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).

Notice: Only Gecko based browsers prior to FF4 support the multipart/mixed "server push" method used by this log reader to auto-update. Since you do not appear to use such a browser, this page will simply show the current log, and not automatically update.

#rockbox log for 2006-01-13

00:05:59peturLinusN: did you get over my coding-rule breaking patch? I redid the recording gain one - it should apply just fine and follow 'da rules'
00:06:50peturbtw, you misdsed my biggest error: that code should be in backlight.c, not pcfxxx
00:11:10 Quit ender` (" Sleep n. An inadequate substitute for caffeine")
00:14:05 Join Xerion_ [0] (
00:14:07 Quit Xerion (Read error: 104 (Connection reset by peer))
00:14:17 Nick Xerion_ is now known as Xerion (
00:17:11 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
00:18:51 Quit alphakiller ()
00:20:00 Quit petur ("here today, gone tomorrow")
00:21:41XavierGrI have a struct with 4 integers the sizeof that struct is 8 bytes right?
00:22:46markun16 I think
00:22:52markuna short is 2 bytes
00:23:12XavierGrI am always bothered with ints in rockbox, I think someone told that ints are 2 bytes no?
00:23:37linuxstb_On the gmini, ints are 2 bytes. On all other targets, they are 4 bytes.
00:24:34linuxstb_Which is why you should use "long" if you need a 32-bit integer in rockbox - or (u)int32_t
00:24:43XavierGrI tried bat[BUF_SIZE / sizeof(bat)]; to be sure but it failed as bat is not already declared. Of course now that I think of it I could just first declare and then create the struct called bat.
00:25:24linuxstb_So what is the type of bat[] ?
00:25:52XavierGr4 ints
00:26:18linuxstb_So you need to use a struct to store those 4 ints, and then sizeof(mystruct)
00:26:44linuxstb_Which will be 32 for most targets, but 16 for gmini
00:27:04XavierGryes but it is an array that I want so the immediate declaration is not right to do this.
00:27:22XavierGrI will declare the struct first and then create the array using sizeof
00:27:51mirak_the sim have problems opening my files
00:28:20mirak_I set the audio buffer as target si maybe the sim have problem with that ?
00:28:57linuxstb_I think that should work.
00:32:36 Join Skydemon [0] (
00:33:19BHSPitLappylinuxstb_: rockbox doesn't have any ipod capabilities that linux doesn't, does it?
00:34:39 Quit Kohlrabi ("Fast alle Menschen sind Regenw├╝rmer")
00:34:53linuxstb_Not at the moment. But when rockbox is fully functional, there are lots of features in Rockbox which probably aren't available in Linux.
00:35:11BHSPitLappyi just mean capabilities specific to the ipod
00:35:19BHSPitLappye.g. hardware
00:35:29BHSPitLappybut i'll still take that as a no
00:35:33linuxstb_No, we're still playing catch-up
00:36:02BHSPitLappythe nano won't listen to taps, even though it can detect them in diag mode
00:36:47XavierGrlinuxstb_: Is the iPod UMS?
00:36:54XavierGr(in the original firmware I mean)
00:37:44linuxstb_Yes and no. You can access the disk using UMS, but the apple firmware ignores files that aren't referenced in the tag database
00:38:18 Quit t0mas (" good night")
00:38:34XavierGrso that means that if you add your own files?
00:39:01XavierGryou can't see them or they don't get in?
00:39:07linuxstb_The apple firmware only lets you browse the tag database - so if the music isn't there....
00:39:52linuxstb_When iTunes (or the third-party equivalent) adds a music track, it will both copy the file and update the database on the ipod.
00:40:03XavierGrLets say that I have my music on an iPod. I stick it to a PC and want to either add or take some music (or other files), without any program. What happens then?
00:40:49linuxstb_You can't add music, and the music that is already there is stored with obfuscated filenames such as F04/4AAD3.M4A
00:41:22XavierGryaakkk that's why I hate Apple!!!!
00:41:32linuxstb_But this is purely a software issue with the apple firmware. Rockbox will of course play them without problems.
00:42:05XavierGrSwitching Firmware would be a problem then.
00:42:07markunAt least the files are not encrypted like the gigabeat does..
00:42:15linuxstb_Why would you want to switch firmware?
00:42:44XavierGrWell in the mid process which iPod can playback more stable.
00:43:01linuxstb_I worried about this a little when I started work on it, but then decided people who wanted to use Rockbox would not want to use itunes.
00:43:27XavierGrYes but Until proper playback is introduced in Rockbox....
00:43:28Skydemonand maybe some Videos (like irivever h300 atm)
00:44:04linuxstb_I think Rockbox will go from no playback to good playback almost overnight. It won't be like the iriver port.
00:44:14XavierGrwhy is that?
00:44:22XavierGrIt is a software one no?
00:44:35linuxstb_The playback system is already working - it just needs a proper audio driver writing.
00:45:04linuxstb_I can play music on my ipod now, it's just that it's silent... :)(
00:45:29XavierGrlol ;)
00:45:45linuxstb_Of course, we may have codec optimisation to do - so maybe it won't quite be overnight.
00:45:48amiconnlinuxstb: What about arm optimisations?
00:45:48XavierGrIt is like the midi plugin.
00:45:59XavierGrIt will proccess the file but no output, (file or audio)
00:46:27linuxstb_amiconn: Yes, just remembered that :) But the fast codecs like FLAC should be fine.
00:46:36XavierGrlinuxstb_: Can you say if iPod will encode faster than iriver targets?
00:46:56Skydemonbtw - is it now save to use bootloader v5 for iriver? i have no changse to repair a blocked iriver - so wanna go as sure as possible )
00:46:57 Join akaidiot [0] (
00:47:08amiconnlinuxstb: Also, using the COP might introduce problems
00:48:07linuxstb_amiconn: Yes, I'm sure it will.
00:48:42linuxstb_XavierGr: encode what?
00:49:02XavierGrlets say mp3
00:49:20XavierGroh no I mean decode.
00:49:33ashridahSkydemon: easiest way would be to check out some of the bootloader v5 threads on misticriver, but so far as i know, it works, the last version just had a bug that made it take too long to turn off an auto shutoff
00:49:41XavierGranyway CPU power, will it be faster?
00:49:46linuxstb_It's hard to compare - a decoding process will just use one of the two 75MHz ARMs.
00:50:05XavierGrno code to use both?
00:50:17XavierGror it is not inteded to do this
00:50:25Skydemonashridah: Thanx )
00:50:31linuxstb_Yes, but the codec thread itself will just use one
00:50:43lostlogicyou want to write task migration for rockbox? :-D :-D
00:50:45linuxstb_The plan is to use one CPU for decoding, and the other for everything else.
00:51:07linuxstb_Our new ipod developer threatened to work on that.
00:51:27XavierGrthat would be tricky for some plugins like rockboy which needs every juice it can get.
00:51:33lostlogicalphasomething? He seems... frighteningly enthused.
00:51:54linuxstb_He may well get distracted with his java OS
00:53:37linuxstb_XavierGr: Going back to your original question, the .m4a files have the metadata (tags) still in them, so Slasheri's new tagcache will be able to index them.
00:54:08XavierGrIs there a patch for the tagcache?
00:55:15XavierGrSlasheri can have many different nicks. amiconn proposed Flasheri but maybe Cacheri is a good one too. 2 cache projects from him so far.
00:56:18*amiconn wonders what happened to his flash project...
00:57:30XavierGrThe Flasheri Project ;)
01:00:06XavierGrOkay lets say that someone tried to benchmark his battery in radio mode. Currently in my battery_bench plugin battery info will be loged on disk access. What about setting a time interval when there is no file playback? And what would that interval be? 1 hour, 2?
01:01:16 Quit muesli__ ("ich will KŘhe!!!")
01:02:44 Quit |Beowulf| ("KVIrc 'Marmalade'")
01:03:28lostlogicLinusN: is the I2C only meant to function at 120mhz on H340, or is 400khz the maximum frequency?
01:06:57 Quit actionshrimp ("a bird in the bush is worth two in your house")
01:07:31 Join webguest96 [0] (
01:11:39 Quit Moos ("Happy new year @ all")
01:12:05LinusNlostlogic: 400khz is max, and the timing loops are adjusted to reach 400khz at 120Mhz cpu frequency
01:12:14XavierGrhow many HZ is an hour in rocbox?
01:12:53XavierGrHZ=60 ticks or 100?
01:13:03XavierGroh 60
01:13:04Skydemonok - h320 -> firmware eu 1.30 -> bootloader v5 = works fine :)
01:13:58 Quit Janco ("CGI:IRC (EOF)")
01:14:10LinusNHZ is HZ ticks
01:14:30LinusNthe whole point of HZ is that you shouldn't care how many ticks it is
01:14:40LinusNHZ is one second
01:14:56LinusNor rather "number of ticks per second"
01:15:12XavierGrokay thanks.
01:15:24LinusNso, an hour is HZ*60*60 ticks
01:16:44 Quit Skydemon ("Nettalk6 -")
01:21:23 Part LinusN
01:21:59 Join Kyl3 [0] (
01:22:16Kyl3how do i get bmp2rb to work?
01:22:48Kyl3im in the tools directory it says command not found
01:23:23lostlogicwhat OS?
01:23:25linuxstb_Type ./bmp2rb
01:23:37Kyl3win in the dev kit
01:24:13Kyl3BASH: ./bmp2rb: No such file or directory
01:24:19linuxstb_Type "make bmp2rb"
01:24:59Kyl3in tools right?
01:26:02Kyl3and the h300 lcd takes 16 bit bmps, for say, the bootscreen?
01:28:01Kyl3bmp2rb -i h300optFULL.bmp -f4 that seems to be correct but its not working
01:29:50Kyl3oh wait
01:29:50linuxstb_You may need a space between the f and the 4
01:29:55Kyl3i did it
01:30:07Kyl3but how do i copy this i cant scroll up
01:30:36 Quit fobbo (Read error: 104 (Connection reset by peer))
01:30:50linuxstb_Type bmp2rb ..... > myfile.c
01:45:26 Quit webguest96 ("CGI:IRC")
01:54:45***Saving seen data "./dancer.seen"
01:56:06 Quit Mark_ (Read error: 104 (Connection reset by peer))
01:56:15 Quit Kyl3 ("( :: NoNameScript 3.81 :: )")
01:57:48manhattanSo like the creative zen stuff is probably never going to be supported by rockbox?
02:00:57 Join Kyl3 [0] (
02:00:58BHSPitLappyanybody care for this at all? −−>
02:02:08linuxstb_manhattan: No-one has shown any interest in it so far.
02:02:29Kyl3whats the size of the boot logo on the h300 screen
02:02:35manhattanno-one as in the coders that be working on this, or it's just not popular
02:02:37BHSPitLappylinuxstb_: i desperately crave your approval
02:02:50Kyl3im changing it for the "optimized" build on Misticriver
02:03:50linuxstb_manhattan: No-one has shown any interest in porting Rockbox to it.
02:04:01manhattangotcha :)
02:04:18 Quit akaidiot (Connection timed out)
02:04:28 Join UmInAsHoE [0] (
02:04:33Kyl3and it worked but didnt display right i guess cause the size was messed up
02:04:48linuxstb_BHSPitLappy: approval of what?
02:05:06BHSPitLappymy little picture a few lines up :P
02:05:16Kyl3I like it
02:05:20linuxstb_Kyl3: 220x68 I think.
02:05:30UmInAsHoEwhats the fwppatcher?
02:05:34Kyl3:) thanks!
02:06:16linuxstb_BHSPitLappy: It's very nice. You should design some graphics for Rockbox games
02:06:30BHSPitLappyha, flattered
02:06:49BHSPitLappywell if i'm actually needed somewhere, i'll help.
02:06:59*manhattan .oO(┐gamez?)
02:07:01BHSPitLappydoubt that i'm the best gimp'er around here though :P
02:07:24*manhattan .oO(┐what's a gimp'er?)
02:07:26 Join einhirn [0] (
02:07:45linuxstb_There aren't many artists amongst the rockbox devs...
02:08:18BHSPitLappymehh :)
02:08:37BHSPitLappylike i said, "if you request it, i will come"
02:09:09BHSPitLappyrather, "I'll try my best."
02:12:58webguest02can someone tell me what the maximum allowed length of a line in a WPS is?
02:14:42linuxstb_I don't think there is one. Are you having problems?
02:15:17webguest02i havent yet tried if it works
02:15:40linuxstb_AFAIK, the only limit is the total size of the wps file
02:15:57webguest02but i was curious, because it contains 370 characters
02:16:22webguest02what is the maximum size of the WPS?
02:18:48linuxstb_I think it's 3072 bytes
02:21:01webguest02ah thanks
02:21:20XavierGrHey BHSPitLappy very nice picture. What's its purpose?
02:21:51BHSPitLappyit's one of those forum signature pics that are all-so-popular
02:22:33XavierGrGet in touch with the (I dont remember its name) that ports colour games in H300
02:23:01XavierGryeah maybe...
02:28:28BHSPitMonkeyif he wants my help, he can contact me...otherwise, i'm not LOOKING for more expenses of my time ;)
02:32:51 Join leftright [0] (
02:34:57leftrightugghh, I'm listening to a album which has some serious artifacts, (mp3), and I put it down to shit mastering/producing, I have tried the original firmaware as well same result pisses me off
02:35:38leftrightall my stuf is riped with EAC- Lame-aps
02:36:02XavierGr is this possible with flash players?
02:36:41 Join aliask [0] (
02:37:13dpassen1leftright: what album? by chance maybe someone else has it
02:37:40leftrightMulholland Drive Original movie soundtrack
02:38:23linuxstb_What kind of artifacts? Have you listened to the original CD?
02:38:34leftrightthe track "The Beast" is the worst
02:39:16dpassen1could it be clipping? are your tracks ReplayGained?
02:39:26leftrightwell at random intervals and on highs or complex passages there's "crackling" (thats the best word for it)
02:39:55leftrightyes they are all replay gained, but it doesnt do it on other albums, nut I'll try it with RG off
02:40:11dpassen1do you have a pre-amp set?
02:40:23leftrightbut low
02:40:32 Quit Kyl3 ("( :: NoNameScript 3.81 :: )")
02:40:41dpassen1try without the pre-amp before you try without replaygain
02:41:15dpassen1also, are you using any bass/treble boost?
02:41:56leftrightonly bass at 8
02:42:10leftrightvol is at -26
02:42:36dpassen1if these files sound alright on your computer, i would just try to obtain as flat a setting as possible (preamp and all boosts at 0) and check
02:42:55 Join DJDD_ [0] (
02:43:09 Quit einhirn (Read error: 113 (No route to host))
02:45:32leftrightnah still there with pre-amp off and then with RG off, no difference
02:46:07dpassen1do the files sound different on a software player (winamp, foobar, etc.)?
02:46:47leftrightgahhhh, I hate shite mastering
02:47:38leftrightdont have those available to me right now, not at home
02:48:01BHSPitMonkeyi keep foobar on my ipod nano :)
02:48:21 Join Jungti1234 [0] (n=jungti12@
02:48:37leftrightwhat makes me certain is that I have some excellently mastered albums, and they sound just fine
02:49:13markunleftright: hm, I think I also here it :(
02:49:31leftrightyou have the same album ?
02:49:51linuxstb_Is that a lame bug?
02:49:52leftrightlisten to 'The Bear" its awful
02:50:05linuxstb_How does the CD sound in a CD player?
02:50:27markunDo you hear it "The Beast" at 1:19
02:50:30markunfor example
02:50:47markunlinuxstb_: I have it in Ogg Vorbis
02:51:17leftrightyep its there all right :-)
02:52:14leftrightoh that is a great movie by the way
02:53:13 Join UmI [0] (
02:53:13 Quit UmInAsHoE (Read error: 104 (Connection reset by peer))
02:53:17markunI also have similar artifacts in "Norah Jones - Don't know why", but that might be the rip.
02:53:43 Quit UmI (Remote closed the connection)
02:54:14leftrightyour files are encoded in ogg ?
02:54:29leftrighthmm mine EAC- Lame-aps
02:55:16markunq3, not very high, but probably not the cause of it.
02:55:31leftrightevery now and then there's a shite album, I would like to think it's bad recording/mastering
02:55:40 Quit linuxstb_ ("CGI:IRC")
03:02:30leftrightthanks for the ears. later
03:02:54Midgey34Yay! My plugin almost works now
03:03:37 Part leftright
03:18:18 Quit mikearthur (Read error: 104 (Connection reset by peer))
03:32:41aliaskLooking forward to it Midgey34
03:41:36lostlogicbah, new H340 charge socket damaged. wher'es my solder.
03:46:51 Join akaidiot [0] (
03:50:09Midgey34at least you have an h300 again
03:54:48***Saving seen data "./dancer.seen"
03:56:10lostlogicugh, now I have to reassemble it with the right assortment of parts from the old and new players so that it might work :-ED
04:08:19 Quit webguest02 ("CGI:IRC (EOF)")
04:08:59XavierGris that poker Midgey34?
04:13:16Midgey34no, blackjack
04:13:28Midgey34just getting the hang of things right now
04:14:55lostlogicI fucking hate it when I forget to connect the battery and have it all screwed back together.
04:21:58 Quit ashridah ("Leaving")
04:22:21 Join ashridah [0] (
04:23:27Jungti1234Hate iriver! Hate iriver! Hate iriver!
04:24:17Jungti1234Mad iriver! Mad iriver! Mad iriver!
04:24:57Jungti1234Iriver that ignore users!
04:27:37 Join ts-x [0] (
04:28:03lostlogic... ... ?
04:28:55ts-xHey does anyone here use the H1xx remote with their H3xx? I could be an idiot, but it doesn't seem to be working with recent builds...
04:30:26ts-xThe buttons work, but the display is blank. I've tried loading several rwps files with no luck.
04:39:28aliaskWhat happened Jungti?
04:47:43Jungti1234I will not buy iriver.
04:48:02Jungti1234Their confidence is lowest.
04:50:09Jungti1234I think that iriver may perish soon if iriver doesn't recognize H300 LCD problem.
04:50:28 Join egotrippen [0] (
04:51:53Midgey34which problem?
04:52:09Jungti1234and I think that they deceived expense of parts.
04:53:06Jungti1234I spoke that there is problem in LCD angle of vision for a long time.
04:53:31Midgey34as in, you can't see it when tilted?
04:54:18Jungti1234And it makes eye easily tiredly.
04:54:34lostlogicI don't think that's a problem they should have to deal with.
04:54:38Midgey34well, that's just the type of lcd it is, now that the hardware is out they can't fix it
04:55:50Jungti1234H300 LCD is new LCD and old LCD.
04:55:57 Part ts-x
04:56:14lostlogicah, didn't know that there were two versions of it
04:56:16Jungti1234I have old LCD and feel that there is problem.
04:56:16lostlogicmakes sense
04:56:35Midgey34oh there are two versions?
04:56:54aliaskIs the new version the one with the problem, or the old one?
04:57:38Jungti1234Old version is problem.
04:57:57aliaskSo you want a new LCD for your old LCD?
04:58:49Jungti1234And H300's pattern is two. fractal and pinspot
04:59:04Jungti1234Pinspot is old version.
04:59:35egotrippeni don't think that's true
04:59:44egotrippenthe LCD is the same, just the outer casing is different
05:00:09Jungti1234Iriver recognized that it exists.
05:00:09egotrippenisn't it? i'd never heard before there were two LCD versions
05:00:52Midgey34I want iriver to fix their stupid RTC
05:01:01Midgey34Feb 30th is not a date
05:01:22Midgey34and Feb 29 does not exist in 2005, 2006, 2007
05:01:35 Join Rob2222_ [0] (
05:04:21 Quit akaidiot (Connection timed out)
05:06:47Jungti1234And H300 portable case is two.
05:07:31Midgey34I thought there was three?
05:07:43Midgey34its not like I use it, its not very good
05:07:55Jungti1234three problem
05:08:27Jungti1234Iriver changed old case by new case to event.
05:08:52Jungti1234Event ends and iriver spoke that don't change old case by new case.
05:09:08Jungti1234But, they changed it!
05:09:25Jungti1234Changed it some people.
05:09:26egotrippeniriverAmerica is even worse
05:09:53egotrippentheir american branch has no idea what the korean branch is doing
05:10:00egotrippenand the korean branch does everything important
05:10:33egotrippeni had one of the iMP CD players, the american firmware in America was version 1.03 or something
05:10:43egotrippenthe korean firmware was 1.25, something like that
05:11:18egotrippensimilar to the H300s, i guess, and they only updated the American firmware once
05:11:25egotrippenbecause a LOT of people complained
05:12:03egotrippenthey're good with RMAs, but not about information or updates
05:14:43egotrippenhow much of that translated properly?
05:14:56egotrippenthose programs can be poor
05:15:12Jungti1234I translate too.
05:15:39egotrippeni read the IRC logs sometimes, i thought i read you used a program
05:16:04Jungti1234Google translation is not good.
05:16:21egotrippeni know
05:16:33egotrippenmachine translation is very rough
05:17:59 Quit Rob2222 (Read error: 110 (Connection timed out))
05:18:00Jungti1234But, we don't speak English.
05:18:13Jungti1234Must use translation.
05:18:28Jungti1234Until learn English.
05:19:06Jungti1234Doesn't US firmware support video?
05:19:12Jungti1234and browser?
05:19:21egotrippenDRM instead of video
05:19:25Jungti1234But, support DRM.
05:19:28egotrippeniriver has said that there's not room for both
05:19:41egotrippenand the browser isn't there because USB-OTG isn't there
05:19:51egotrippennot space, memory
05:20:09Jungti1234But, it's possible.
05:20:25egotrippenthe US models can use korean firmware
05:20:32egotrippenbut they lose DRM permanently
05:20:49Jungti1234It's dangerous action.
05:20:56 Nick drumrboy is now known as DrumRBoy320|Away (
05:20:59egotrippenonly if you use DRM
05:21:39Jungti1234Korea model can't use US firmware.
05:21:52egotrippendon't know why you'd want to, anyway
05:22:02Jungti1234korea = International?
05:22:07ashridahso reflashing with the US firmware again doesn't regain you the DRM key?
05:22:35egotrippenUS can flash to US and keep it, but flashing to international firmware loses it permanently
05:22:36ashridahstrange effect.
05:22:53egotrippenit must overwrite the part of the memory that keeps the key
05:23:02Jungti1234use US firmware in international model, don't find DRM key.
05:23:26Jungti1234Music isnt played.
05:23:34egotrippenonce a player, american or international, has international firmware on it it cannot use DRM
05:24:17Jungti1234Don't support video in EU firmware.
05:24:34Jungti1234But, I saw.
05:24:42Jungti1234Yesterday I talked.
05:25:11Jungti12341.29EU supports video. but, don't support in 1.30EU.
05:25:31egotrippeni don't know why they would do that. seems very strange
05:25:33Jungti1234I used all firmwares.
05:25:45egotrippenkorean seems the best
05:25:59egotrippenno volume cap like EU, and more radio frequencies than J
05:26:20egotrippenbut i've been using Rockbox since it came out, anyway
05:26:46Jungti1234But, it's Korean.
05:27:35egotrippeni'm working on fixing a little bug
05:27:40egotrippenor, adjusting a feature, i guess
05:27:59egotrippenit's complicated and i don't know C very well
05:28:17egotrippenbut not important enough for a serious programmer to work on, i think
05:28:29Jungti1234I dont know entirely C.
05:28:44egotrippeni know very little
05:28:52egotrippeni'm guessing and checking
05:29:00Jungti1234I can't work in rockbox. :(
05:29:11Jungti1234Translate only.
05:29:29egotrippeni can understand some of it, but not write really
05:29:45Midgey34what are you trying to do?
05:29:47egotrippeni read the part i want changed, and ask for what i want
05:30:00egotrippenon the H300, you can change the colors....
05:30:15egotrippenand with the 'inverse line selector,' it looks bad
05:30:35egotrippenwith black background, orange foreground, i get white background and light blue text
05:30:49egotrippenso i'm trying to make it reverse, just swap the BG and FG colors
05:30:54egotrippeninstead of inverting
05:31:06egotrippenthe code is in lcd-16bit.c
05:31:19egotrippenand this is some general stuff
05:31:31Midgey34ah ok
05:31:37egotrippeni'm fiddling with it, trying to get it to do what i want
05:32:00egotrippenit's mostly a hassle because i'm on XP, and making a build is slow
05:32:17Jungti1234me too
05:32:24egotrippenalso, the inverse bit seems to be connected to some other things
05:33:45egotrippenit might need more of a change to seperate the line selector from the rest of the color code
05:33:50egotrippenwhich would definately be over my head
05:36:45Jungti1234Translation is strange.
05:37:19egotrippenit says korean to english is in beta
05:37:26egotrippenso, even worse than usual
05:37:58Jungti1234Many Koreans know gravity of problem.
05:38:45egotrippenwell, i got the foreground to become the background
05:38:52egotrippenit's still inverted, i just can't read anything
05:39:05egotrippenand the WPS screen disappears, too
05:39:30egotrippeni think i'm giving up on this one
05:40:15Jungti1234I can't help to you. :(
05:40:44egotrippenit's alright
05:40:55egotrippeni'll post what i found to the rockbox and H300 forums
05:41:08egotrippenmaybe someone who knows more will take it upon themselves
05:41:59egotrippenthe real programmers seem happy that someone tried, at least, before making a request
05:43:20Jungti1234I want that use color BMP in WPS.
05:43:50egotrippenthe code i'm messing with will probably have to be rewritten for taht to happen, too
05:44:01egotrippenchanging the colors i was changes the colors of the WPS bitmaps, also
05:44:07egotrippeni was changing^
05:44:21egotrippener, lemme try that again
05:44:32egotrippenchanging the colors i was changing changes the colors of the WPS bitmaps, also
05:45:35aliaskI'm messing around with colour bitmaps at the moment, adding transparency and some functions, but because I suck so badly at C, I'll probably never do anything useful.
05:46:03aliaskAnd also now that rockboy is running much better, more of my time is dedicated to pokemon. :(
05:46:19Jungti1234me too
05:46:56egotrippeni haven't had the chance to used the improved rockboy yet
05:47:07egotrippenmy battery is dying, i think. just ordered a 2200 mAh one
05:47:39aliaskIs the original 1300? You'll get a nice boost in life if it is.
05:47:50egotrippenyeah, that's right
05:48:02egotrippena fresh 2200 brings it to 20-24 hours
05:48:10Jungti1234Many Koreans changed battery.
05:48:34aliaskFor iRiver firmware, at least.
05:48:55Jungti1234It's 2100mAh battery of ipod.
05:48:55egotrippenit is tempting to blame rockbox for the decline in battery time i've seen recently, but i think it's a coincidence
05:49:14egotrippenit's a couple months more than a year old, about time for the battery to go
05:49:34aliaskDon't say that, my battery is a year old too D:
05:49:48egotrippenwell, good luck, but the new batteries are cheap
05:49:59Jungti1234egotrippen, where are you from?
05:49:59aliaskHow much?
05:50:19egotrippeni'm from America, boston and just started going to college in new york
05:50:47egotrippenthe 2200 mAh are $30 plus shipping on real online retailers
05:50:52egotrippenthe brand is newertech
05:51:04aliaskHrm not bad at all.
05:51:04egotrippenbut they go on eBay for around $20 shipping included
05:51:17Jungti1234A person who have H300 in Korea exceeds ten thousand people.
05:51:38egotrippenipod is 98% of market share in america
05:51:40egotrippensomething like that
05:51:46Jungti1234Perhaps, I think so.
05:51:54egotrippenwith ipod batteries, you have to change the wires
05:52:00egotrippenthere are a couple guides on Misticriver
05:52:23egotrippenit doesn't seem too hard, and no soldering involved inless you screw up :)
05:52:57 Nick DT291 is now known as DreamTactix291 (
05:52:58Jungti1234Many Korean teenagerses have mp3 player and cellphone. :)
05:52:59egotrippeni just ordered a new battery, i'm hoping it comes before i have to make the drive back up to school
05:53:10egotrippen4-8 hour drive, depending on traffic and how lost i get
05:53:23egotrippenhere too
05:53:33Jungti1234um..? really?
05:53:50eEyeshould i go for a h300 or an X5L?
05:53:55egotrippenmaybe not as many
05:54:32Jungti1234Because there is no LCD's problem, X5L is better than H300.
05:54:52***Saving seen data "./dancer.seen"
05:54:58egotrippeneEye: with no knowledge of X5L specs, the H300 is a great machine and has rockbox already
05:55:35eEyeim just a little iffy with the size of the h300
05:55:38egotrippenat a year old, it's held its own very well in the market. i don't know how well it compares specifically, but i continue to be impressed with it
05:55:39lostlogicX5L better battery life, H300 better LCD, X5L smaller, better processor.
05:56:41egotrippeni find the H340 size comfortable, without a case
05:56:48egotrippeni go for the adhesive screen protectors
05:57:01egotrippenbut battery life is a major consideration
05:57:15eEyeis that a good price?
05:57:50 Join Paul_The_Nerd [0] (
05:57:50egotrippen believe so
05:58:08egotrippen$200-$250 seems like a good range, but mind i haven't done much research
05:58:21egotrippeni got the H340 a year ago, never really looked at the '20
05:58:32egotrippeni've got too much music
05:59:14egotrippenoh, it's got USB host
05:59:40egotrippenthat's an international model, if it's got USB host, which makes it worth more. USB-OTG and instead of DRM
06:00:07eEyehow much music?
06:00:08Jungti1234H320 is about $205 in the Korea.
06:00:21eEyeprices in canada suck ;/
06:00:23egotrippenyou won't be able to play songs rented through napster, but if you get a card reader you'll be able to store digital camera photos or whatever on it
06:00:48Jungti1234It's no longer produced.
06:00:59aliaskX5 also has USBOTG i think.
06:01:06Jungti1234H340 is $285.
06:01:17egotrippeni've had ~30-35 gigs for a long time
06:01:40eEyei just hit 60gb ;/
06:01:56Jungti1234X5 is 60GB, and USB Host(USB 2.0, OTG) support
06:01:57eEyethere is about 20gb of stuff i probably never even listened to yet
06:02:08egotrippenword is they're about to release 80 gig drives that fit the H340
06:02:09eEyeor that i currently dont listen to anymore
06:02:17eEyeoh really?
06:02:30egotrippenso if you want to pay for it, you'll be able to get a 40 gig upgrade for the H320
06:02:32 Join webguest25 [0] (
06:02:50egotrippennew drives that fit 'em cost a lot, though. almost as much as the player itself
06:02:50Paul_The_NerdThe 40 gig exists already, though I don't think it's on sale individually yet.
06:02:56Jungti1234hey egotrippen
06:03:06Paul_The_NerdThe 80 gig dual platter should be in players *very* soon.
06:03:07Jungti1234Where is it?
06:03:41egotrippenthis is just what i've heard from misticriver. i don't think they're on the market yet
06:04:03egotrippenword is, i guess, "*very* soon"
06:04:11Paul_The_NerdMK8007GAH I believe
06:04:18Paul_The_NerdYes, perpendicular.
06:04:33webguest25will they work though? besides the physical fitting prob?
06:04:45egotrippen60 gig drives have been used already
06:05:09webguest25so for h120, only upto 40 would be possible then, as of now?
06:05:16egotrippenand i believe someone has plugged a regular sized 80 gig drive into a H100 and it recognized it
06:05:26Paul_The_NerdApparently the 80gig is on sale in at least one place, supposedly
06:05:46webguest25400 bucks!
06:05:53egotrippenthey're pricy
06:05:54webguest25i'd just get another player :p
06:06:04egotrippenespecially because they're brand new - they'll come down a bit
06:06:11Paul_The_NerdYeah, give it a little time.
06:06:12webguest25yeah i guess
06:06:14egotrippenbit they'll still be, well, pricy
06:06:19egotrippenbut they'll^
06:06:23Paul_The_NerdBut then so will players with them in it.
06:06:35webguest25high cap batt + those hdd would be nice lil upgrade for good ol h1xx's
06:06:43egotrippenthe 60 gig ipod photo was around 600 when it came out, if i remember correctly
06:07:05Jungti1234very slow
06:07:20webguest25now where would i get those color screens... *takes a look at the cellphone in a menacing way*
06:08:47egotrippeni'm amazed at how well the H300 holds up to today's market, seeing as it's over a year old
06:08:47 Quit webguest25 ("CGI:IRC (EOF)")
06:09:07 Part Midgey34
06:09:15egotrippeniAudio has offered some competition, but nothing clearly, no-argument-possible better
06:09:17aliaskIt was pretty ahead of its time when it came out.
06:09:51aliaskVideo and USB Host were completely unheard of.
06:10:11Jungti1234They differentiate it.
06:10:54Jungti1234Vice corporation
06:15:39Jungti1234As there is English, read.
06:16:14egotrippenyeah, it's true
06:16:20egotrippenenglish is rediculously complex
06:16:49Jungti1234what is 'rediculously'
06:17:01egotrippenridiculously, i think
06:17:14egotrippenin this context, 'very very'
06:17:28 Quit Nibbler (Read error: 110 (Connection timed out))
06:17:43 Join Nibbler [0] (
06:17:46egotrippeni think because it has been influenced by so many other languages
06:17:50Jungti1234Korean is easy!
06:17:57egotrippenit's germanic, but heavily influenced by french
06:18:10egotrippeni have a friend learning japanese
06:18:17egotrippenthe grammar is so much more straightforward
06:19:02aliaskFro what I hear, the grammar is identical.
06:19:12egotrippenin japanese and korean?
06:19:46egotrippeni'll be taking chinese in school soon
06:19:48Jungti1234Korean is beginning of Japanese.
06:19:49egotrippeni'm a little scared
06:20:07egotrippenit's all pictorams, no phonetics like japanese
06:21:38aliaskKorean is all phonetics, the alphabet is pretty logical too and easy to learn.
06:21:49egotrippenjapanese has both
06:22:07aliaskYeah, I've learnt japanese for about 10 years now.
06:22:23egotrippen:) cool
06:22:23 Quit egotrippen ("CGI:IRC (EOF)")
06:23:03Jungti1234web irc? :)
06:24:08 Join egotrippen [0] (
06:24:38aliaskAh that's why I thought you were from sweden, you were using web irc...
06:24:54egotrippennah, i just don't use IRC much
06:24:57Jungti1234Do you use Web irc?
06:25:30egotrippeni am now
06:25:34egotrippeni don't have a real IRC client
06:26:04egotrippeni was about to say, i'm taking chinese to read about Zen
06:26:44egotrippeni've read some books by Alan Watts, and i want to know more about the older practices
06:26:52egotrippenwhen it was still mainly chinese, and not yet japanese
06:27:32egotrippenkorean either, i suppose
06:27:54egotrippenit began in india, i read, migrated to china, and from there to japan
06:28:37Jungti1234If you understand it, can do easily Korean.
06:30:22Jungti1234I think that had spoken it to aliask.
06:30:38aliaskYes, I started learning Hangul.
06:31:04egotrippenfrom what i've read so far, though, the masters i am interested in learning from spoke and wrote chinese
06:31:12egotrippenthough, i may well move on to other languages after that
06:31:16 Part Paul_The_Nerd
06:31:20Jungti1234Chinese is hard.
06:31:35 Quit egotrippen ("CGI:IRC (EOF)")
06:31:35 Quit stamppot ("CGI:IRC (EOF)")
06:31:35 Quit adamdullenty ("CGI:IRC (EOF)")
06:31:36 Quit adamd ("CGI:IRC (EOF)")
06:32:08 Join egotrippen [0] (
06:32:14egotrippeni should get a real IRC client
06:32:21egotrippenyes, i know it's hard. i'm nervous
06:32:28egotrippenbut, i'm willing to try
06:32:51egotrippenwill anyone be mad we're not discussing pure rockbox-related material?
06:33:03Jungti1234Korea, China, Japan learns Chinese character.
06:33:15aliaskWe're wasting space on the IRC logs :O
06:33:30aliaskAnyway, I'm going to work, cya.
06:33:42 Quit aliask (""Hi ho, hi ho..."")
06:33:56 Quit egotrippen (Client Quit)
06:34:22 Quit Jungti1234 ()
06:34:42 Join egotrippen [0] (
06:34:55egotrippeni'm gonna go too, the web client is getting uppity
06:35:22egotrippennice talking with you, i hope you can get your screen issue sorted out
06:37:36 Quit egotrippen (Client Quit)
06:46:43 Join akaidiot [0] (
06:49:38 Quit DJDD_ (Read error: 110 (Connection timed out))
06:56:51 Quit RotAtoR ()
06:59:29 Quit DreamTactix291 (Read error: 110 (Connection timed out))
07:04:32 Quit Maxime ()
07:35:01 Join B4gder [0] (
07:35:39 Join Membrillo [0] (n=sam_kill@
07:38:04Membrillodoes anyone know if you could fit a 40gb single platter hardrive and a 2100\2200 mah battery in an H320?
07:38:51B4gderI would assume so
07:39:45Membrillois a 40gb HDD the same physical size as the original 20gb in the player/
07:40:16B4gderafaik, yes. single-platter 1.8" disks should be the same physically
07:40:42Membrilloi think i might splash out and 'pimp my player'
07:40:58Membrillo60\80 arent realeased yet are they
07:41:30B4gderthey're very hard to get at least
07:41:44B4gderthe 80 ones are still only talk
07:41:54Membrillowell i wouldnt need an 80
07:41:59Membrillobut 60 might be nice
07:42:11Membrillohow would you track one down? eBay?
07:42:17B4gderno idea
07:42:25B4gderebay is probably a good idea
07:42:37Membrillodo you have the model number of the 60gb in your head?
07:42:48Membrillomight have a look around
07:43:02B4gderno I don't
07:45:14Membrillothats it
07:45:33B4gderis that single platter?
07:46:21Membrillooops sorry it isnt. Thats a notebook HD
07:50:31Membrillowhat countries internet code is .ch?
07:54:04Membrillomaybe i could buy a bricked h340
07:54:10Membrilloand steal the HD
07:54:19B4gderno, it doesn't fit in a h320
07:54:33B4gderthe 340s use dual-platter disks
07:54:51B4gderat least that's how the h120 h140 differs
07:54:55***Saving seen data "./dancer.seen"
07:54:57Membrilloreally? oh ok
07:55:05Membrilloi thought h340 used single platter too
07:55:17B4gderI'm actually not 100% on this
07:55:53Membrilloi dont think ive hears anywhere about h340s using dual platter
07:56:08Membrilloive heard a lot of hype about single platters though
07:56:15B4gderI guess you can find out the model name and then lookup the specs
07:56:54Membrillomight look around MR for a while
08:04:13MembrilloB4gder: check this. it needs tidying up but its pretty useful i think
08:04:22Membrilloreplace the old exe with that
08:04:31 Quit akaidiot (Read error: 110 (Connection timed out))
08:04:43 Join Kohlrabi [0] (
08:06:28B4gderI'm a linux dude, you know
08:07:55Membrilloi thought you were commenting on the app before
08:08:01Membrillomust of been someone else
08:10:55 Quit Slasheri (Read error: 145 (Connection timed out))
08:12:36 Join Slasheri [0] (
08:13:02BgerB4gder, Membrillo : yes, the h340 is with dual platter HDD too
08:14:09Bgermorning, amiconn
08:14:33Bgeror mooning, if you prefer
08:15:26 Quit darkless (Client Quit)
08:17:03amiconnAfaiu, the Toshiba 1.8" HD names use the following convention: MK-nnmmGAx where nn is the capacity in GB, mm is some kind of generation and x indicates the number of platters
08:17:29amiconn...GAL are single platter, ...GAH are dual platter
08:18:43Bgermaybe (L)ow and (H)igh ..
08:22:38Membrillookey dokey
08:24:45 Join ender` [0] (i=ychat@
08:27:38Membrilloso what single platters hds are easily available
08:27:42Membrillowhat size (gb)
08:29:07 Join Janco [0] (
08:30:17Membrilloapparently not 60gb
08:30:29 Join LinusN [0] (
08:31:44 Quit Slasheri (Read error: 145 (Connection timed out))
08:34:07 Join Slasheri [0] (
08:35:59 Quit Membrillo ()
08:38:52 Join DreamTactix291 [0] (
08:42:55 Join linuxstb_ [0] (
08:44:20LinusNanyone what?
08:44:30 Quit linuxstb (Read error: 110 (Connection timed out))
08:44:36BHSPitLappyanyone wanna PARTY
08:45:59LinusNBHSPitLappy: sure, i always start my day with a party
08:46:15*BHSPitLappy techno music
08:46:19B4gderthat's what life in the fast lane is all about
08:46:43BHSPitLappyeverybody whip out your mp3 players! time for a digital party!
08:47:27dwihnoEnjoy the digital ... flow?
08:47:38dwihnoWhat _is_ the iriver slogan?
08:47:46dwihnoFeel the digital snow?
08:47:55*ashridah gets his player out and tosses on some beethoven
08:47:59B4gder"eat the digital cow"
08:48:03*ashridah starts headbanging
08:48:04BHSPitLappyG.I. the digital Joe?
08:48:17*BHSPitLappy cranks up some Gershwin
08:48:30*BHSPitLappy dances like an iPod commercial maniac
08:53:22 Join Zagor [0] (
08:54:11 Join einhirn [0] (
08:55:57 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
09:02:01 Quit DreamTactix291 (Read error: 110 (Connection timed out))
09:15:24*preglow joins in on the partying!
09:17:40 Quit Janco ("CGI:IRC (EOF)")
09:23:44*LinusN falls asleep in a corner
09:25:07*BHSPitLappy passes preglow a questionable substance
09:25:16*ashridah slips a roofie in LinusN's drink
09:25:21*BHSPitLappy then proceeds to write on LinusN with a sharpie
09:25:31BHSPitLappyhe's already asleep, ASS
09:27:14ashridahBHSPitLappy: so? doesn't mean he won't be thirsty when he gets up
09:27:49LinusN"i'll never drink again..."
09:35:15 Join darkless [0] (
09:39:08*preglow tries coding on questionable substances
09:42:42 Quit YouCeyE ("Leaving")
09:44:56 Quit solexx (Read error: 110 (Connection timed out))
09:46:01LinusNwhisky on the rocks (baaaad joke)
09:46:45 Join akaidiot [0] (
09:51:24preglowanyone actually working on recording to other formats than wav?
09:52:05B4gderI don't think so
09:54:58***Saving seen data "./dancer.seen"
09:56:50ashridahdidn't someone just submit an encoder to cvs?
09:57:47LinusNwe have two encoders in cvs
09:58:07B4gderwe do
09:58:09LinusNthat doesn't mean that someone is working on realtime encoding
09:58:27B4gderthey're both encoder
09:58:31B4gderpost encoders
09:58:38ashridahonly one i can see in cvs is the shine one, but i don't knw how far back of my archives go or if the filenames include "enc"
09:59:00 Join Paul_The_Nerd [0] (
09:59:09B4gderwav2wv is an encoder as well
09:59:13 Join DangerousDan [0] (
10:02:05perplexityHrm.. Bootloader V5 on h340 says it's in "bootloader usb mode" but does not actually appear to enable the usb hardware
10:02:40ashridahapparently it takes a few because it shuts down and starts up the hd again for some unknown reason
10:02:41preglowat least both of them are capable of realtime encoding
10:02:43ashridahcould be that
10:02:48preglowand the shine one can see more work
10:02:50preglowboth in quality and speed
10:03:46LinusNperplexity: yes, there seems so be issues with the bootloader usb mode
10:04:02perplexityOk, ta LinusN. at least it's a known issue :\
10:09:32preglowrealtime audio in linux is... questionable...
10:13:29 Join amiconn_ [0] (
10:14:05preglowit skips when i switch desktops...
10:14:08preglowand scroll in windows
10:14:31preglowi'm talking low-latency here as well, but not too low, i'm on around 20ms latency now
10:14:54perplexityI solved the scroll problem years ago with enabling PCIretry in my BIOS, but it still skips if I swtich to/from X while it's playing
10:15:25preglowdo i still need to patch 2.6 for decent realtime audio?
10:16:14perplexityyou could certainly apply ingo
10:16:19perplexityingo's rt patchset
10:16:38preglowwhy isn't this in the kernel yet?
10:16:43perplexityit's supposed to be the dogs bollocks for low latency audio
10:17:15perplexitybecause it is still being knocked in to acceptable shape.. currently it still touches a couple of core subsystems in not quite acceptable to Linus ways..
10:17:42perplexityit's slowly being merged through -mm and it's a lot better than it used to be
10:18:52 Join Skydemon [0] (
10:20:59preglowi'd rather not compile my own kernel here
10:21:00preglowbut we'll see
10:22:11 Join Janco [0] (
10:23:04perplexitythe other option is just don't scroll or switch around windows while playing music :) Just switch the monitor off, put the mouse away and enjoy the toons
10:23:23preglowit's a bit heavier than that, i'm afraid
10:23:46preglowif i can't count on being able to even play music without skipping, then what i'm going to do later is not going to work at ll
10:24:10B4gderI don't have those problems on my linux box
10:24:59 Join adamd [0] (
10:25:06perplexityI wrote some software for DJ'ing years ago, and kept the interface nice and simple in text mode... Even on my K6-II 333 with 32 Meg of ram I could still beatmix 2 MP3's with no skips or crackles..
10:25:08 Quit adamd (Client Quit)
10:25:56B4gdergo to and make sure your IP info is right! ;-)
10:26:02 Join muesli__ [0] (n=muesli_t@
10:26:19muesli__B4gder sorry, i couldnt resist ;)
10:26:22muesli__hi @ll
10:26:38preglowi'm going to do a ton of realtime convolution
10:26:39muesli__i meant obviously
10:26:58preglow2 mp3s decoding are peanuts compared to it, so i'm not having high hopes
10:27:06preglowanyway, worst come to worst i'll just make my own kernel
10:27:11perplexityyou need the rt patches then preglow.. seriously.. these guys are getting good 5ms latency
10:27:12B4gderI hate forums
10:31:44 Quit amiconn (Read error: 110 (Connection timed out))
10:31:44 Nick amiconn_ is now known as amiconn (
10:35:46preglow2.6 really is one damned slow beast by default
10:36:27preglow50ms scheduling latency has been seen
10:37:06zewith ingo's patches + realtime-lsm (on 2.6.13), when recording to shm (as opposed to HD), i can do realtime convolution with 2.67ms latency @ 24bit/96khz without xruns on my 1ghz athlon system
10:37:48zeif i'm not recording it then shm vs disk doesn't matter obviously, but if i try to write to disk while doing it then i get tons of xruns heh
10:38:43preglowrealtime-lsm also seems to be missing here, yes
10:39:07zei think it might be obsolete now but i haven't looked into it
10:39:30zeyour distro might have a package for the realtime-lsm module
10:39:34 Join DreamTactix291 [0] (
10:39:40zegentoo does at least
10:40:18 Quit Janco ("CGI:IRC (EOF)")
10:40:20zeof course the length of your impulse will effect it also, but i've used some fairly longish impulses
10:41:05preglowit does have a package for it, yes
10:41:18preglownot too long impulses, but very many channels
10:41:35preglowplus i'll need a couple of iir filters and a matrix mixer as well
10:41:39preglowwill be fun to see if it works well
10:43:05preglowof course, if it doesn't, i have to optimise...
10:43:15preglowwhich sounds strangely familiar in a way
10:45:42zewell it all sounds pretty pertinent to the usual goings ons in #lad (linux audio developers), though i dunno how much benefit you may derive from the channel... just to let you know of its existance, if you're not already aware of it
10:49:55ashridahsounds like it'd be cheaper to just buy some hardware to do the job
10:50:20ashridah(in time V cost terms anyway)
10:52:16preglowyou don't get too much hardware that does ambisonics
10:57:39 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
10:57:44kernelsenseihi there
10:58:03*ashridah nods
10:58:35kernelsenseiI want to patch the H300.hex, but I can't find the tools on the wiki, where can I find them ?
10:59:09linuxstb_Which operating system are you using?
10:59:18 Nick linuxstb_ is now known as linuxstb (
10:59:25kernelsenseiI need descramble and mkboot
10:59:30kernelsenseiaccording to the doc
10:59:36B4gderthey're only available as source
10:59:42linuxstbYou need to download the Rockbox source code, and you will find those in the tools/ directory.
11:00:44darklessany precompiled linux binaries available? It's time consuming to make a complete cross-dev toolchain
11:01:06B4gderI don't think anyone has made any available
11:01:22B4gderits even more time consuming to make packages ;-)
11:01:24kernelsenseiwhen you said "sources" you mean the cvs, or is there a tarball ?
11:01:30B4gderthere's both
11:01:53B4gderkernelsensei: the daily build page has daily source tarballs
11:01:56kernelsenseioh yes, the daily
11:04:02 Quit akaidiot (Read error: 110 (Connection timed out))
11:04:34darklessB4gder: in gentoo, it's as easy as "quickpkg cross-devel/gcc" and then upload the resulting file ;)">
11:04:56kernelsenseigentoo roxes ;)
11:04:58perplexityjust the three executables for firmware munging
11:05:14B4gderouf of all people that get problems on linux with cross compiling
11:05:22B4gder99.5% uses gentoo
11:05:32B4gderso please go ahead
11:05:50darklessdoes that say something about gentoo or about its users? :p
11:05:57darkless_average_ users
11:06:05B4gderboth I'd say
11:06:27darklessdamn... now I have to make a cross-dev toolchain just to prove you wrong
11:06:28*amiconn will probably look into the sim issue on amd64
11:06:53Paul_The_NerdThere are sim problems on AMD64?
11:07:02amiconnI'm awaiting my amd64 box any day.
11:07:08 Quit Vlad0man (Read error: 110 (Connection timed out))
11:07:16amiconnFirst challenge will be the linux installation...
11:07:24B4gderPaul_The_Nerd: it seems so
11:07:40preglowamd64 was very well-behaved for me
11:07:42 Join Vlad0man [0] (
11:07:43B4gderit is claimed to crash when built 64bit
11:07:47preglowit does
11:07:49preglowi remember trying it
11:07:58preglowwouldn't even compile until i fixed a couple of things
11:08:30B4gderit compiles now
11:11:04preglowwhich header is size_t ordinarily in? i thought it was a builtin type
11:12:37 Join vger_ [0] (
11:13:04B4gder<stddef.h> perhaps
11:13:09linuxstbSo what types are different on AMD64? Are integers and pointers 64-bit? (/me hasn't ventured beyond the IA32 world)
11:13:19B4gderlongs and pointers are
11:13:25 Quit thegeek (Read error: 113 (No route to host))
11:14:11*B4gder has curl working 64bit on ia64,sparc,HPPA,MIPS,alpha,AMD64 and more
11:14:50ZagorB4gder is da porting ninja
11:16:30B4gderas a curiousity, time_t is not 64bit on all of them ;-)
11:16:57B4gderthat's an OS choice of course
11:18:17preglowcstddef it is
11:18:21 Quit CrookedRain ()
11:21:28*preglow does the c++ dance
11:21:41*B4gder looks scared
11:23:23amiconnpreglow: I tried ubuntu on VMware. Wasn't able to make it work correctly...
11:23:31amiconn(unlike debian testing)
11:24:21preglowjust did itself here
11:24:25ashridahsurprising really. most accounts have ubuntu being fairly automagic. it just really irritates me as an experienced unix sysadmin in general
11:24:27preglowhas always done
11:24:47preglowinstalls as easily as windows
11:24:47amiconnIt installed fine, but I had 2 problems
11:25:19amiconn(1) I wasn't able to build vmware tools, b/c the kernel was built using gcc 3.4.5 vs. the installed gcc is 4.0.3
11:25:44amiconn(2) I wasn't able to make rhythmbox work although sound itself worked fine
11:25:56ashridahamiconn: hrm, that's usually just a matter of telling the vmware script to use gcc-3.4 to build with on debianalikes (assuming that's even handy)
11:26:15ashridahah. can't say i've used rhythmbox tho, i tend to be a kde user
11:26:19ashridah(hence, amarok instead)
11:26:25amiconnashridah: debian testing kernels are built with 4.0.x, so no problem there
11:26:38mirak_amiconn: you want to install a 64bit distibution ?
11:26:41*amiconn doesn't like kde at all
11:26:47*mirak_ as well
11:26:49ashridahamiconn: yeah, but i'm surprised ubuntu hasn't installed both gcc 4 and 3.4 if it compiled the kernel with 3
11:26:51amiconnmirak_: yes
11:26:52preglowrhythmbox is shit anyway
11:26:55*mirak_ not much in fact
11:26:58ashridahbut i haven't looked recently
11:27:00mirak_preglow: deep shit
11:27:16preglowi bloody prefer xmms to rhythmbox, and that's saying something
11:27:17mirak_preglow: try quodlibet it's a fine equivalent, with a library etcetera
11:27:23amiconnpreglow: It's working fine on debian testing
11:27:31mirak_preglow: quod-libet
11:27:54mirak_amiconn: prepare a 32bit backup install, or even a dualboot
11:27:54preglowwhy must everyone name their programs after what toolkit they use?
11:28:39amiconnmirak_: Why?
11:29:01*amiconn doesn't like dualboot
11:29:05mirak_preglow: because different toolkit apps looks like shit often on the other. Well not really like shit, but well that's not as good looking
11:29:13kernelsenseiOMG !!
11:29:38mirak_amiconn: because you won't have flash and some binary stuff
11:29:43kernelsenseirockbox boots a lot faster than the official firmware !
11:29:44mirak_amiconn: their is instabilities
11:29:51*amiconn doesn't need flash
11:29:53mirak_amiconn: well, I gave up personnaly
11:29:56 Join Moos [0] (
11:30:22*amiconn considers getting vmware ws for linux as well
11:30:26mirak_amiconn: but if you like to post bug report go ahead :)
11:30:45ashridahkernelsensei: yes. it doesn't go on with the 'lets scan the entire disk even tho you probably care about <2% of it this time i'm running' stuff
11:30:46mirak_amiconn: there is wmware for 64 bit ?
11:31:31pregloweven when scanning the entire disk, it's tons faster
11:31:40amiconnvmware ws 5.5.1 even supports 32 and 64 bit guests on 64 bit hosts
11:32:14mirak_amiconn: also there is no performance boost really noticeable
11:32:15kernelsenseilets read the user's manual ... rockbox is great !! @devels : thanks ;)
11:34:47 Join webguest35 [0] (
11:36:03 Quit webguest35 (Client Quit)
11:40:17preglowthe manual is somewhat outdated
11:44:47 Join Vladoman [0] (
11:46:37linuxstbDoes anyone else have audio problems in the SDL sim? For me, it always seems to stop playback after 1 or 2 seconds.
11:46:53Paul_The_NerdI still haven't gotten it to build for me in Cygwin
11:47:39 Quit Vlad0man (Read error: 110 (Connection timed out))
11:48:23mirak_linuxstb: me too
11:55:02***Saving seen data "./dancer.seen"
11:55:03ashridahPaul_The_Nerd: if you give me a few, i'll give it a shot
11:55:03 Quit Strath (Read error: 104 (Connection reset by peer))
11:55:17ashridahactually pretty sure i've got cygwin around now
11:55:33 Join Strath [0] (
11:56:21 Quit vger_ ("User abortion with 5 coathooks")
12:00:55 Quit DreamTactix291 (Read error: 110 (Connection timed out))
12:01:09Paul_The_NerdI have to go now, seeing as it's about 5AM and I really should get a little sleep. But yeah, I installed SDL 1.2.9 using the tarball from the official site and followed their instructions for using it in Cygwin, and when compiling the sim it runs into a bunch of errors and warnings when it can't find pthread.h.
12:01:11 Part Paul_The_Nerd
12:02:26ashridahhmm. my guess is it's configured itself to build as a non-cygwin library
12:15:34ashridahyou know? cygwin in vmware is kinda slow :)
12:22:50preglowno shit
12:33:41 Join mikearthur [0] (
12:40:11 Quit Paprica (Read error: 110 (Connection timed out))
12:40:55 Join petur [0] (
12:44:12ashridahwhoever worked on the SDL-sim port needs to replace the simulator's use of pthreads' with SDL's wrapper implementation
12:44:22preglowoh yes
12:46:54 Join akaidiot [0] (
12:47:49LinusNpetur: please use the -u switch when creating patches
12:47:54preglowoooh, i've got a heisenbug!
12:48:10peturwhich patch?
12:48:21LinusNpetur: the gain patch
12:48:49peturI resubmitted that one - I'm quite sure I used -u
12:49:06 Join adamd [0] (
12:49:21amiconnpreglow: ?
12:49:25LinusNpetur: i downloaded it 3 minutes ago
12:49:41preglowamiconn: you know, one of those bugs that vanish when you try to debug it
12:50:45peturI don't understand it anymore. I did this: cvs diff . -u -r > patch
12:51:19peturcan't double-check - am at work now :(
12:51:42peturshall I make it again tonight?
12:52:54LinusNpetur: -r?
12:53:13LinusNcvs diff is always recursive
12:53:21peturis it? ok
12:53:32LinusNand the "." isn't needed either
12:53:36LinusNcvs diff -u
12:53:49peturthanks... always happy to learn something
12:54:10B4gderput the -u in your .cvsrc file and you won't have to do that either ;-)
12:54:27B4gder$ cat ~/.cvsrc
12:54:27B4gdercvs -z3
12:54:27B4gderupdate -dP
12:54:27DBUGEnqueued KICK B4gder
12:54:27B4gderdiff -u
12:54:49peturI had troubles in cygwin so I came up with that syntax after some trial-and-error...
12:55:14LinusNthe -r switch to cvs diff is not for recursion, it's for selecting a specific version to compare against, -r <tag>
12:55:32peturIf I did 'cvs diff -u somedir' it didn't work
12:55:46peturI found the -r switch in the RB wiki :(
12:56:01LinusNpetur: that was for "diff" not for "cvs diff"
12:56:04 Quit mirak_ (Remote closed the connection)
12:56:12peturwoops again
12:57:08ashridahdamnit. you can't use errno in a threaded app.
12:57:12CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
12:57:12*ashridah smacks forehead
12:57:23peturI see you're correcting my wiki change
12:57:43linuxstbWhilst we're in the Unix classroom, does anyone know an alternative way to browse info files than "info" ?
12:58:15B4gderoh how I don't like info files
12:58:49linuxstbI don't like "info"
12:59:03linuxstbBut I don't hold it against the files themselves...
12:59:17B4gderI prefer a good old man page
12:59:57linuxstbYes, but very large man pages are hard to navigate.
13:00:13B4gderso are info files
13:00:38linuxstbThat's why I was asking if there is a better browser than "info"
13:00:49B4gdernot to my knowledge
13:00:55peturLinusN: I was thinking of adding the combined L/R gain change also for the other targets... are you applying my patch or rejecting?
13:01:01 Join webguest74 [0] (
13:01:19webguest74linuxstb: pinfo
13:01:33 Nick webguest74 is now known as nnick (
13:01:41LinusNpetur: neither
13:02:12adamdis it me or does changing channels to mono not actually output in mono
13:02:21peturshould I redo my homework then ;)
13:02:23LinusNpetur: remember that the archos only has 8 lines of text including the status bar
13:02:43adamdthat'd be very useful for two people using the same earphones... borrowing an earpiece each
13:03:14preglowinfo files would not be hard to navigate given a decent browser
13:03:23ashridahhrm. slight problem. SDL expects threads to return ints, not whatever the hell pthreads is expecting a thread's function to return
13:03:24preglowinfo files themselves are fine when you want a reference manual
13:03:40preglowthinking they should replace man files is lunacy
13:04:22peturLinusN: you win one line: before = gain / gain L / gain R. after = gain L / gain R
13:04:27ashridahhmm. iriver's scramble and descramble seem to fail to build in cygwin
13:04:42LinusNpetur: how do you solve it?
13:05:12peturif you move down to gain, it first selects both lines, then L, then R
13:05:41LinusNso both lines are highlighted?
13:05:53peturagainst the rules?
13:05:55adamdis this a known bug, or does setting "Channel configuration" to Mono actually do something different than what I understand it should do [... output in Mono] ?
13:06:27LinusNadamd: which platform?
13:07:11LinusNi don't think those settings are active on the iriver platform
13:07:37adamdany plans to make them active?... would be very handy
13:07:42adamdor is it even possible?
13:07:47LinusNyes, we have plans for that
13:07:52adamdcool :)
13:08:31 Quit nnick ("CGI:IRC")
13:12:06ashridahhm. can i do #ifdef WIN32 && !SDL ?
13:12:51amiconn#if defined(WIN32) && !defined(SDL)
13:13:48ashridahthe win32 SDL sim has a couple of places where it doesn't make the distinction between win32 and SDL
13:14:01ashridahnamely, uisimulator/common/lcd-common.c
13:14:16ashridahthreads.c is fairly simple modification
13:14:31ashridahthread.c rather
13:15:03 Join Febs [0] (
13:17:18 Join baobab68 [0] (
13:17:32baobab68hi all
13:19:23ashridaham i so out of my depth :)
13:20:00ashridahIn function 'Lang_load': /rockbox/apps/language.c:80: undefined reference to '_debugf'
13:20:05ashridahlang_load rather
13:20:55ashridahi can't find that anywhere in the source tho
13:23:23 Join aliask [0] (
13:25:03 Part petur
13:26:03 Part baobab68
13:27:38 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
13:32:15linuxstbashridah: debugf is defined (for the sim) in uisimulator/common/io.c - it is wrapped by the DEBUGF macro
13:32:45ashridahstrange that it isn't being found when linking rockboxui.exe or whatever it is
13:33:04 Join eli_sherer [0] (
13:33:44ashridahout of my depth tho, just going to post my thread.c modifications to the tracker for now (got a few warnings in it i'm not sure how to solve without being potentially architecture dependent
13:34:55linuxstbYou're using cygwin?
13:35:22ashridahtrying to get the SDL sim working for Paul_The_Nerd
13:35:29ashridahthe sdl sim was still using pthreads
13:35:41ashridahthat's an easy fix, just replaced it with the SDL threads wrapper functions
13:36:24ashridah(there's no pthread implementation in cygwin itself, it's an extra download, so we may as well just use SDL's wrapper for sanity's sake)
13:40:44ashridahdamnit. can't remember my sourceforge password, or what email address i used for it
13:40:59 Join DreamTactix291 [0] (
13:41:59 Join DrMoos [0] (
13:42:31 Quit Moos (Read error: 104 (Connection reset by peer))
13:42:34 Nick DrMoos is now known as Moos (
13:45:49 Quit Moos (Read error: 104 (Connection reset by peer))
13:45:52 Join Moos [0] (
13:48:17ashridahlinuxstb: feel like looking at ?
13:51:39linuxstbashridah: Looks fine to me (but I don't know much about threading). I'm just compiling it under Linux now.
13:52:26ashridahyeah. there's two issues i'm not sure about, one SDL'ism, (SDL_GetThreadID is basically effectively giving me a massive negative number, not sure if it's a casting bug or otherwise
13:52:47linuxstbDoesn't that just mean it is unsigned?
13:53:13ashridahah, yeah, it probably is :)
13:53:15linuxstbI see it's giving some warnings.
13:53:18ashridahit's defined as Uint32
13:53:39ashridahyeah, those are the differences between the function prototype for the functions pthreads accepts and the functions sdl accepts
13:53:52ashridahpthreads returns void* (i think) and sdl threads return int's
13:54:05linuxstbAccording to the man page, the SDL prototype is: SDL_Thread *SDL_CreateThread(int (*fn)(void *), void *data);
13:55:03***Saving seen data "./dancer.seen"
13:55:23XavierGranyone on gygwin right now?
13:55:33ashridahlinuxstb: yeah
13:55:54ashridahlinuxstb: and the problem is, int could potentially be different from void *
13:56:36 Join tvelocity [0] (n=tony@
13:57:07B4gderbut should we use the SDL way unconditionally?
13:57:12XavierGrHi tvelocity!
13:57:59preglowyes, we should
13:58:14linuxstbashridah: I think you should change the definitions of the functions to match SDL's type for CreateThread - i.e. "int myfunc(void* p);"
13:58:38linuxstbAnd then you can remove the messy cast from the function call.
14:04:09 Quit akaidiot (Read error: 110 (Connection timed out))
14:04:39ashridahhrm, are those functions defined in the simulator?
14:05:13ashridahrofl, they are too
14:06:17Bgerstupid q, but what's the quickest way to find where a string is situated in a mysql db with 20~30 tables ?
14:07:31Nibblerwrite a script that gets all the tables+varchar stuff
14:07:35linuxstbDump the database to a text file and use "grep"?
14:07:53*linuxstb hates databases
14:07:55Bgerphpmyadmin ...
14:10:51ashridahlinuxstb: hrm. that's not really going to be possible, thread.c provides "create_thread"
14:12:18 Join Membrillo [0] (n=sam_kill@
14:12:41ashridahwhich, annoyingly, hrm, except that's not exported.
14:13:55linuxstbashridah: I've done a little work, I'll show you my patch...
14:14:19ashridahheh, i was poking at it a bit myself. not sure if my modifications are really completely safe tho
14:14:32ashridah(since i can't remember the keybindings :) )
14:15:13preglowBger: select id from tablename where stringmembername = 'stringvalue' ?
14:15:26linuxstb(this is a patch against CVS)
14:15:38linuxstbapply it in the uisimulator/sdl directory
14:16:44Bgerpreglow all tables at once ? :)
14:16:46ashridahaah, i see, it was bitching about the argument to the function, not what the function returns.
14:16:58ashridahBger: there's not really anything that's going to get you that in a neat fashion
14:17:10ashridahnot with mysql's limitations, anyway
14:17:28Bgerah :( ok, 10q
14:17:43preglowhaha, all tables at once i have no idea on how to do
14:17:46ashridahyou'll be able to do each table individually, of ocurse
14:17:48preglowi never had to search a whole databsae at once
14:18:03linuxstbjust dump it to a text file...
14:18:30ashridahlinuxstb: which is horribly inefficient
14:18:37linuxstbI mean forever...
14:18:46preglowPlease wait for search result: dumping database to text file...
14:19:03ashridahlinuxstb: no, even if it's there permanently
14:19:13preglowsearching a text file = not fast
14:19:18Bgerlinuxstb no access to the dbfiles ...
14:19:22linuxstbUnless you have a lot of data, I would argue that text files are more efficient than running a DB server
14:19:26preglowi'm not very fond of databases myself, but they do work when you have tons of info
14:19:29ashridahthe correct answer is "dump it to a text file *then import it into postgres and use PL/SQL, just for fun*"
14:19:54ashridahBger: you don't need access to the files themselves to run mysqldump :)
14:19:58preglowlinuxstb: you wouldn't need a very big text file before searching became wildly inefficient with text files as opposed to a proper db
14:20:08ashridahjust select access for whatever database you're using
14:20:10preglowand believe me, i've tried
14:20:11 Join Jungti1234 [0] (
14:20:23*ashridah concurrs
14:20:29ashridahtext indexing is HORRIBLE
14:20:48Jungti1234hi ho
14:20:52 Join Polo_o [0] (
14:20:57ashridaha properly indexed database will be a zillion times more efficient if you ever need to run the same style of search multiple times
14:21:10 Quit kernelsensei (Remote closed the connection)
14:21:22ashridahthe annoying part is getting decent text matching in SQL :)
14:22:02 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
14:22:03ashridahlinuxstb: that patch looks okay to me. solves part of the SDL sim issue (at least, it compiles that file). it seems to run under X11
14:22:14linuxstbOf course - it all depends on the amount of data. Bger?
14:22:17ashridahjust no idea how to get the menu up to test plugins or whatever
14:22:41linuxstbJust copy a text file into the archos/ directory and play it.
14:22:51linuxstbOr check button.c for how the buttons are mapped
14:23:13Bgeroh, never mind :)
14:24:39ashridahseems to be okay.
14:24:54linuxstbDo you have commit access, or shall I commit it?
14:25:18preglowsure, it depends on the amount of data, but that amount of data doesn't need to be big at all
14:25:20ashridahi don't, which is why i was looking for my sourceforge login
14:25:22ashridah(which i cna't get)
14:25:57Rob2222_What was the difference of the H300 old and new display? The viewing angle?
14:26:52preglowthe way angle at which it's meant to be viewed, i think
14:27:28Rob2222_horizontal and vertical?
14:28:06Rob2222_I had 2 H300 and I saw a difference at the displays, but I didnt knew, what it was ...
14:28:13preglowafaik, the original one is a taller than it is wide, so iriver just turned it 90 degrees
14:28:19Rob2222_Well, 1 unit was selled.
14:29:12Rob2222_So which display is that, that has different colors when viewing from up and down?
14:29:36 Join YouCeyE [0] (
14:30:04Jungti1234I have vertical LCD
14:30:06Rob2222_When viewing from up, its almost blue, when viewing from down, its almost white.
14:30:29Jungti1234vertical LCD
14:30:30Jungti1234vertical LCD
14:31:15Jungti1234It's old LCD
14:31:33Jungti1234horizontal LCD
14:31:36Jungti1234It's new LCD
14:31:40linuxstbashridah: Is this your first patch? You don't appear in the CREDITS file.
14:32:17 Quit Jungti1234 ("CGI:IRC")
14:32:27Rob2222_so the difference is the color changing and the viewing angle?
14:32:42ashridahlinuxstb: yeah, it is
14:33:11 Quit kernelsensei (Remote closed the connection)
14:34:02 Join kernelsensei [0] (n=boris@unaffiliated/kernelsensei)
14:34:12adamdthose images don't load :-\
14:34:47ashridahthey load for me.
14:34:57adamdhm strange
14:35:14adamd"The page you are looking for might have been removed, had its name changed, or is temporarily unavailable."
14:35:24 Join webguest83 [0] (
14:35:26perplexityworks for me
14:36:06adamdvery strange
14:36:42adamdit works now
14:37:08adamdhm, weird. wonder why they're using a different screen
14:37:38ashridahadamd: manufacturer of the first one probably shafted them, or there were unexpected high defect rates or something
14:38:03ashridahhell, maybe they were just 5c cheaper per unit
14:38:24adamdhaha, yeah
14:38:35adamdi've got the new one anyway... bought mine at xma
14:38:44linuxstbashridah: patch committed.
14:38:56 Quit tvelocity ("Leaving")
14:39:03ashridahyay ^.^ n stuff.
14:39:07linuxstbI removed all references to pthread in thread.c as well, just to finish cleaning it.
14:39:31ashridahaah, yeah, there migth have been a few bits i missed, like that block of three commented lines that were in the middle of no-where
14:39:39linuxstbBut cygwin still gives you that debugf problem?
14:39:54ashridahlast i checked. which is strange, i don't get it in linux
14:40:00ashridahi'll poke more tomorrow i guess.
14:40:02*linuxstb should try compiling the SDL sim on Mac OS X
14:40:14ashridah(although it won't be for a while, i need sleep, then i'm going birthday shopping
14:40:17 Quit YouCeyE ("Leaving")
14:40:31ashridahlinuxstb: i actually think it's leftover win32ism's
14:40:42ashridahthere's some in uisimulator/common/lcd-common.c for instance
14:41:04ashridahwhere it just has #ifdef WIN32 #include <lcd-win32.h> which isn't in the include path for the SDL sim
14:41:16ashridah(just including a !SDL deals with that perfectly fine
14:41:31ashridahalthough i'm wondering if that's why i'm getting DEBUGF issues
14:41:36*eli_sherer is announcing that his gif viewer has improved a little...
14:41:47ashridahpossibly we need an lcd-sdl.h with generic stuff
14:42:22ashridahanyway, it's 20 to 1 in the morning. i need sleep :(
14:42:23linuxstbThere are some #ifdef WIN32 in the firmware/ directory
14:42:42 Quit ashridah ("sleep.")
14:43:14linuxstbI see the problem - look for WIN32 in firmware/SOURCES
14:46:08aliaskHey wow it's 20 to 1 in the morning here too! Same timezone as someone else!
14:46:25preglowyou're both aussies
14:46:26preglowso go figure
14:46:37aliaskAh, I never realised before...
14:46:38XavierGrQuestion: I have this loop: for(i = 0; i <= last; i++). Variable "last" is 0. Am I right that the loop will run 1 time and then "i" will be 2?
14:47:06XavierGrI mean i =1
14:47:22linuxstbThen yes - the loop will run once, and i will be 1.
14:47:26aliaskThen that's the way I see.
14:47:29XavierGryeah my bad. So in the end of the loop i will be incremented...
14:47:47Bgerbut you shouldn't rely on this ...
14:47:58aliaskWhy not just execute the code once, not in a for loop?
14:49:18 Join eagle730 [0] (
14:49:32 Quit eagle730 (Client Quit)
14:49:59XavierGrbecause the loop is variable
14:50:19BgerXavierGr what do you want to do ?
14:50:39XavierGrWell it is for the battery_bench plugin.
14:51:15aliaskOh yeah sorry, I was thinking for(i=0;i<=0;i++)
14:51:46XavierGrIf there is no memory log the loop must run once. But if between disk accesses there is battery info I have to run the loop more times
14:56:37 Join tvelocity [0] (n=tony@
15:01:28 Join akaidiot [0] (
15:02:45 Quit linuxstb ("Leaving")
15:04:23adamdthe radio tuner in rockbox has so much better reception than the iRivier firmware's radio tuner
15:04:41Membrilloadamd: id say thats placebo
15:05:01Membrilloits hardware based meaning realistically it shouldnt make a difference
15:05:09adamdbut I can't normally get reception in the room with all the printers in
15:05:27 Join linuxstb [0] (
15:05:37adamdwhen I swap back to iriver it's all fuzzy... unless there's.. something else going on in the iriver firmware
15:05:38Membrillodoesnt make much sense to me. But if thats true, nice
15:05:40adamdor something
15:05:51adamdheh, no it doesn't make sense, if it's hardware as you say
15:06:02Membrillowell, i dont know anything about it
15:06:05adamdlet me swap back and see if it's just environmental changes maybe
15:06:08Membrilloim just speculating
15:06:08aliaskCould be interference from other hardware which we are not using.
15:06:08linuxstbWhich iriver?
15:06:20Membrilloi know there is a physical tuner chip which manages it
15:06:35adamdremember 98.9MHz
15:06:43aliaskI think if you set the CPU at certain frequencies with the radio on you can get interference I think.
15:06:48jelle-kis the cpu in the iriver firmware running on 11 mhz too?
15:06:58jelle-kwhen in FM tuner mode
15:06:58aliaskWho knows.
15:07:27adamdI'm getting more fuzz with the iRiver firmware
15:07:37eli_sherermy gif viewer only works on the sim :(
15:07:39Membrillosounds good then
15:07:41adamdnot as bad as earlier but there's a difference
15:07:43eli_shererand i dont use memory allocation anymore
15:08:17aliaskWhat are the differences between sim and player?
15:08:22eli_shereri guess it's something to do with endian stuff
15:08:36adamdthat's why I wanted the more exact tuning, because at work I always get fuzz
15:08:39XavierGraliask: many
15:09:26aliaskI thought the sim emulated the hardware, or am I hallucinating?
15:09:44aliaskGah whatever, I'm going to bed, too tired to think straight.
15:09:44XavierGrit is a simulator
15:09:49XavierGrnot an emulator
15:09:50aliaskG'night everyone.
15:10:01 Quit aliask ("Chatzilla [Firefox 1.5/2005111116]")
15:12:48eli_shereri'm trying to do a endian change in my prog to see if it's the problem
15:13:42linuxstbWhat happens when you try and use it on the target?
15:15:05eli_shererit shows only a background color like some people said it happens...
15:15:37XavierGrticks / HZ = seconds right?
15:15:40eli_shereron the sim it draws a fuzzy picture that resamble the original picture (ill fix in in time)
15:16:00markunXavierGr: yes
15:16:09linuxstbWell, GIF files are little-endian, so you will have to take care of that in your code.
15:16:44eli_shereri know that they are LE but i use the readshort and readlong functions
15:16:44preglowprobably just the header where it matters, yeah?
15:16:49XavierGreli_sherer: You are porting animated Gifs?
15:16:59preglowplease, god, no!
15:17:18eli_shererXavierGr: in time it will be animated (it's just more pictures after the first one)
15:17:37eli_shererand the animated header, i read anyway
15:17:41XavierGrwow, what about grayscale?
15:17:42linuxstbeli_sherer: Another difference between the sim and targets is stack size (I think).
15:17:50XavierGrfor the H100
15:17:54eli_shereri use my own stack
15:17:59eli_shererLZW decoding
15:18:16preglowyour own stack?
15:18:30eli_shereri wish there was some how i could have a DEBUGF but runtime on the target...
15:18:42B4gderlogf is your friend
15:18:46eli_shereroreglow: meaning an array...
15:18:59pregloweli_sherer: that's not a stack
15:19:08eli_shereri use it as a stack...
15:19:25eli_shererthe same i used it with xobox...
15:19:42eli_shereran array with a top pointer...
15:19:42linuxstbDo you use plugin_get_buffer() ?
15:20:16eli_shereri did before the memory i just allocate in advance by pixelstack[MAXSTACKSIZE]
15:20:31eli_shererinstead of *pixelstack
15:20:47linuxstbIs your source available to download?
15:21:13eli_sherernot yet...but i can upload it if you want...i really want some break from it
15:21:13 Quit Membrillo ()
15:21:30linuxstbI don't mind having a quick look
15:22:15eli_shererok wait a minute
15:23:34eli_shereryou need to add it to sources... and copy it to viewers and add to viewers.config file
15:23:38 Join stamppot [0] (
15:23:48linuxstbThat's OK, I probably won't compile it.
15:23:49stamppotOK, here's a thought...
15:24:33stamppotDoes an iPod sync the playcounts for the different files over to iTunes, to determine track popularity?
15:25:14eli_shereri just need to know the difference between the sim and the target that make it don't work on the target the same as on the sim
15:25:40stamppotI'm using Amarok under Linux as my Audio player, and it would be nice if I could sync the "popularity" of my tracks over to Amarok.
15:26:13B4gdereli_sherer: why shouldn't they work the same?
15:26:24B4gderthe whole point of the sim is to have the same code
15:26:29stamppotThat way, I can have a unified view of the most popular songs.
15:27:07markunlinuxstb: how big is the rockbox bootloader for the ipods?
15:27:10stamppotDoes anyone know if it works like that in iTunes? Or doesn't the iPod have a notion of popularity?
15:27:17linuxstbstamppot: Yes, I think the ipod does that, but I don't use itunes unless I have to.
15:27:35linuxstbWhen listening to a song on the ipod, you can rate it.
15:27:54linuxstbBut I don't know if the info is sent back to itunes - I would imagine that it would be.
15:28:01preglowmarkun: about 60k
15:28:10markunok, thanks
15:28:51linuxstbThat's with a couple of logos in, and I'm sure it has some dead code that could be stripped out if size was a problem.
15:28:56stamppotLinuxstb: yeah, but that is something different. Rating is a user-set property. Popularity would be determined by the number of times you play it (playcount in Rockbox parlango)
15:28:59 Join NicoFR [0] (
15:29:25linuxstbIt knows about "recently played" tracks - not sure about popularity.
15:29:36*preglow votes for removing the ipl logo
15:29:48preglowand rockbox logo, for that sake, it's a bootloader :>
15:30:20linuxstbBut the rockbox logo is so pretty....
15:30:32preglowlinuxstb: do you have any idea how retailos manages to boot straight like it does after a long sleep?
15:30:32stamppotPlaycount is nice, so that you can generate an on the fly playlist in Amarok containing unpopular or "forgotten" songs and send them to Rockbox
15:30:40preglowyou see the apple logo, so the bootloader is run
15:30:45preglowbut then our bootloader is skipped completely...
15:30:50linuxstbIs it?
15:31:01linuxstbI never noticed that.
15:31:39preglowbut yeah, rockbox logo can stay if it must, but not much point in having the ipl logo anymore
15:31:53preglowcredits are there anyway
15:32:13markunOn the gigabeat we can also put our bootloader where the background images are now
15:32:23stamppotI find my self using my H300 at work, and using Amarok from home (working from the same Rsynced collection). It would be nice if I could keep them unified a bit more...
15:33:34linuxstbmarkun: So you are able to modify the background images?
15:33:42stamppotHow experimental is the playcount stuff still? It is still marked as experimental in the menus...
15:34:03preglowvery experimental
15:34:04 Join DrMoos [0] (
15:34:06preglowi doubt it works
15:34:11preglowit's been pretty much abandoned
15:34:18stamppotI might dust off my C/C++ skills and try to write a module for Amarok that accomplishes this...
15:34:41stamppotAi.... Any plans on re-instating such a functionality?
15:35:33stamppotpreglow: what is the state of the tagdatabase?
15:35:34markunlinuxstb: I could. They are stored as 16 bit bitmaps. Probably RGB 5:5:5:1, as that is the mode the firmware sets for the LCD controller.
15:35:56B4gderstamppot: we await slasheri's tagcache
15:36:20stamppotB4gder: eagerly awaiting, I presume?
15:36:40B4gderwell, some might say so sure
15:36:53markunlinuxstb: still didn't write to the flash. I'll try to change a few bytes with dd next week.
15:37:11stamppotsorry if this has been asked before, but what is tagcache trying to accomplish beyond tagdatabase?
15:37:14linuxstbmarkun: Good luck.
15:37:40B4gderstamppot: it builds the tagdb entirely on target
15:38:05B4gderand it is a new stab at it
15:38:12B4gderperhaps it'll work better this time
15:38:26B4gderthe problem with the existing code is that no one care for it
15:38:49stamppotWell, I did...
15:39:13B4gderstamppot: then go ahead, build a tagdb install it and test it
15:39:19stamppotI'm using it all the time. It has some bugs, but I always presumed that they would be ironed out over time.
15:39:22B4gderit might work
15:39:45preglowno, as it is, tagdb is pretty much abandoned
15:40:00preglowlooked like something might have happened to it, but nothing came of it
15:40:23preglowso now it's probably a half-baked mess
15:40:32stamppotOK, to avoid confusion: with tagdb I'm referring to the java-based songdb.jar stuff.
15:40:32preglowwhich in turn might explain why noone wants to touch it
15:40:50B4gderthere's a perl tool as well
15:40:51preglowthe database tool being java doesn't help either...
15:40:56stamppotHmmm, it is still working...
15:40:57B4gderthey both build a tagdb
15:40:59linuxstbmarkun: Are there any kernel modules loaded from disk on the gigabeat? Could you patch one of those to start your bootloader?
15:41:19preglowB4gder: does the perl tool do as much as the java one does, currently?
15:41:38stamppotI'm running a rockbox from 3 days ago and I can at least build playlists using the artist/album/genre tags...
15:42:12B4gderpreglow: hcl claims the java version has a better id3 parser though, which might be true I don't know
15:42:14stamppotAgreed, I don't use the search functionality (too awkward to do that with an on-screen keyboard, imho)
15:42:28preglowoh well
15:42:36markunlinuxstb: they are, but I can load my own kernel module with insmod of course.
15:42:40preglowat least tagcache will support _exactly_ what rockbox supports
15:42:43preglowfor obvious reasons
15:42:58stamppotOf course...
15:43:13B4gderyes, and any new host-side tools could use the *same* code for it
15:43:24stamppotDo you know if tagcache will also include playcounts?
15:43:25preglowdidnt' slasheri mention that would be a bit hard?
15:43:58 Quit Moos (Read error: 104 (Connection reset by peer))
15:44:17stamppotIf so, I will hold off of building such functionality for Amarok till tagcache is implemented
15:45:25 Join ^BeN^ [0] (
15:45:32 Nick ^BeN^ is now known as Paprica (
15:46:02eli_shererlinuxstb: have you looked into my code?
15:46:09Slasheristamppot: it will include that soon
15:46:22lostlogicLinusN: The iRiver can operate without a battery on charger voltage alone, I'm not sure if this means that it will also do this when the charger and battery are connected, but it is at least possible.
15:46:39Slasherithat is trivial thing to do once the tagcache is working
15:48:09Slasherimaybe i have a day or so at weekend to work with the tagcache
15:49:02markunSlasheri: what will you do with non unicode tags? Convert them to unicode while building the db?
15:50:03preglowSlasheri: would be really neato
15:50:24lostlogicLinusN: the datasheet would seem to indicate that this is impossible and it should go to SAVE state, but the device powered on when I tested a resoldering job on my H340's power jack last night w/o a battery connected.
15:50:31Slasherimarkun: hmm, possibly. Or i just hope metadata parses does the proper conversion
15:50:35preglowSlasheri: wouldn't happen to have a working patch? i haven't tested it yet
15:51:04Slasheripreglow: at the moment, nope. But maybe in two days :)
15:51:36stamppotSlasheri: OK, that would be neato...
15:51:54markunSlasheri: Better to leave them non unicode. People with different encodings mp3s can then change the default codepage for some files even after the db is built.
15:52:35Slasherihowever, the first version of tagcache will not include runtime db information, but that feature will be added soon
15:52:53preglowthen all looks well, if you ask me
15:52:59preglowhow does it work on archoses?
15:53:14Slasherimarkun: ok, i will not touch the encoding then
15:53:32Slasheribut now time to leave the bus :) ->
15:53:39NicoFRlostlogic: with your A-B repeat patch, AB_REPEAT_ENABLE set to 1, pressing the rec button causes the track to go in pause for a split second
15:53:52NicoFRwhen A-B repeat isn't enabled, obviously
15:54:54NicoFRwould it be possible to make it have absolutly no effect ?
15:55:04***Saving seen data "./dancer.seen"
15:55:31lostlogicNicoFR: yeah, that's a bug, I'll try to look at it today, now that I have an H340 again :)
15:55:31BHSPitLappyso preglow, linuxstb, how's it going
15:55:57NicoFRlostlogic: already a new H340 ? that was quick :)
15:56:11 Quit darkless (Client Quit)
15:57:38preglowall's well, having some quality time with c++ right now
15:58:17linuxstbeli_sherer: Just browsing it now. One thing that looks wrong is that your readshort and readlong routines (for big-endian) are returning signed values, but I think they should be unsigned.
15:59:30linuxstbIf I was you, the first thing I would check would be to make sure those two functions are working correctly in the sim - i.e. always use them, instead of only for big-endian targets
16:01:10linuxstbIn fact, all you need to do is to replace the calls to those functions with le2he32() and le2he16() - they are defined in system.h
16:04:14LinusNlostlogic: will you work on the power management on the h300?
16:04:32 Quit akaidiot (Read error: 110 (Connection timed out))
16:05:55amiconnlinuxstb: letoh32() and letoh16() ...
16:06:41linuxstbamiconn: ?
16:06:49 Quit jelle-k (Remote closed the connection)
16:08:05 Join jelle-k|offline [0] (
16:08:40 Join alphakiller [0] (n=bohas@
16:08:46linuxstbamiconn: Sorry, I misspelt "to"
16:08:46alphakillerhey :P
16:09:02jelle-k|offlinei hate irc bounsers :P
16:09:13alphakilleryesterday I ported a microkernel to iPod :P hehe just in a couple of hours ...
16:09:13linuxstbamiconn: In fact, just ignore me, I need my afternoon coffee.
16:09:24 Nick jelle-k|offline is now known as jelle-k (
16:09:27linuxstbIs it working?
16:09:32alphakilleryeah :D
16:09:36lostlogicLinusN: yeah, I will be this weekend.
16:09:56alphakillerhmm .. how rockbox does memory allocations ?
16:10:15alphakillerdo you give entire pages for the plug-ins ? or they are only allowed to use arrays ?
16:10:46lostlogicNicoFR: yeah, fast shipping of a broken product that I had to solder back together last night when I got it... but at least I'm back in business.
16:11:18XavierGrhey where is the alpine_cdc.c from the source?
16:11:52 Quit Rick (Read error: 104 (Connection reset by peer))
16:13:04 Join Rick [0] (
16:13:49alphakillerlinuxstb: I could code a page generator for plug-ins :D
16:14:18BgerXavierGr ?
16:14:27alphakillerso we could have dynamic memory allocation
16:15:12XavierGrBger: Sorry I thought that it was removed. Dumbass I was looking at the wrong folder!
16:15:41B4gderalphakiller: you would first have to persuade us that we need it
16:16:01markunhere we go again..
16:16:38preglowwhat would we need dynamic memory allocation for ?
16:16:42linuxstbalphakiller: There is a single block of memory (at a fixed location in RAM) where plugins are loaded to. Only one plugin is ever running at one time, and they are compiled to execute from that fixed address.
16:16:50preglowwe need to write a new developer faq or something
16:17:22alphakillerhmm .. I see
16:17:28alphakillerso no need
16:17:53preglowour goal, when it comes to memory management, is to maximise the size of the leftover memory, since that becomes the mp3 buffer
16:18:18preglowif we start using dynamic memory allocation, that will, contrary to what you might first thing, give us less left-over memory
16:18:32alphakillerI got the point :D
16:18:49 Join Kohlrabi [0] (
16:18:52preglowi bet we could write a twenty page treatise on this if everyone pitched in, heh
16:19:00alphakillerI just say my microkernel core on top of the rockbox booting process :P
16:19:11markunalphakiller: will you be working on rockbox or porting your java os to the ipod?
16:19:44alphakillernah, what I've made was just to explore the ARM ... heheh coz I'm gonna run it on another arm based system ...
16:19:52alphakilleri'll work on rockbox
16:20:12alphakillerI'm just saying that I've ported to learn a little bit more about ARM :P
16:20:19linuxstbalphakiller: How much of the hardware is your kernel using? e.g. have you enabled the cache?
16:20:42alphakillerlinuxstb: just the higher level core
16:21:25alphakillerI'm not using cache yet, nor any processor stuff .. I just fixed up some memory distribution stuff ... though, I just saw that ARM, as MIPS, needs memory addresses aligned to 4 bytes
16:21:41alphakillerso I fixed that on my microkernel core ... just improving the real core porting ..
16:21:43alphakillernothing else
16:22:15alphakillerbut yes, I'm very interested on rockbox :D
16:22:21preglowit'll vomit quite copiously if you give it unaligned addresses yes
16:22:21XavierGrdo I need to rb->button_get() to see if the user connected the USB cable?
16:22:38alphakilleryeah, but I didn't know that :D
16:23:19alphakillerso I had to adapt some parts of the core to do that .. and cleaned code ..
16:23:27alphakiller(my microkernel)
16:23:34alphakillerjust part of my "normal job"
16:23:59alphakillerI prefer playing with my own stuff, before working with other's :D
16:24:27alphakillerbut I think I'm able to help ..
16:24:42alphakillerjust need to get more experience now on arm asm ...
16:25:02alphakillerand read a bit more on architecture details
16:26:33linuxstbWhat ARM is your "real" target using?
16:29:15amiconnpreglow: arm needs data aligned, like sh1? Or just instructions (which is common afaik)?
16:29:35preglowlike sh1
16:30:00amiconnAt least this won't be ignored then :)
16:30:16preglowall data accesses need to be aligned to their natural size
16:31:42B4gderI believe that's right
16:32:32linuxstbamiconn: Have you tried the SDL sim in Windows?
16:33:16preglowcode alignment is actually used to indicate if the branch target is arm or thumb code, when using the bx instruction. if the lsb is 1, it's thumb, if 0, arm
16:33:16 Join DangerousDan [0] (
16:33:47preglowthe address isn't used like that for fetching code, of course, it's just an indicator
16:34:07preglowsince arm code uses fixed size instructions, you don't need the lower bits anyway
16:40:18 Join _lars [0] (
16:41:47 Join gtkspert [0] (
16:43:19amiconnlinuxstb: no
16:44:09XavierGrWhat's the SDL sim exactly? What is the difference with the default simulator?
16:44:29B4gderthere is no default
16:44:36linuxstbIt's a new (third) simulator - the intention is to replace the X11 and win32 simulators with the SDL one.
16:45:10XavierGrso we will have a unified one then?
16:45:35linuxstbThat's the idea.
16:45:57linuxstbIt should also compile on Mac OS X - I'm hoping to give it a go this weekend.
16:48:21 Nick alphakiller is now known as alphakiller|away (n=bohas@
16:50:31 Nick _lars is now known as lars|offline (
16:50:35 Nick lars|offline is now known as _lars (
16:50:44jelle-kit works lars :P
16:51:05adamd... sorry
16:51:26adamdno offence meant tto any blind people in here
16:51:54adamd9 minutes til I finish work :-) woooo
16:52:04 Quit B4gder ("time to say moo")
16:52:12XavierGrwhat sort of wok?
16:52:31adamdfaffing around with bills of materials, in the drawing room of an enginnering works
16:52:38adamdit's boring and tedious but easy
16:52:42adamdsuits me :P
16:53:08 Join darkless [0] (
16:53:09XavierGrand it is friday too! ;)
16:53:37 Quit Zagor ("Client exiting")
16:54:23adamdgoing back up to uni on Sunday too
16:56:12adamdtalk to you later
16:56:48 Quit adamd ("CGI:IRC (EOF)")
16:58:17SkydemonLinusN: i readed that you self have a h300 remote - atm i wait that my arives from Korea and now ask myself if i should start to look into how to patch RockBox for the remote or if the implamantation will be in of of the next rockbox dailies? :)
17:04:02BgerSkydemon there are patches for this...
17:04:22 Join vger_ [0] (
17:04:38Bgeri'm working on this, but i don't have time
17:04:43Bgermaybe next week ...
17:04:51Skydemoni know - just never worked with patches - and no i just wonder if i should try to get that runing or if that comes in 2-3 days anyway )
17:04:59Skydemoncool )
17:05:10BgerSkydemon what remote do you have?
17:05:20Bgernon-lcd or with lcd
17:05:33Skydemoniam not sure when the remote will arive - iam damn happy to get one after nearly 6 month serching )
17:05:40Skydemonwith LCD
17:05:43Bgerah :)
17:05:52Bgercongrats :)
17:05:56Skydemonty )
17:06:15Bgerfor now i suggest you to use the "rockbox optimized for h300" on MR
17:06:28 Quit BHSPitLappy (Read error: 110 (Connection timed out))
17:06:32Bger(if you don't want to patch yourself)
17:07:39 Nick _lars is now known as lars|offline (
17:07:39 Nick lars|offline is now known as _lars (
17:08:56Skydemonwow - greate - dident saw that before - thats perfect - Thanx a ton )
17:09:42Bgerfor nothing
17:09:58Bger;) nite ;)
17:11:16NicoFRi'm trying to test the sdl sim, but i don't really see how i'm supposed to install sdl...
17:11:22 Join ripnetuk [0] (
17:11:30NicoFRwhen i download a build, all i have is a dll
17:11:48NicoFR(i'm on win32)
17:12:01NicoFRcould someone give me a hand ?
17:13:36lostlogicNicoFR: I found the problem with the AB_REPEAT patch, will post a new one shortly.
17:13:57NicoFRbut SF is down atm
17:14:29 Quit ripnetuk (Remote closed the connection)
17:14:45linuxstbeli_sherer: I've got your gif viewer running on my ipod (which is little-endian) - it doesn't work properly on there either, at least not with my two test images.
17:15:06linuxstbOne of them is very corrupted, and the other is fine, but the colours are wrong.
17:16:06linuxstbI don't know if it's a co-incidence, but the one that's corrupt is a gif 87a, the one with bad colours is gif 89a
17:18:29NicoFRnoone to help me with sdl ?
17:20:15linuxstbWhat help do you need?
17:20:46NicoFRi just want to get the sdl sim running
17:20:58NicoFRbut i don't seee how to install sdl
17:21:03NicoFRon win32
17:21:16linuxstbDoesn't cygwin have a package for SDL?
17:21:44linuxstbi.e. can you just run the Cygwin setup.exe and install libsdl (and libsdl-dev)
17:22:36NicoFRi'm using th devkit... i suppose i can still use the cygwin setup ?
17:22:57linuxstbYou may be fighting a losing battle with the devkit.
17:22:59eli_shererlinuxstb: have you changed anything so it work on your ipod?
17:23:16NicoFRie trying to get sdl running with the devkit ?
17:23:41linuxstbeli_sherer: I just turned pluggif.c into a standalone plugin - I didn't want to worry about changing the Makefile etc to handle your "extra" directory.
17:23:52linuxstbBut apart from that, I think it's identical.
17:24:23eli_shereri think i know what's the problem...
17:24:33linuxstbYes, I've just done a diff - the code is identical.
17:25:12eli_shereri didn't load with read????() the array cells from the "block" array...that might be it
17:25:35XavierGrBger are you here?
17:27:27eli_shererwhat h stands for in letoh16()...?
17:27:41linuxstbi.e. the endian-ness of the cpu
17:27:56linuxstbSo if the host cpu is little-endian, letoh16() does nothing.
17:28:23eli_sherergif is LE hence the "le" and and H300 is BE hence the "h" for host?
17:28:49linuxstbWe ignore the fact that h300 is BE and always use that macro for every target.
17:32:17 Join sanitarium [0] (
17:32:21sanitariumhey all
17:32:23eli_shererlinuxstb:but when i read a char i don't need that whole Endian thing...
17:32:33sanitariumi'm sure somepeople use linux here
17:32:57sanitariumi am using kubuntu, and the .rockbox folder doesn't show up
17:33:03sanitariumin the iRiver
17:33:06linuxstbeli_sherer: No.
17:33:09gtkspertls -lah
17:33:11sanitariumand i cant extract any daily builds
17:33:26XavierGrchoose to show hidden files
17:37:43sanitariumwill that work
17:38:20lostlogicNicoFR: new ab repeat patch posted.
17:43:05NicoFRlostlogic: here ? :
17:43:51NicoFRok, i just saw it be updated...
17:43:52 Quit Ismo ("Lost terminal")
17:44:33 Join adamd [0] (
17:46:44 Join akaidiot [0] (
17:48:02 Join solexx [0] (
17:48:24linuxstbNicoFR: You could try these instructions:, skipping step 1.
17:48:44lostlogicNicoFR: a warning (I just realized this) the abrepeat patch probably hurts performance somewhat, because it adds like 9 instructions to every elapsed time call from the codec. I'm trying to figure out a way to reduce that to more like 2 instructions.
17:49:04 Quit edx (Read error: 110 (Connection timed out))
17:49:17 Join ismo [0] (
17:49:28NicoFRlostlogic: even when a-b repeat is not being used ?
17:50:01 Quit einhirn ("Miranda IM! Smaller, Faster, Easier.")
17:50:15NicoFRlinuxstb: this will only give me sld.dll, won't it ?
17:50:21lostlogicyeah −− because it still has to make the JSR to the code to _check_ if it's enabled.
17:50:31 Quit webguest83 ("CGI:IRC (Ping timeout)")
17:50:51linuxstbNicoFR: Yes, but you will need that before you can attempt to compile the SDL simulator.
17:51:21NicoFRbut i can download the dll on the sdl website
17:51:33linuxstbYou will also need the include files
17:51:44NicoFRi'll try thay then
17:52:09eli_shererwell the animation is working too now but still not on the h300
17:52:13eli_shereronly sim
17:52:47linuxstbDo the gifs display perfectly in the sim?
17:53:00*linuxstb tries the sim
17:55:06***Saving seen data "./dancer.seen"
17:55:06XavierGrBger or anyone who is interested: I have updated my battery_bench plugin. Testers are welcome. You can find it on the tracker
17:58:12eli_shererlinuxstb: no... :(
17:58:31 Join actionshrimp [0] (
17:59:24 Nick _lars is now known as lars|offline (
18:01:03linuxstbeli_sherer: I'm getting exactly the same corruptions in the simulator (both h300 and ipod) as on my real ipod - which in a way is good news.
18:01:58eli_shereri too have this weird looking images on my sim but on my "real" h300 i see only bg color
18:02:42eli_shererby the way there's a bug it the code (in the function readGraphicControlExt you should delete the last read)
18:02:43 Quit DreamTactix291 (Read error: 110 (Connection timed out))
18:02:59eli_shererthen you could enable the animation...
18:03:42eli_shererwhen i read a lot of bytes at each 2 bytes gets swapped??
18:05:37linuxstbDo you mean the rb->read() function?
18:13:55eli_shereri am so dumb...i forgot lcd_update...
18:13:55linuxstbThen the answer is no - read just copies a block of data
18:16:05 Join DreamTactix291 [0] (
18:16:45linuxstbeli_sherer: You have lcd_update() at the end of loadgif()
18:17:25sanitariumstill cant extract or see the .rockbox folder
18:17:43eli_shereryes but inorder to see animation i need to insert it between frames
18:17:53linuxstbOK :)
18:18:02eli_shererand that's when it's starting to get interesting :)
18:18:10 Quit NicoFR ()
18:19:14linuxstbsanitarium: Just type "unzip -d /mnt/iriver/" (assuming your player is mounted as /mnt/iriver/)
18:21:16sanitariumCit says it extracted it
18:21:20sanitariumbut I dont thinkso
18:21:26sanitariumI will have a look though
18:21:35linuxstbHow are you looking?
18:21:44sanitariumoh, yeah, Is there a safely remove thing for linux?
18:21:48linuxstbI mean do you use the "ls" command, or a file manager?
18:22:19 Join NicoFR [0] (
18:22:20linuxstbsanitarium: "unmount /mnt/iriver/"
18:22:31 Quit Strath (Read error: 104 (Connection reset by peer))
18:22:38eli_shereri now notice that the colors aren't correct...interesting maybe it's my solution
18:22:57linuxstbsanitarium: I meant "umount"
18:23:17sanitariumand, i cant see the .rockbox folder in ls
18:23:35linuxstbYou need the "-a" option to ls
18:23:43linuxstbie. "ls -a /mnt/iriver/"
18:24:18sanitariumyep, its there
18:24:23sanitariumbut, it was always there
18:26:02sanitariumand H300 doesn;t automatically mount in bootloader USB
18:26:16sanitariumwhere, H300 default FW, it always did
18:26:18 Join Mmmm [0] (
18:26:38linuxstbYou mean the iriver firmware, or when Rockbox itself is running?
18:26:48sanitariumwhen the iRiver is off
18:26:52sanitariumand plug it into the USB
18:27:02sanitariumit doesn't show up in kubuntu
18:27:40linuxstbAnd if you plug it into USB with Rockbox running, does it show up?
18:27:59MmmmLinusN: I was thinking about the bpm steps and thought an accelerated change would be best... so
18:28:05sanitariumbut it does in iRiver Firmware
18:28:12sanitariumBTW, it worked
18:28:17sanitariumthe extraction
18:28:35linuxstbOK, so it's not just a bootloader problem, it's a general Rockbox problem as well.
18:29:58MmmmLinusN: so i've made this patch what do you think?
18:30:53MmmmLinusN: it includes the fix for pcm_stop too
18:31:14sanitariumdoes it mount for you linuxstb?
18:33:47 Quit darkless (Read error: 104 (Connection reset by peer))
18:34:02 Join darkless [0] (
18:34:46 Quit darkless (Client Quit)
18:35:58linuxstbsanitarium: I don't have a h300, and I don't use the automounter anyway.
18:38:33 Quit alphakiller|away (Read error: 110 (Connection timed out))
18:39:14 Quit Skydemon ("Nettalk6 -")
18:40:50linuxstbeli_sherer: I'm going to try to get the sim working on my Mac (big-endian) this evening. If I succeed, then I can test your plugin there. But I can't see any obvious endian issues in your code.
18:43:43 Quit NicoFR ()
18:46:13eli_shereri got the bug!!!!!woooooooo hooooooo!!!!!!! it now works perfect!!!!!
18:46:30eli_shererbut the animation still gitters...
18:47:03eli_shereri can see clear gifs as if i opened them on the computer with acdsee
18:50:10 Join NicoFR [0] (
18:52:05NicoFRthe build process on my devkit just stopped working :(
18:52:12NicoFRlibplugin.a takes forever
18:53:10NicoFRwould you guys have any idea what's happening ?
18:53:58sanitariumhappend to me before
18:54:08sanitariumit is you computer has been on for a while and is slow
18:54:14sanitariuma restart fixed it for me
18:54:18NicoFRi did that
18:54:42NicoFRi already had the same kind of prob and restart fixed it
18:54:48NicoFRbut this time it doesn't
18:54:54eli_sherernow i fixed the animations also...whoo hoo!
18:55:06eli_shererlinuxstb: are you here?
18:55:18sanitariumany special beta version for the IRC'ers?
18:55:43eli_shererthe thing is: it doesn't work on the target only simulator!
18:56:07 Part Polo_o
18:56:11linuxstbeli_sherer: What was the bug?
18:56:16eli_shererand apperantly it works on ipod so it just a processor thing we honna figure soon
18:56:56eli_shererwell there were planty of them but i'm gonna send to the same address the new one and i want you to test it for me
18:57:14eli_sherernow i use the dynamic allocation again cause it's better and faster
18:57:52NicoFRnow it works again
18:58:12NicoFRit appears rockcalendar was stopping the build from working
18:58:38eli_shereri still need to get the hand of the disposal code so some animations will have cut-outs...
18:59:14eli_sherer (it's there)
18:59:32NicoFRwell the last think i had done was update to that version
19:00:00NicoFRand then i removed rockcalendar.c from SOURCES and it worked again
19:00:21NicoFRi'll rty again
19:01:15 Join _FireFly_ [0] (
19:02:14*eli_sherer asks for help (PLEASE HELP! WHY DOES THIS GIFPLUG DON'T WORK ON THE "REAL" H300?)
19:02:18NicoFRlostlogic: i finally got to trying yout naw ab repeat patch
19:02:38NicoFR* "yout naw" -> "your new"
19:02:45NicoFRworks fine
19:03:24NicoFRbut i think i'll revert it because i don't really use that feature and you said it had an impact on perf...
19:03:30sanitariumoh, yeah, one more question about about linux, when I type "make" into konsole, it says Make: Bash not found
19:04:36linuxstbeli_sherer: Your new version doesn't work at all on my ipod now. But it does display both my test gifs correctly in the sim.
19:04:40 Quit akaidiot (Read error: 110 (Connection timed out))
19:05:09eli_shererwait...i have a newer version
19:05:43 Quit vger_ ("Erection reset by queer")
19:05:48eli_sherer(+ i guess the dynamic memory allocation is still an issue)
19:09:04linuxstbeli_sherer: Yes, it's a malloc problem. I replaced the call to plugin_get_buffer() with a static (512KB) malloc buffer, and it's working fine on the ipod now - displaying the images correctly.
19:09:21lostlogicNicoFR: new version posting shortly that will have minimal impact on performance.
19:10:15NicoFRand also your patch 3 is a bit messy: it changes apps/settings.h.orig
19:11:23 Nick lars|offline is now known as _lars (
19:14:08NicoFRbtw how do you make a diff with only the changes concerning the particular feature you are working on ?
19:14:53eli_shererlinuxstb: how did you do that?? (in what words exactly?)
19:15:50lostlogicNicoFR: yeah, I noticed that, 4 won't do it.
19:15:55eli_shererlinuxstb: has it inflated the rockfile to be a 512KB file?
19:16:15lostlogicNicoFR: I usually copy the files that apply to this feature to a clean source copy and diff that against another clean copy, pulling out any non-relevent changes.
19:17:04NicoFRyou don't diff against CVS ?
19:17:13 Join webguest69 [0] (
19:18:03 Nick DrMoos is now known as Moos (
19:21:47lostlogicNicoFR: no, because I'm too lazy to find those cvsutils to fake add files and stuff.
19:22:04NicoFRwell it seems as soon as i add rockcalendar.c to the SOURCES file, i can't "make" anymore
19:22:36NicoFRlostlogic: in the devkit i can just use "cvs diff -u"...
19:22:44 Join linuxstb_ [0] (
19:22:48 Quit Rob2222_ ()
19:22:56_FireFly_NicoFR: but not if you have added new files ;)
19:22:58 Quit linuxstb (Nick collision from services.)
19:23:02 Nick linuxstb_ is now known as linuxstb (
19:23:04_FireFly_and new-dirs
19:23:12_FireFly_in which the new-files resists
19:23:21linuxstbeli_sherer: No, my "pluggif.rock" is 5028 bytes.
19:23:36linuxstbI just define the buffer as: "unsigned char static_malloc_buffer[512*1024];"
19:23:45_FireFly_the static buffer is afaik only generated on runtime as a static buffer
19:23:55linuxstbAnd then initialise malloc_buffer and malloc_size to refer to static_malloc_buffer instead of calling plugin_get_buffer().
19:24:55 Join Rob2222 [0] (
19:25:16eli_sherernow i get it
19:25:25linuxstbBut this obviously should only be a short-term fix until the plugin loader is fixed.
19:27:33eli_shererthe buffer_size is an integer doesn't it means there is only 65k for storing things?
19:29:00 Join edx [0] (
19:31:27 Nick _lars is now known as lars|offline (
19:32:07eli_shererlinuxstb: my rock file gets to 512KB when compiled (real h300)
19:33:18 Nick lars|offline is now known as _lars (
19:37:57 Join SteveL__ [0] (
19:38:33 Quit DangerousDan ("Miranda IM! Smaller, Faster, Easier.")
19:40:58 Quit gtkspert ()
19:54:42 Nick _lars is now known as lars|offline (
19:55:07***Saving seen data "./dancer.seen"
19:56:05lostlogicnew version of swcodec ab_repeat patch posted. Can someone on a hwcodec machine test to make sure I didn't break that? I moved more code around this time to reduce the performance penalty of the patch on swcodec (since the feature enabled check must be done anytime the patch is present on swcodec currently).
19:56:16NicoFRi'll test it
19:56:28NicoFRbtw i'm the one who made patch n░ 1404763
19:56:35NicoFRhowever i didn't submit it
19:56:38lostlogicNicoFR: you are on H3x0... I need someone on a machine with hwcodec.
19:56:44NicoFRoh sorry
19:56:53NicoFRi read too fast :)
19:56:58lostlogicbut feel free to test too :-D
19:57:03NicoFRi will
19:59:13 Join Coldtoast [0] (
19:59:26ColdtoastI had a thought
20:00:04Coldtoastwhen I change the file display and there's no data, like if I change to ID3 Db nd there is none, it shouldn't just display a lank screen I reckon
20:00:27Coldtoastshould be a message telling you there's nothing to display
20:01:38ColdtoastI accidentally changed mine to ID3 DB when I was moving stuff around and bunping buttons and when I went to select different music, I just saw a blank screen and was all like "WTF!?? It's crashed?"
20:07:05 Join mirak [0] (
20:11:09 Quit Coldtoast ()
20:13:32 Quit sanitarium ("Leaving")
20:13:48 Join linuxstb_ [0] (
20:14:10linuxstb_eli_sherer: Are you using the current CVS version of Rockbox? It sounds like you haven't updated recently.
20:18:46eli_shererwhy is that?
20:19:29 Join PaulJ___ [0] (
20:20:49 Nick PaulJ___ is now known as PaulJ (
20:22:53 Quit XavierGr ("Trillian (")
20:24:05 Join PaulJ_ [0] (
20:25:22 Quit uwe_ (Read error: 110 (Connection timed out))
20:26:30 Join uwe_ [0] (
20:26:53linuxstb_eli_sherer: That's the reason your 512KB buffer is in the .rock file. amiconn changed that a few days ago (8th Jan I think)
20:27:43linuxstb_But he also introduced the plugin_get_buffer() bug, so you may be better off not updating until that's fixed.
20:28:11eli_shererbut im not using it for now
20:28:19eli_shererand 65k isnt enough for me
20:28:43miraklinuxstb: he changed what ?
20:28:56eli_sherer+ i found the bug in real h300 and the background thing in the ani gifs
20:30:13 Nick SteveL__ is now known as SteL (
20:31:38linuxstb_static buffers can be up to 4GB in size (in theory)
20:32:36linuxstb_mirak: The bss section (which store uninitialised variables) is no longer in a plugin's .rock file.
20:33:25linuxstb_eli_sherer: What stopped it working on a real h300? Was it an endian problem?
20:38:42eli_shererno...something with palette loading
20:39:01eli_shererwait...i meant yes!
20:39:14 Quit PaulJ (Read error: 110 (Connection timed out))
20:42:06 Join Febs_ [0] (
20:42:26 Quit Febs (Read error: 110 (Connection timed out))
20:42:50 Quit PaulJ_ (Read error: 110 (Connection timed out))
20:44:58 Join Menollo [0] (
20:46:25 Join akaidiot [0] (
20:48:36 Join Lear [0] (
20:49:15eli_shereri mean 65k because malloc_free (integer) can't hold more then 65536 (so it will run out sooner)
20:49:31 Join petur [0] (
20:50:15Bagderan int is 32 bit on our platforms
20:54:55linuxstb_Also, malloc-type functions should use "size_t", which is "unsigned long" in Rockbox.
20:57:38miraklinuxstb: where is it then ?
20:58:16linuxstb_size_t is in <sys/types.h>
21:04:41 Join DangerousDan [0] (
21:10:56preglowanyone experienced in compiling 2.6 realtime patched kernels here? :/
21:12:49linuxstb_That sounds like a fun way to spend Friday eveniing...
21:13:53 Quit webguest69 ("CGI:IRC (Ping timeout)")
21:14:03 Join sanitarium [0] (
21:14:26sanitariumwhat should I install on linux so I can do the "make" command
21:14:47linuxstb_Which distribution?
21:14:54sanitariumcause in kubuntu, i type "make" to install a tar.gz and it says "BASH: Make not found"
21:15:39dpassen1you have to extract the files before you can ./configure && make && make install
21:16:09sanitariumi did
21:16:21sanitariumtar zxvf filename.tar.gz
21:16:24sanitariumcd filename
21:16:34sanitarium"bash, make not found"
21:16:38linuxstb_Try "apt-get install make"
21:16:42dpassen1do ./configure first
21:16:51sanitariumyeah, I do configure aswell
21:16:55sanitariumok, linuxstb
21:17:02dpassen1then sudo apt-get install make
21:19:36sanitariumwhats the "zxvf" thing for tar.bz2 again?
21:22:18sanitariumthat looks like it worked
21:22:28sanitariumbut, where did it install to?
21:22:43sanitariumsorry for all the stupid questions
21:22:55ender`in the current directory
21:25:16sanitariumconfigure: error: Can't find "javac" in your PATH
21:25:25sanitariumDo I need to install java?
21:25:51linuxstb_What are you trying to install?
21:26:24sanitariumthat is java, isn't it?
21:30:07sanitariumany solution?
21:30:34 Join elisherer [0] (
21:30:55 Quit Mmmm ()
21:32:19linuxstb_Doesn't look a straightforward installation - but that's because iriverter relies on lots of external programs.
21:34:31sanitariumwell, I have mPlayer
21:34:39sanitariumand java just installed
21:34:46sanitariumlooking for swt ATM
21:34:48 Quit Menollo ("CGI:IRC (Ping timeout)")
21:34:54sanitariumso, I hope it will work out
21:35:44peturLinusN: in case you're still interested: I remade the brightness bugfix patch (do tell me if I'm still screwing up) and also cleaned it up a bit (no more backlight method in pcf50606.c)
21:37:34peturalso, do you approve of the way L+R gain is modified? I can change it for the other platforms as well...
21:40:46elisherer[thought] if we could write a jvm for rockbox it'll rock (as rockbox do)!
21:42:13 Quit vektek ()
21:42:37preglowlinuxstb_: very fun, actually
21:42:40preglowat least i've got beer
21:43:12 Join vektek [0] (
21:43:45 Quit linuxstb_ ("CGI:IRC")
21:43:49 Join linuxstb_ [0] (
21:43:55linuxstb_preglow: Any luck?
21:44:31preglowthe latency in the stock 2.6 kernel is... shocking
21:44:43preglow100ms audio buffer, and it _SKIPS_ when i change desktops
21:45:22vektekHi, I would just like to say to anyone that has been involved in the development of this software ..... Well done!!! you guys are amazing!!!
21:46:10preglowwhy, thanks
21:47:20vektekMy pleasure
21:48:10vektekAre dumb questions allowed here?
21:48:37Bagderfire away
21:49:22 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
21:49:41vektekThanks. Why does the volume now show from a minus to a 0 instead of the percentage?
21:49:45 Quit dpassen1 ()
21:49:50preglowbecause the volume is in decibels
21:50:01preglowwhich is a rather meaningful way to show volume changes
21:50:22preglowwe're aware that people who don't know what decibels are might be confused, but still, it's not hard to get used to, is it?
21:50:42vektekI guess it would make more sense if I was clever
21:50:54preglow-6 = half the volume of 0
21:50:59preglow-12 = half the volume of -6
21:51:00preglowand so on
21:51:07preglowthis way the numbers actually mean something
21:51:53vektekso is there such a thing as + decibels?
21:52:06preglowbut 0 db is commonly used as "max level"
21:52:19preglowif we did 6db gain, for example, the result would be distortion
21:53:09vektekWorking as a preamp?
21:54:24preglowand btw, if i get it my way, you will have positive db gains some day
21:54:32 Join linuxstb_ [0] (
21:54:35preglowfor those extra quiet tracks that can stand some extra amplification before they distort
21:54:36mirakpreglow: I don't like this numbers
21:54:42preglowmirak: what numbers?
21:54:53miraknegative number of volume
21:54:57vektekCool. because mine is quiet
21:55:07preglowvektek: i've got a lot of quiet tracks as well
21:55:09***Saving seen data "./dancer.seen"
21:55:14preglowmirak: well, ok, you're not alone in that
21:55:34preglowbut for once i'm happy with how the volume works :)
21:55:37mirakpreglow: just wanted to express my voice :)
21:55:43 Join solexx_ [0] (
21:56:01mirakpreglow: problem is that we are used to have 0 like zero sound
21:56:22markunI'm happy with it
21:56:25mirakI guess the negative value is just the attenuation
21:56:26vektekWhat player are you using?
21:56:28lostlogicvektek: if you use the replaygain, you can adjust the preamp... I use a +5dB preemp with replaygain's default settings to get the range I like.
21:56:47preglowmirak: not me, i'm used to 0 meaning max
21:56:47mirakI heard the sound chip only reduce the nominal gain
21:57:01preglowlostlogic: i know, i want that functionality moved out of replaygain
21:57:08mirakpreglow: means we must learn the lowest
21:57:13mirakI don't remember what is it
21:57:15preglowlostlogic: problem with the replaygain settings is they only work when the files you play are tagged replaygain
21:57:16 Join Ryles [0] (
21:57:41mirakhow is it that there is more gradutations in rockbox than on the original firmware ?
21:57:52lostlogicpreglow: *nod* that's fine for me, because the songs that aren't replay gained are already louder in general sine replay gain's default setting pretty much reduces everything.
21:58:12markunpreglow: but having a seperate preamp for replaygained files is not such a bad idea
21:58:14vektekThanks lostlogic
21:59:00markunat least for people who don't have all their files tagged.
21:59:00 Quit edx (Read error: 104 (Connection reset by peer))
21:59:51preglowvektek: they just decided to drop a bunch of them
22:01:48linuxstb_So can I replaygain mp2 files for use in Rockbox?
22:02:08 Quit uwe_ (Read error: 110 (Connection timed out))
22:02:27 Join einhirn_ [0] (
22:03:03vektekdrop a bunch of what?
22:03:03markunlinuxstb_: you could add it in a apev2 header with foobar, but that's probably not wat you want, right?
22:03:36preglowvektek: drop a bunch of volume levels
22:03:46preglowthey've only got 40, the hardware allows tons more
22:04:08 Quit akaidiot (Success)
22:04:12 Quit linuxstb_ ("CGI:IRC")
22:04:18 Join linuxstb_ [0] (
22:04:31 Quit linuxstb (Read error: 110 (Connection timed out))
22:04:36 Join linuxstb [0] (
22:05:52linuxstb_markun: So Rockbox supports ape tags in mp2/mp3 files?
22:05:56vektekWhy is that?
22:06:24markunno, but we you can add it to your build very easy
22:06:38linuxstb_My build? Why not add it to CVS?
22:06:46markunask preglow
22:07:07linuxstb_preglow: Why don't we read ape tags from mp2/mp3 files? :)
22:07:15vektekwhen you relpy to someone do you type it in or is there a clever way of inserting it?
22:07:32markunvektek: my client has tab-completion
22:07:42preglowlinuxstb_: big boys are against it
22:07:46preglowi believe
22:08:00lostlogicBig Brother Rockbox is watching.
22:08:09linuxstb_Do I have any other options for replaygain tags in mp2 files?
22:08:15 Join Midgey34 [0] (
22:08:22preglowbut no, i think we can have ape tags tomorrow if someone just deals with how it should work with existing id3 tags
22:08:43markunignore the id3 if there are ape tags would be my guess
22:09:22markunalso no more encoding problems, always UTF-8
22:09:43linuxstb_When foobar adds ape tags, what does it do with any existing id3/id3v2 tags?
22:09:54Learlinuxstb_, vertek: d3v2 tags?
22:10:02 Join xmixahlx [0] (n=xmixahlx@
22:10:30markunWhere's the gapless info in lame files stored btw?
22:10:43preglowin the xing heder
22:11:05preglowthat is, it's been retrofitted to be a lame info header
22:11:42 Quit solexx (Read error: 110 (Connection timed out))
22:14:15psi`i can't seem to record in stereo with rockbox on an iriver h120. am i configuring something wrong or is it not supported yet?
22:15:08linuxstb_Lear: Thanks, id3v2 seems to be the choice. Next question is if there is any software to do it?
22:15:18linuxstb_(adding replaygain to mp2 files)
22:15:21LearOther than foobar you mean?
22:15:28LearAt least I assume it can do it... :)
22:15:51 Join webguest82 [0] (
22:16:01linuxstb_I thought foobar used ape tags, or can you specify the tag format?
22:16:17webguest82hi there! just want to say a BIG thanks to all rockbox devs for such a great peice of software (i have an h100)
22:16:36LearSure can; you might need to install an id3v2 component, but that's at least in the 0.8.3 "special" package.
22:16:40preglowfoobar uses idv1/idv2/ape
22:16:47preglowwhat you want it to
22:17:01webguest82and does the iriver h100 version play mp4 files, even if it is an mp4 video it will play the audio?
22:17:03 Join XavierGr [0] (
22:17:17*Lear is looking for that mp2 test file he once had...
22:17:27preglowi've got mp2 files if you want
22:17:42linuxstb_And presumably I can just set it going to recursively process a disk full of files, only processing the .mp2 ?
22:18:00webguest82do i need mp2?
22:18:37Learlinuxstb_: well, you do need to create a playlist...
22:18:39peturpsi`: internal mic is mono, line-in should be stereo
22:18:45markunwebguest82: they were not talking to you
22:19:14linuxstb_Lear: Could I just use "find" to create a playlist?
22:19:16markunI don't know if it will play mp4 video files. Did you try it?
22:19:24webguest82is there a list of codecs that work?
22:19:35preglowLear: still need a test file?
22:19:42Learsure, then load that in foobar.
22:19:57linuxstb_webguest82: The "SoundCodecs" wiki page - see the table at the bottom.
22:20:02webguest82my boy freind got me some mp4s, and i just thought i could copy them onto rockbox?
22:20:05Learnot important really, just to see if foobar does add replaygain tags to mp2 files...
22:20:16preglowi'm in linux, so can't chec
22:20:26Learbut it seems I've deleted the file I had...
22:20:51linuxstb_webguest82: Assuming your .mp4 files are AAC (which is most likely), then no, Rockbox can't quite play them yet.
22:22:00webguest82thanks linux person!
22:22:13webguest82will try and convert or somehting lool
22:22:21psi`aah thanks peter
22:22:43peturis the same ;)
22:22:47linuxstb_If you do, then keep the originals. You'll lose some quality when you convert, and hopefully Rockbox will play mp4 files in the future.
22:23:15 Join einhirn [0] (
22:23:29 Quit sanitarium (Read error: 110 (Connection timed out))
22:24:15Learbtw, anybody working on faad?
22:24:21 Join sanitarium [0] (n=sanitari@
22:26:01Learlooks good, yes (foobar tagging mp2 with id3v2 replaygain tags, that is)
22:26:16 Join ashridah [0] (
22:26:16linuxstb_Thanks for checking.
22:26:44linuxstb_And no, I don't think anyone's working on faad. Someone should run it through lostlogic's profiler.
22:26:54 Join Mmmm [0] (
22:27:38Learwasn't it that one that caused ice:s?
22:28:14linuxstb_You mean with the profiling enabled?
22:28:23linuxstb_If so, I think you could be right.
22:29:32 Join Shebb [0] (
22:30:24 Part Midgey34
22:30:27 Join TCK [0] (
22:30:39 Join PaulJ [0] (
22:31:51LearYes, the patch says that faad and alac had such problems ("I think", so it might be worth trying, after all).
22:32:39lostlogichehe, if I wrote that, then I couldn't get them to compile with it :(
22:33:04ShebbIs anybody working on AB support for the iRivers?
22:33:09 Join PaulJ_ [0] (
22:33:17lostlogicShebb: done, and patch posted on sourceforge
22:33:25Shebbis it seemless?
22:33:30lostlogicShebb: no.
22:34:12ShebbI was wondering if it would be able to implement a seemless AB with the current audio codec
22:34:27lostlogichmm... it should be possible...
22:34:48lostlogicwould have to make sure the PCM buffer is not flushed...
22:34:49ShebbI do not know the audio codec at all and it is not documented much
22:35:13lostlogicYou want to give it a try, or would you rather I take a look?
22:35:22ShebbI was guessing some of the seek functionality could be used
22:35:35lostlogicseek isn't seemless
22:35:48Shebbno, your are right...
22:36:20ShebbI really do not know anything about the audio codec so you would probably manage before me
22:36:43lostlogichehe, I'll see if I can do it pretty easily... can't spend much time on it, I want to get working on H340 power management.
22:36:55NicoFRlostlogic: i tested the new A-B repeat patch
22:36:58NicoFRseems to work fine
22:37:27lostlogicNicoFR: good deal
22:37:31NicoFRonly one thing : when you browse files and then press play to return to WPS, the repeat points are lost
22:37:39 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
22:37:54NicoFRthat's the only thing i noticed
22:38:09lostlogichmph, can you comment to that effect on the patch tracker... I don't feel inclined to fix that, but someone else might :)
22:38:27 Join kkurbjun [0] (
22:38:44NicoFRok i'll do it
22:39:18RylesHello all
22:39:20ShebbI am not sure if the current codec could manage this...
22:39:34 Quit PaulJ (Read error: 145 (Connection timed out))
22:39:38 Join webguest57 [0] (
22:39:53kkurbjunamiconn: I'm looking into implementing an auto frameskip, but I was wondering if there is any functions offered to plugins that will provide more accurate timing then current_tick
22:39:58 Quit einhirn_ (Read error: 113 (No route to host))
22:40:34 Join webguest24 [0] (
22:40:34 Quit webguest57 (Client Quit)
22:40:44RylesI've been considering joining you guys and trying to get the gmini port going again.
22:41:09webguest24shebb: any chance of your playlist patch being commited ?
22:41:30 Join lowlight [0] (
22:42:12lostlogicShebb: the lack of seemless might have been just because I built it on the old hwcodec code with minimal modifications −− let's see if a quick change makes it work.
22:43:19 Quit sanitarium (Read error: 110 (Connection timed out))
22:43:38 Join sanitarium [0] (
22:43:51Learhm... faad filterbank.c looks like it could benefit from some assembler...
22:44:39Mmmmkkurbjun: how about timer_register ?
22:44:45lowlightI think I found a bug in the mpa codec where sometimes the last frame is skipped.
22:45:18RylesThe only hardware I've got is an xs 200. The info in the wiki only has base addresses for 120 and SP and I don't want to turn it into a brick.
22:46:01BagderRyles: the good old 'jyp' had an emulator
22:46:12Bagder(jyp being the main gmini man)
22:46:27webguest24lowlight: I have experianced that as well, the skipping or missing of the last few seconds on a mp3 track
22:47:08RylesBadger: I take it he hasn't been seen for a while?
22:47:17Bagderno, he vanished
22:47:26Bagderhis computer broke down and he couldn't get it replaced
22:47:41webguest24thats why he isn't typing now
22:48:06 Quit lowlight ("CGI:IRC")
22:48:14 Join lowlight [0] (
22:49:06RylesPitty. I've only just started looking at it. Will see if I can get the emulator going.
22:49:12lowlightwebguest24: I mean just the last frame (not a few seconds).
22:50:20 Quit PaulJ_ (Read error: 110 (Connection timed out))
22:50:41vektekanyone know how to get the replayagain preamp to work?
22:51:36webguest24set RG to on, then set preamp to value
22:51:55webguest24assuming your files have been RG'd
22:52:14vektekI have done that but the level doesn't change
22:52:23 Join Vapour-X [0] (
22:52:31lostlogicvektek: your files have to be replaygain tagged for that preamp to be applied (says preglow)
22:52:54webguest24are your files RG'd to ID3 V2 or V1 tags
22:53:09vekteksorry I'm a bit stupid.... How do I do that?
22:53:13Vapour-Xhi san, eli & febs
22:53:24Vapour-Xoh and paprica =]
22:53:55webguest24vektek: there is a guide over at Mistic under Rocbox forum general
22:54:03preglowLear: so faad doesn't work with profiling, no?
22:54:20lostlogicShebb: seemless works, I think I might need to tweak the adjustments, but the basic functionality is fine.
22:54:24LearI think not, but ask lostlogic to be sure.
22:54:27preglowi sorely wish the ffmpeg people started working on an aac decoder
22:54:32preglowLear: he says he hasn't tried it
22:54:44vektekthanks.... Sorry if I seem dumb but I'm very new to this
22:54:54mirakit's incredible how most mobile phones screen look like shit in comparison of the H300 screen
22:54:57Learwell, I haven't either. :)
22:55:14mirakI have seen the ipod video for real
22:55:21webguest24vektek: np. the guide at mistic river is good for this
22:55:24Learpreglow: why should they make an integer decoder though...
22:55:36mirakI was impressed by the size of the screen. Bigger than H300, but the ipod is not bigger
22:56:27LearHm... Not much done on faad with regards to iram...
22:57:19mirakpreglow: do you think you will ever be able to exploit the ipod video decoding chip ?
22:57:45preglowmirak: don't know, it'll depend on tons and tons of work
22:57:53preglowLear: well, they've done so before
22:58:06mirakthe chip is doing hardware decoding or is it software ? preglow ?
22:58:29 Quit Mmmm ()
22:58:30preglowthe chip itself is hardware, like any other chip
22:58:34preglowbut it is a cpu/dsp/whatever
22:58:40preglowso you program it like any other cpu
22:58:42mirakprobably that's like the shaders of the video cards
22:58:43preglowat least, as far as i've seen
22:58:51preglowdon't expect anything like that
22:58:55Shebblstlogic: welldone!
22:59:07preglowif it was anything like that, the job would be impossible without specs
22:59:33preglowi think it'll be pretty much impossible as it is
22:59:39lostlogicNeed help: To make AB repeat work seemlessly, I calculate when to jump without hte PCM buffer latency so that it jumps when the PCM buffer has been filled to the right frame. Thsi is all fine and dandy, but the UI is wrong, because it draws _with_ accounting for the latency, and also shows the bar jump LATENCYms before the audio jumps.
23:00:17DrumRBoy320|Awayir Rockbox Nano compatable?
23:01:09preglowdoesn't work 100% yet
23:01:31webguest24vektek: link to Replay Gaining
23:01:48lowlightpreglow: I think I see a bug with the mpa codec where the last frame isn't always decoded (not related to gapless).
23:02:40preglowwith or without seeking?
23:02:57lowlightdoesn't matter
23:03:05preglowexample file?
23:03:50lowlightProbably any file without an id3v1 tag.
23:04:00preglowthe id3v1 tag doesn't enter into it
23:04:06preglowat least it shouldn't
23:04:37*webguest24 hopes someone commits shebb's playlist patch
23:04:38lowlightLet me explain...
23:05:11 Part webguest24
23:06:01lowlightIf there's no id3v1 tag, the size of the last request_buffer (mpa.c line 170) is the size of the bytes left in the file.
23:06:14 Quit ashridah ("shopping")
23:06:22preglowsounds like you could fix this one yourself :-)
23:07:00 Join linuxstb_ [0] (
23:07:02mirakpreglow: this chip cannot be found for consumers ?
23:07:16preglowmirak: no idea, but the information on it sure can't
23:07:17lowlightBut line 401 in libmad/frame.c wants the buffer MAD_BUFFER_GUARD bytes larger or else it generates a MAD_ERROR_BUFLEN error.
23:07:29preglowmirak: they say the manufacturer (broadcom) is known for never giving out info unless they have to
23:07:40lowlightAnd it gets skipped.
23:08:09mirakpreglow: how is it possible that we cannot acces to docs of the hardware we use ?
23:08:14preglowmirak: ask them
23:08:16lowlightThe if there's an id3v1 tag at the end, there's enough buffer space to decode the last frame.
23:08:33lowlightpreglow: follow that?
23:08:41mirakpreglow: Linux have this problem in general
23:08:50preglowi'm not sure, i've had beers
23:08:54preglowi need to look at the code in question
23:09:06preglowbut anyway, you got a proposed solution? you sound knowledgable enougj
23:09:20preglowand anyway, i don't know much about that part of the code
23:09:22lostlogicShebb: patch with seemless AB repeat posted to sourceforge. As I said in here, the UI is wrong currently, but the sound is right.
23:09:24mirak"knowledgable enougj" how many beers ?
23:09:35preglowhaha, fuck off, i always type like swine
23:10:01mirakok 5 25cl beers is my bet
23:10:08preglowtoo low
23:10:22mirak4 50cl ?
23:10:46lostlogicgah, damned partial litre measurements! Whatever happened to a pint of ale!?
23:10:48ShebbThanks a lot for that! I will patch it up now
23:10:56preglowlostlogic: can't get pints here :///
23:11:09preglowthat is, unless you're in a posh pub
23:11:23preglowwith luck you get 0.5
23:11:37*preglow packs bags and moves to england
23:12:17miraklostlogic: wasn't sure it was called a pint there
23:12:37mirakpints cost less
23:12:54mirakwell the cl cost less at least ...
23:12:55*Lear worked in england a year ago from now...
23:13:09preglowlowlight: hmm, if that part of the code is triggered, i'm not sure there even is enough data left in the buffer to contain a full frame
23:13:15preglowlowlight: i guess it depends on bitrate
23:13:21RylesHa ha. England is great. We only get pints (or bigger) and now we have 24 hour licencing :)
23:13:24lowlightpreglow: I don't know the proper solution as I'm not sure why the extra MAD_BUFFER_GUARD is necessary.
23:13:38*petur could ship preglow some Belgian beer ;)
23:13:46preglowshould is the word, should
23:13:55 Quit mirak ("Ex-Chat")
23:14:20preglowRyles: plus you have bloody excellent beer
23:15:20RylesThere's nothing wrong with Belgian beer. But yes we are great.
23:15:20preglowat least i have some family left in england i can use as an excuse to go there
23:15:34preglowand live for free :>
23:17:05lowlightpreglow: it get's triggered, probably all the time for mp3's with no id3v1 tags at the end...
23:17:11lostlogicrockbox devs: In the latest AB repeat patch, I added a function to the audio.h and playback.c for swcodec called audio_seemless_seek, because it's something that is possible on SWCODEC that was not supported by the API, and will probably come in handy for future features (like seek-peeking)
23:17:28Learlowlight: hm... my mp3:s don't have id3v1...
23:17:43preglowlostlogic: problem is, id3v1 _should_ get stripped by the playback engine
23:17:51preglowahh, lowlight i mean
23:17:57preglowso libmad should never see them
23:18:04preglowso there's probably another bug afoot
23:18:31lowlightpreglow: hmm
23:18:33preglowlostlogic: no protest from me
23:22:54LearInteresting... Standard FAAD decodes a Nero encoded track, Rockbox doesn't...
23:23:18preglowwhy's that interesting? the two don't exactly share parsers, heh
23:23:31LearBoth use libfaad2?
23:23:39preglowwe use our own mpeg container parser
23:23:47preglowand that might be were we miss out
23:23:52LearYou talking metadata or codec here?
23:24:04preglowlowlight: anyway, i'd like a faulty file if possible
23:24:05DrumRBoy320|Awayhello, has linus released a bootloader for h300's that wouldnt usually work right before? my h300 has had problems twice when i upgraded (2/2 times)
23:24:45Learpreglow: well, I use my version that groks non-iTunes files...
23:24:59LearSo the codec starts, but fails after half a second or so.
23:25:00preglowgod knows
23:25:01preglowi hate faad
23:25:08preglowit's a crap codec
23:25:17 Join Skydemon [0] (
23:25:18lowlightpreglow: strip the id3v1 tag from any file.
23:25:30preglowlowlight: like i said, that should be our current situation
23:25:34 Join akaidiot [0] (
23:25:37preglowid3 tag or no, playback should strip it before load
23:26:30 Quit elisherer ("CGI:IRC (EOF)")
23:26:32lowlightpreglow: I just id3v1 tag the last buffer_request returns 313 which is the bytes left and the final frame size
23:26:58lowlightpreglow: but libmad returns a MAD_ERROR_BUFLEN error so it's not decoded.
23:28:07 Quit Vapour-X ()
23:28:09lowlightpreglow: same file with an id3v1 tag, buffer_request returns 441 (313+128) and the final frame is decoded
23:28:38Learlowlight: right, mad might need some padding to decode last frame. read something about it on the mad mailing list.
23:29:13Leardidn't think much about it at the time...
23:30:35 Quit NicoFR ()
23:30:41 Join mirak [0] (
23:30:43preglowand they haven't felt the need to bugfix this yet?
23:31:00 Join Skydemon_ [0] (
23:31:11mirakdamn, no wonder why I didn't knew what are the gigabeats. They are not distributed in france
23:31:25 Join vger_ [0] (
23:31:41lowlightLear: I only noticed because I've been messing with auto-removing the encoder gaps for non-lame files and the end gap was always a frame off.
23:32:00 Quit Skydemon_ (Client Quit)
23:32:22preglowhave you now
23:32:37 Quit stamppot ("CGI:IRC (EOF)")
23:32:37preglowthat's weird, gapless mp3 playback has been working okish for me
23:32:50preglowbut then again, i don't think i have many id3v1 files
23:32:53 Join uwe_ [0] (
23:33:03 Join RotAtoR [0] (
23:33:05lowlightbtw, audacity has the same problem (the final frame is missing).
23:34:59preglowprobably uses libmad, then
23:35:11lowlightpreglow: I've noticed a lot of lame mp3s have enc_padding > 1 frame, so you might not notice.
23:35:25 Part mirak
23:35:29 Quit Skydemon ("Nettalk6 -")
23:37:17linuxstb_Lear: Did I understand that you have your own mp4 parser?
23:37:29lowlightpreglow: Just to be clear, it appears to effect files _without_ id3v1 tags.
23:38:18Learlinuxstb_: yes, I do.
23:38:18 Quit _FireFly_ ("Leaving")
23:38:19 Part Ryles
23:38:34BHSPitMonkeyman, i can't wait for the rockbox emulator for iPL
23:39:00Bagderbuild the rockbox sim and run it on ipl ;-)
23:39:11linuxstb_Does SDL run on the ipod?
23:39:34Bagderright, there's probably a lack of ui lib
23:39:48perplexitywhy not.. ipl uses an fb does it not? Just might need some work on the kbd handling
23:39:56linuxstb_Lear: Do you plan to incorporate it into Rockbox? The one that's there is just taken from the alac decoder - so it's not very complete.
23:39:58BHSPitMonkeyhmm that's actually a thought -provoking idea
23:40:44Learyes, the idea is to do that, once it is more tested...
23:40:52uwe_linuxstb: AFAIK SDL does run on ipl
23:41:39linuxstb_Lear: Good. I assume you're testing it with ALAC as well?
23:42:01LearWell, if you could provide me with a file, sure. :)
23:42:13LearBut I don't see why it wouldn't work. Same container, right?
23:42:55linuxstb_There's some minor differences IIRC, but shouldn't be hard. I'll upload a file for you now.
23:43:19lowlightLear: Found this regarding decoding last frame:
23:43:42preglowlowlight: well, strange, my files are mostly without, but this does explain an ancient problem i had with libmad, where it somehow never would give me the last frame.....
23:44:30Learlowlight: read a shorter, more recent, comment about it.
23:44:42Learpreglow: yes, that would explain that comment...
23:45:05 Quit linuxstb (Nick collision from services.)
23:45:17 Nick linuxstb_ is now known as linuxstb (
23:45:24 Join linuxstb_ [0] (
23:45:28BHSPitMonkeylinuxstb: userper!
23:45:47 Quit linuxstb_ (Remote closed the connection)
23:46:17preglowLear: ahh, yes, i did make a comment about it
23:46:44preglowit's even still there
23:48:20 Quit eli_sherer ()
23:55:14***Saving seen data "./dancer.seen"
23:55:24 Part lowlight
23:58:21 Join edx [0] (

Previous day | Next day