless to a temporary buffer). IRAM is mainly used for commonly used tables and buffers and small chunks of decoded audio (during processing).
19:19:38LearThere is only 96 kB, of which the codec gets to use half of it.
19:19:41amiconn_FireFly_: I can see problems for 2 targets with my virtual page idea.
19:19:43mirakLear: yes of course I understand that
19:19:58_FireFly_amiconn: yes ?
19:21:07amiconnThere are 2 targets which use keyboard modes because there aren't enough buttons to allow separate button (combos) for moving the cursor within the input line and deleting chars.
19:21:26amiconnThese use keyboard modes today
19:21:35Learmirak: So, what parts do you not understand?
19:21:51_FireFly_amiconn: i have seen it in the sources
19:22:01MoosLear: hi, did you look at the lostlogix's patch for optimise Tremor a bit again?
19:22:46amiconnOn Ondio, LEFT at the left border and RIGHT at the right border are used to switch pages, as there is no separate button for that. UP at the top border and DOWN at the bottom border switches to line mode, allowing to move the cursor and delete chars within the input line
19:23:13Learmoos: again and again. I've applied one part of it (though in a different way), and I plan to do the rest too (the iram tweak bit).
19:23:28_FireFly_i think on these 2 targets a button could switch between line-editing and char-selects
19:23:49amiconnOn the player, the virtual page is one line only. ON switches between line edit at picker
19:23:49 Join Paul_The_Nerd [0] (
19:23:56MoosLear: good to hear you're in this area
19:23:57mirakLear: when it happens in the code
19:24:14amiconn_FireFly_: Which_ button on Ondio is the biiiig question
19:24:21mirakLear: I am in tremor and try to search wich datas are put into the iram
19:24:52amiconnLEFT/RIGHT/UP/DOWN are used for moving. MODE adds a char (or deletes a char in line mode). Long MODE accepts, OFF aborts
19:25:01amiconnNo other button available...
19:25:11_FireFly_and what about combos ??
19:26:06mirakLear: I am thinking how it can be used for xvid
19:26:06amiconnWe could keep the move-out-of-picker-area system, but the picker area will be bigger with a virtual page, potentially way bigger
19:26:16Learmirak: the copy I mentioned is where it happens. The rest is done during linking. Through the ICODE_ATTR/IDATA_ATTR/ICONST_ATTR macros, different bits of code and data are put in separate sections, that are located after each other in the binary file. The copy does the final part.
19:26:18amiconn(thinking chinese/japanese)
19:27:23Learmirak: Similar approach, I guess, though you need to make sure no codec is running first, obviously...
19:27:23 Quit DreamTactix291 (Read error: 104 (Connection reset by peer))
19:27:41 Join DreamTactix291 [0] (
19:28:26_FireFly_the otherside at least for the player it would be hard to display on them non-latin chars
19:28:31amiconnCombos might be possible, but combos involving MODE forbid the convenient 'Long MODE accepts' assignment. Otherwise the user has to be really fast...
19:28:49mirakthere is no japanese participating rockbox ?
19:28:52amiconn_FireFly_: The player doesn't need to display non-latin.
19:28:53 Join einhirn [0] (
19:29:04_FireFly_ah ok
19:29:12amiconnIt should just share the virtual page (and morse) code
19:29:59_FireFly_you mean the builtin virtual page
19:31:31 Quit tvelocity (No route to host)
19:34:25 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
19:36:42_FireFly_amiconn: can only MODE be used to have two button combos on the ondio??
19:37:06amiconnMODE or OFF, but OFF is very inconvenient
19:37:40amiconnCombos involving OFF need to be pressed in the 'reverse' order, to avoid switching the unit off
19:37:41_FireFly_yepp but a long press OFF could be used but if should is a other question
19:38:08_FireFly_oh yes
19:38:18amiconnWe already have 2 places which use long OFF, but that's ugly
19:38:33_FireFly_ok long press OFF is no option
19:39:20mirakLear: long oggpack_look_full(oggpack_buffer *b,int bits) ICODE_ATTR; what is that doing ?
19:40:08LearTells the compiler to put that function into a specially named section.
19:40:21 Quit akaidiot (Read error: 110 (Connection timed out))
19:44:19 Join saratoga [0] (
19:44:51 Quit JazzBone ("Miranda IM! Smaller, Faster, Easier.")
19:54:17mirakLear: so generally the use of iram is not dynamical
19:58:42mirakLear: but for exemple, the window on wich the idct are applied is copied in a buffer at run time
19:59:50mirakfor exemple
19:59:57LearWhere, more exactly? I guess tremor somewhere...
20:00:29mirakfor jpeg or xvid, I guess the blocks needs to be copied
20:01:17LearYou refer to the window data in window_lookup.h?
20:01:17mirakyou can get an improvment of how much when using iram ?
20:02:37LearI've no exact measures, but quite a lot. When Slasheri played with dithering, it was soon noticed that it was quite slow. I then tried moving most of the data used by the dithering code, for a large decrease in boost ratio (could've been halved or something, don't remember exactly).
20:03:50Learit all depends on how much the data in question is used.
20:04:23Learbtw, anyone who knows what kind of latencies you get from normal ram?
20:05:27_FireFly_amiconn: on you can find a file which defines the structur of the vkeyboard and used data-structs
20:11:05dropandhou rock firefly!
20:11:10dropandhosooo great
20:11:14amiconnmirak: Data that is only read once doesn't benefit from copying to IRAM, quite the opposite would happen
20:11:22dropandhois there anything else that needs to be looked at for remote?
20:11:29amiconnAs soon as it is used more that once, the copying pays off
20:11:45mirakamiconn: have you done some benchmark with the same function in ram and iram ?
20:12:12lostlogicxLear: any thoughts on my next version tremor optimization patch? Things that I did that are big no-nos, other approaches you think I might have better luck attacking?
20:12:13_FireFly_dropandho: recording-screen
20:12:18amiconnLear: Copying a large block of data from DRAM to DRAM takes ~4 times as long as copying a same-sized block from IRAM to IRAM
20:12:27amiconnThat's already _with_ burst mode
20:12:56Learlostlogicx: next version?
20:13:01lostlogicxLear: in case you missed it yesterday, I like how you did the inlining, definitely better than just merging the .cs like I did. Only concern is the repetition of that add, but I guess it doesn't hurt much.
20:13:18amiconnmirak: Putting code into IRAM doesn't give nearly as much boost as putting data into IRAM
20:13:31amiconnThe coldfire has an instruction cache, but no data cache
20:13:32lostlogicxLear: I posted to patch tracker my latest optimizations, rebased up to what you put in CVS.
20:13:42LearIIRC, the compiler re-used the result of it, so it might not be as bad as it looks.
20:14:03lostlogicxLear: sweet!
20:14:33lostlogicxI didn't actually check the assembly results for those lines.
20:15:44lostlogicxI'm beginning to think that we can get a lot of performance improvement by converting everything that _is_ unsigned in the codec to be explicitly unsigned. Looks like a lot of time and code size is wasted jumping around to deal with positive vs. negative results that are impossible.
20:16:10LearWell, I'm not sure if I checked the "CVS"-version, but I had that in one version at least...
20:18:38lostlogicxthe addition is reused.
20:20:29Learhm... do all those asms for stuff like add really help?
20:21:13Learand at least those in oggpack_look seem to be in the rarely executed branch...
20:23:20lostlogicxgood point on oggpack_look and yes, they do −− addq.l is much faster than move, add, move −− it depends how often the code is called... I threw them in everywhere they'd help at all, without really considering only doing it in critical code paths.
20:25:19LearI did some quick checks on that, and the rare branch is executed in something like once every 200 calls or so.
20:26:29lostlogicxyeah, it's much much less, even GCC knows that, and compiles the code so that it's the further jump. the ADDQIs in framing.c are hit about 30k times each in a song
20:26:42LearBtw, I did try assembler optimizations for the often executed branch in oggpack_look (i.e., the inlined part). Turns out it was difficult to improve the common cases. Easy to improve the uncommon case though... So I dropped that bit.
20:26:45lostlogicxfor comparison, oggpack_read is called 206k times
20:27:07Learoggpack_look far more than that, afaik.
20:27:07lostlogicxLear: yeah, you're right on that, I discovered the same after much sweat and blood over it.
20:27:22lostlogicxyeah, oggpack_look was in the millions
20:27:50lostlogicx15 million on oggpack look and 11 million on oggpack_adv
20:29:18lostlogicxpraps just the render_point part is worth taking out of that patch ;) that had a tangible gain alone, because it was 1) written stupidly and 2) compiled stupidly.
20:30:26lostlogicxoh, that and maybe the (unsigned)s in look.
20:31:42LearI was plannig on at least doing iram tweaks first (separate commit) and then render_point and perhaps render_line.
20:32:01_FireFly_amiconn: i have updated the file
20:32:13lostlogicxLear: rock on.
20:32:19_FireFly_amiconn: does it reflect your idea
20:32:39LearNow, if we only could get someone to port the FFT patch... :)
20:32:51_FireFly_Lear: FFT ??
20:33:20LearUsing an (I?)FFT to do the IMDCT in Tremor.
20:33:42lostlogicxLear: I think perplexity is working on that
20:33:48_FireFly_this doesn*t say anything to me ;)
20:33:51lostlogicxand teaching GCC to automagically do some of the ASM stuff I've done.
20:34:27mirakLear: ifft yes :)
20:34:39lostlogicxFast Fourier Transform
20:35:13mirakthe inverse one
20:35:13 Join akaidiot [0] (
20:35:18lostlogicxInverse Modified Discrete Cosine Transform
20:35:26lostlogicxI looked them up on google / wikipedia
20:35:30 Join NicoFR [0] (
20:35:40mirakfourier transform a wave to a spectrum
20:35:50mirakinverse do the opposite
20:36:06ratpack91i always feel sorry for the cpu having to do DFTs.
20:36:15mirakmaybe we could use the same
20:36:22mirakfor all the codecs
20:36:32lostlogicxI think MAD already uses an FFT
20:36:42mirakoptimised ?
20:36:53miraklostlogicx: it's an ifft
20:37:19lostlogicxdon't know how optimized it is, but MAD is pretty fast...
20:37:37mirakmad is a mp3 codec ?
20:37:37_FireFly_amiconn: ??
20:37:42lostlogicxmirak: yeah
20:39:14 Join linuxstb [0] (
20:39:16LearOn the lowmem branch of Tremor, the FFT patch speeds things up by about 30%, when run on a PC...
20:39:46 Join arkascha [0] (
20:40:58lostlogicxwow. That'd be impressive as hell.
20:43:48lostlogicxLear: What makes that patch incompatible with Tremor in rockbox?
20:44:13 Join DrumRBoy320|Away [0] (
20:44:23*amiconn takes a look at the design doc
20:44:25LearThe lowmem branch is quite different, in many ways...
20:44:55LearIt's not trivial to move it over at least; things about the windowing is different, for instance.
20:45:59 Join linuxstb_ [0] (
20:46:11 Quit linuxstb (Nick collision from services.)
20:46:15 Nick linuxstb_ is now known as linuxstb (
20:46:15*_FireFly_ hopes the doc is written in a understandable way
20:47:07amiconn_FireFly_: Some details are missing, and some more details are tbd
20:47:17_FireFly_tbd ??
20:47:22amiconnto be determined
20:48:09amiconnOne thing is that it should be made clear that there are 2 virtual pages, one for the 'classic' picker mode and one for morse input (which is actually only a help page)
20:49:29_FireFly_i thought this makes it clear: "a page for latin-chars and morse-codes are builtin as the standard pages"
20:49:40amiconnThe thing to be determined is whether the virtual page should be fixed both in X and Y dimension, or if it would be better to rewrap it to the display width, scrolling in Y direction only
20:50:21amiconnI would like to put latin and non-latin all into one page, and morse into the second
20:50:39_FireFly_this is better ? "a page for latin-chars and a page for morse-codes are builtinas the standard pages"
20:50:43amiconnI tend to like the rewrap idea more, this would ease implementation on Ondio
20:51:43amiconnThen we could imagine the 2 virtual pages as being next to each other. Moving right from the picker page would switch to morse input, and moving left again would switch back
20:52:01_FireFly_that can be easily be done to define the "len" var in the virtual-page struct to be the lcd-width
20:52:55amiconnWe could even handle the line edit problem on Ondio the same way - moving left from the picker page or right from the morse page would switch to line mode.
20:53:02amiconnUp or Down could switch back
20:53:17_FireFly_"There are two pages builtin on for the latin-chars and
20:53:18_FireFly_one for the morse-code-input(currently only a help-screen)"
20:53:22amiconnAll this could be indicated by a buttonbar-like indicator line
20:55:55_FireFly_if the cursor is located either on the right side or the left side the page is switched
20:56:15_FireFly_when pressing to the left or right depending on the current page
20:58:52 Join hshah [0] (
21:00:39amiconnYes, and for the Ondio there's a third 'page', the line edit mode
21:02:03amiconnThe player should keep the line oriented layout, imho. This is better because (1) the picker can be only one line anyway and (2) there's '-' and '+' as the equivalent of LEFT and RIGHT, but no equivalent for UP and DOWN
21:02:37_FireFly_should the third "page" for the ondio be reachable from both pages ??
21:03:39amiconnI think it could work like this: (actually a ring) <->picker<->morse<->line edit<->
21:03:59_FireFly_or so
21:04:03amiconnErm, slightly wrong
21:04:23_FireFly_ah right
21:04:35amiconn<-picker-> <-morse-> ^-line edit-v
21:04:39_FireFly_in line-edit mode you need to press UP or DOWN ;)
21:06:47_FireFly_has the player also the morse-code-page ??
21:06:53 Part Paul_The_Nerd
21:07:21amiconnImho it should. Perhaps w/o help, just saying 'morse input'
21:07:38_FireFly_and how to reach ??
21:09:11amiconnImho there are 2 options. Either we keep the current player keyboard, where ON switches between line edit and picker, and MENU flips through picker lines (currently 3)
21:09:31amiconnThen we could simply add morse input as a 4th 'picker line'
21:09:53_FireFly_this option would be the easiest because this code needs not be rewritten to the new "design"
21:10:47amiconnOr we want unified code, then MENU could flip through modes (picker / morse/ line edit) and ON could scroll through lines in picker
21:12:25 Join jlo [0] (
21:12:36jlohi all
21:13:18dropandhoyour work sounds perfect!
21:13:40dropandhois it hard to code?
21:14:33***Saving seen data "./dancer.seen"
21:15:16_FireFly_i have update the doc
21:21:05 Quit saratoga ("CGI:IRC (EOF)")
21:21:49_FireFly_i think it doesn't really matter which option is used but the second option has mayby the advantage that the overall source-code size can be reduced a bit
21:23:11San||Awayanyone working on a "top 25 played songs" for rockbox?
21:23:15 Nick San||Away is now known as San (n=test@
21:23:24_FireFly_for the second methode we could split the device-dependent code from the device-independent code which could reduce the usage of #ifdefs
21:23:47peturSan: make that an audioscrobler plugin ;)
21:23:56Sanwell, yes
21:24:00_FireFly_so that the independent keyboard code includes the device-dependent header
21:24:09Sanare they petur?
21:24:33Sanmaking it? I heard some about it but saw no progress
21:24:43peturno idea...
21:26:08peturlinuxstb: are you here?
21:26:39_FireFly_amiconn: we could define some interface-functions e.g. get_next_page, handle_button, draw_picker, draw_edit_line
21:38:31 Join Rob2222_ [0] (
21:43:00*petur thinks we should do something about the way multi-platform support is implemented
21:43:19 Quit hshah ("Leaving")
21:44:20peturadding rtc for h3xx is going to be clumsy...
21:44:29 Quit Rob2222 (Read error: 110 (Connection timed out))
21:44:45 Quit NicoFR ()
21:45:26peturdriver must be split just like LCD was...
21:45:49 Join Jungti1234 [0] (n=jungti12@
21:46:39amiconnpetur: Isn't the driver already split? H3x0 RTC is part of PCF50606 iirc
21:47:18*petur goes looking
21:48:05peturnope, and it shouldn't... it only contains general access methods
21:48:26petureven my brightness stuff doesn't belong in there...
21:49:16peturreason: if it's register compatible with the PCF50605, that code can be shared...
21:49:52peturrtc was included in the pcf50605 driver...
21:50:50 Join dupczas [0] (
21:52:07dupczascicho tu jakos
21:53:27Jungti1234What mean is it?
21:53:46dupczashmmmm i think that you understand polish
21:53:48dupczasi mean
21:54:05LearYay, finally found why mpa.codec handled bad files so badly. :)
21:54:05dupczasthat is very quiet here
21:54:17Jungti1234yes :)
21:54:22LearNow I just need to figure out how to fix it wihtout borking the seeking. :)
21:56:02Jungti1234sleep all?
21:56:18dupczasno ;)
21:56:43dupczaswhat time is it ??
21:56:43Jungti1234What time is there?
21:57:04dupczasrotfl !
21:57:15dupczas9 57 pm ;)
21:57:28dupczaswhat you come from ?
21:57:42Jungti1234Just because.
21:58:14Jungti1234I don't understand.
21:58:15dupczasi don't get it ??
21:58:21dupczasme too ;)
21:58:34dupczaswhere do you live ?
21:58:44Jungti1234I live in South Korea.
21:59:02dupczaslol nice
21:59:11jlohi preglow : did you have time to listen to files or think about shelving filters ?
22:00:23dupczaseverobody here lives in South Korea?
22:01:12Jungti1234Only one .
22:01:51Jungti1234But, Rockbox is many korean people.
22:02:04_FireFly_dupczas: most are in europe
22:02:13 Quit linuxstb (Read error: 110 (Connection timed out))
22:02:45 Join linuxstb [0] (
22:03:31Jungti1234dupczas: Where do you live?
22:03:44dupczasin Poland :)
22:04:25Jungti1234ah. :)
22:04:54DBUGEnqueued KICK dupczas
22:04:54dupczasJungti1234 you are female or male ?
22:04:54 Quit edx (Read error: 104 (Connection reset by peer))
22:05:16LearAh, that looks promising...
22:05:57Jungti1234I'm male
22:06:05dupczashehe me too ;)
22:06:29Jungti1234Did you want that I'm female?
22:07:36Jungti1234Nothing. :)
22:08:09peturconversations sure have a high peak-to-peak value here ;)
22:09:42 Quit jlo ()
22:09:48Jungti1234I go to theater today.
22:10:29 Part dupczas
22:10:37 Quit Jungti1234 ("bye")
22:11:47 Join solexx_ [0] (
22:12:14MoosLear: what will?
22:12:49 Quit Febs ("CGI:IRC (EOF)")
22:13:39Moosdid you have discovered something? :)
22:14:15LearThe mpa codec fix.
22:15:01Moosyou found one way for fix it?
22:15:07Moosalready I mean
22:15:53Learalready? spent about two hours digging...
22:16:25Moosoops sorry :)
22:22:50 Quit solexx (Read error: 113 (No route to host))
22:25:24linuxstbpetur: Have you seen this patch?
22:26:32linuxstbThat's basically all that's needed to enable the RTC on the h300. LinusN has seen it, but he said that the i2c commands to read the RTC need protecting from interrupts.
22:29:40 Join LinusN [0] (n=linus@
22:29:46peturhmmm... and why don't the other I2C calls need protection from interrupts?
22:29:54peturlook who's there ;)
22:30:54amiconnhi LinusN :)
22:31:07LinusNpetur: the pcf50606 is used in the adc interrupt
22:31:11LinusNamiconn: hi
22:31:45LinusNso you have to protect all pcf50606 communication that can be interrupted by the adc (and button) interrupt
22:31:56peturah ok
22:32:45amiconnLinusN: Two questions regarding H300 lcd: (1) Do you have an idea why lcd DMA interferes with recording?
22:32:48 Quit San (Read error: 110 (Connection timed out))
22:32:58amiconnDid you perhaps use the same DMA channel?
22:33:01LinusNamiconn: of course i have, it uses the same dma channel
22:33:10amiconnHow many DMA channels are there?
22:33:20amiconnBah, not good :(
22:33:25LinusNnot very
22:33:33amiconnThe other is used for playback?
22:33:56amiconnHmm. Why does the SH1 have 4 DMA channels, but the coldfire only has 2? :/
22:34:06LinusNwe could use the same for playback and recording
22:34:07peturcan recording and playback not share a DMA channel?
22:34:27LinusNhowever, it might be nice to be able to play and record simultaneously
22:34:32amiconn(2) Do you think a non-DMA lcd driver that uses burst-reading would be as fast (or even faster) than your DMA version?
22:34:37LinusNespecially for voice
22:35:07amiconnI still have no H300, hope to get one from ebay after xmas
22:35:11peturdoesn't it use the same bus between controller and uda?
22:35:19LinusNamiconn: i don't think you can make it as fast
22:35:30LinusNbut definitely faster than the current one
22:35:42amiconnIs it possible to write the GRAM backwards?
22:35:56LinusNi think so
22:36:04amiconnGood, that'll save some insns
22:36:54amiconnBurst-reading 4 longwords and then writing as 8 words is easier when starting with the low word of each longword
22:37:00 Join tvelocity [0] (
22:37:01LinusNpetur: there are different in and out busses
22:37:06peturbtw, Linus, you probably saw my recording patch is ready :)
22:37:20LinusNpetur: nice
22:37:49peturso we leave the dma alone for lcd then?
22:37:56LinusNi think so
22:38:10LinusNthe 5250 has 4 channels :-)
22:38:27LinusNiaudio here i come
22:38:47LinusNtoo bad the 5250 data sheet pdf is corrupt... :-(
22:38:57LinusNor rather, the user's manual pdf
22:39:17amiconnmovem.l and then 4x move.w / swap / move.w
22:39:29peturI do think that the number of times we'll need recording+voice is _far_ less than the times we want quick lcd updates... tradeoff?
22:40:51Learblind users might not agree...
22:41:00LinusNpetur: that depends on how much slower the lcd is
22:41:11LinusNwe could change it dynamically
22:41:19LinusNuse dma when not recording
22:41:26amiconnYou said the DMA solution was 20% faster?
22:41:33LinusNsomething like that
22:41:54petur40 -> 50 fps
22:42:09LinusNit could have been a lot faster if we could use burst accesses
22:42:18LinusNwhich we can't
22:42:51LinusNhowever, your solution can use bursts in dram, which the dam couldn't
22:43:07lostlogicxcould the LCD DMA potentially have caused playing problems when playing music and disk reading?
22:43:12LinusNso your solution might be just as fast
22:43:36LinusNlostlogicx: i doubt it
22:47:13lostlogicxthis is distressing. I'm getting strange semi-random freezes playing Kelly Clarkson - Thankful. I think my iRiver is telling me to listen to better music.
22:47:53dropandhoFireFly- you workin that why you got quiet?
22:47:55lostlogicxand it's not my hackage of Tremor that's causing it either, because I just reverted to a clean codebase.
22:48:56_FireFly_dropandho: not really :)
22:49:50dropandhohehe- its ok!
22:49:57dropandhoim just being nosey
22:50:02dropandhobut it is all pretty exciting
22:50:09dropandhoi appreciate your commitment to the remote issues!
22:50:14_FireFly_LinusN: what do you think about this idea to make the vkeyboard multiscreen-capable:
22:50:16amiconnLinusN: Reading line-bursts from DRAM is 2.5 times as fast as reading longwords
22:50:36amiconn...and even more compared to reading words
22:53:46 Join Rob- [0] (
22:55:13amiconnLinusN: H3x0 uses the 80-system 16bit interface?
22:56:50Rob-Hi, is it a known bug that the latest daily build won't record on h340?
22:57:08peturuse the bleeding edge
22:57:35Rob-is it fixed now?
22:58:03Rob-I'll probably wait for the daily as I don't have a build environment setup.
22:58:34_FireFly_no need for a build env
22:59:06_FireFly_after each cvs commit this build will be updated
23:00:09Rob-excelent, thanks!
23:02:36 Join San [0] (
23:03:59 Join bazz [0] (
23:04:40 Quit arkascha (Remote closed the connection)
23:05:49 Join DrMoos [0] (
23:06:12petur(iRiver) the two gain settings (recording settings, recording screen) are very confusing as one is the decimator, the other the analog gain.
23:06:38peturapart from the naming issue, shouldn't it be possible to change both from the recording screen?
23:07:04 Quit Moos (Read error: 104 (Connection reset by peer))
23:07:11 Nick DrMoos is now known as Moos (
23:11:59Papricawhat do you think?
23:12:15mirakhello ! hello !!!
23:13:16Papricait will going out on some days
23:14:36***Saving seen data "./dancer.seen"
23:14:47_FireFly_LinusN: ? any commments about the idea?
23:14:59dropandhoyou scared him away!
23:15:49_FireFly_mayby he is on the way to me to kick my ass into the space ;)
23:16:08LinusNamiconn: yes, it uses the 16-bit 80-system
23:17:15LinusN_FireFly_: looks much like the one recently discussed in this channel
23:17:45LinusNseems to be a good idea
23:18:34_FireFly_yepp it is the idea from amiconn i have put it only into the file and added some suggestion how the data-struct good be defined
23:20:16LearI hope my recent mpeg.c fix actually works. Can't test them in the simulator, and I have no SH1-compiler around at the moment... :)
23:20:20amiconnHaha, the H3x0 display controller offers an inverse display mode. Might look funny with complemented colours :)
23:20:34amiconnFlip display in hardware seems to be possible as well
23:20:58LinusNPaprica: cool
23:21:08amiconnDo we want 8 colour mode for speed? ;)
23:21:10CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
23:21:10*amiconn hides
23:21:12LinusNPaprica: i want it for h100 and archos as well .-)
23:21:26LinusNamiconn: 8-color mode doesn't give a speedup
23:21:34LinusNit's only a power saving mode
23:21:55Papricalet me finish it for h3xx first =]
23:22:57amiconnAh I see.
23:23:23_FireFly_LinusN: btw in the same dir is also a first attempt to define a token-based wps-handling
23:25:49LinusN_FireFly_: the wps paper doesn't mention what problem it attemps to solve
23:27:23dropandhobut the real question is...does it solve it?! hehe
23:28:51_FireFly_it doesn't attemps to solve any problems it is only an idea how the current wps-handling could be rewritten as someone in this channel had said that it should be rewritten
23:29:25_FireFly_the idea was orginaly from TiMiD
23:30:05LinusNyes, it should be rewritten, but not just because we feel the need to rewrite something
23:30:45LinusN_FireFly_: btw, this kind of "papers" really belongs in the wiki
23:31:00Bagderindeed, as that make people actually discover them
23:31:23 Part ratpack91
23:32:31 Join matsl [0] (
23:33:12 Quit matsl (Client Quit)
23:33:32_FireFly_ok i will add these two ones to the wiki
23:36:15peturany dev with an H1xx willing to try the pre-recording patch? (just to be sure it also works ok on that one as well)
23:36:24dropandhothanks for the work tho!
23:36:29dropandhohope they didnt discourage u
23:37:35 Quit San (Read error: 110 (Connection timed out))
23:39:10 Quit Lear ("Chatzilla 0.9.69 [Firefox 1.5/0000000000]")
23:40:09PapricaLinusN, there is a option now to build a clock for h3xx?
23:40:38Papricaok =]

Previous day | Next day