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

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

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

#rockbox log for 2005-10-17

00:00:47amiconnThe question is, what resolution are the replaygain values, and what range do we need to handle?
00:06:25 Quit matsl ("Leaving")
00:07:02 Join matsl [0] (
00:09:35preglowi believe they are floats
00:13:49 Quit preglow (
00:18:24 Join muesli- [0] (
00:19:20 Quit Moos (Read error: 110 (Connection timed out))
00:25:31amiconnThere. Shiny new math routines in mandelbrot.rock. SH1 low precision: up to 4x speedup, SH1 full precision: up to 3x speedup, coldfire full precision: up to 6x speedup
00:33:16 Join muesli__ [0] (
00:39:34 Join thomjoha [0] (
00:40:14 Nick thomjoha is now known as preglow (
00:40:17preglowamiconn: nice
00:40:36preglowamiconn: noticed any visible differences with the new routines?
00:43:51amiconnThe reduced delta-x and delta-y resolution leads to slight shifting of the axes at higher zoom levels
00:44:23amiconnNot directly related to the new iteration routines, but rather to the reduction of the variables to 32 bit
00:45:21 Join fubar_ [0] (
00:45:21 Quit fubar (Read error: 104 (Connection reset by peer))
00:47:05 Quit goa ("Client suicide")
00:47:52 Quit Febs (" HydraIRC -> <-")
00:55:38 Quit muesli- (Read error: 110 (Connection timed out))
00:59:34preglowlooks good to me
00:59:42preglowand damn, it's fast
01:00:25amiconnHehe, using your emac trick on coldfire... and inlining
01:04:20preglowbut btw, like i said the other day, grayscale plugins seem kind of sluggish after grayscale stopped cpu_boosting
01:04:46preglowdoes the grayscale lib screen updates and so on use lots of cpu?
01:07:31preglowlike i suspected
01:07:43preglowslasheris overflow fix halves the volume, no wonder it no longer overflows
01:12:46preglowhmm, it is quite strange it actually succeeds in getting rid of the overflows... just rescaling the audio should retain the overflows like nothing happened
01:17:40 Join ripnetUK [0] (
01:18:18 Join webguest97 [0] (
01:18:54 Join DMJC-L [0] (
01:19:08 Quit ripnetUK (Client Quit)
01:23:34webguest97I've just put a file to experiment on my iHP-140; it was SACD WavPack [Metallica Black Album BTW :)] 8745kbps; 96000Hz; 6 channels; - obviously it froze my player and a hard reset did the job, but I was wondering if such files could ever be playable on this hardware; just a thought...
01:23:58 Quit matsl (Remote closed the connection)
01:24:06 Quit muesli__ (Read error: 110 (Connection timed out))
01:24:58preglowsacd wavpack?
01:25:01preglowwavpack supports sacd?
01:25:49webguest97I've just downloaded the file from a hub; yes: sacd wavpack
01:25:55preglow96khz sacd doesn't make sense
01:26:22preglowsure it's not sacd converted to pcm wav, then converted to wavpack?
01:26:30 Join linuxstb [0] (
01:26:32preglowbut anywho, i'm not certain 96khz sample rate is supported yet
01:27:10webguest97I don't know whether it was converted in any way
01:27:10preglowalso, i'm not certain how multi channel files are handled
01:27:21webguest97I see
01:27:58webguest97do you think the iriver's hardware could handle such a beast?
01:28:30preglowit depends on wavpack
01:28:45preglowif wavpack has to decode all six channels, then no
01:29:01linuxstbThe iriver can only play back 16-bit/44.1KHz/Stereo, so it's a bit of a waste of resources to decode anything better than that.
01:29:05preglowif it can only decode two of them: then yes
01:29:14 Quit markun ("leaving")
01:29:25linuxstbI think you would have to decode all 6 channels and then downmix to stereo.
01:29:52preglowyes, then you're chanceless
01:30:07preglowwith the overhead of resampling on top of that again, there's no chance
01:30:32webguest97got it :); it would be fun though
01:32:04preglowsure, but also ultimately a waste
01:32:56linuxstbYou would get much higher quality playback by doing a one-time conversion on your PC with high quality (i.e. slow) resampling.
01:33:51preglowoh yes
01:34:00preglowthe current resampler will do a very bad job of resampling 96khz
01:34:07preglowit was more or less only made for 48->44.1
01:35:06webguest97will foobar e.g. be able to resample it to say: 48?
01:35:53linuxstbProbably, but for the iriver you will want to resample to 44.1KHz - that's all the hardware can handle.
01:36:13linuxstbRockbox resamples 48KHz files to 44.1KHz during playback.
01:36:54webguest97didn't know that; thanks a lot for your help :)
01:37:34preglowbut damn
01:37:54preglowplain libmusepack, with no emac opts whatsoever, still generates overflowing audio on target
01:38:00preglowi must be doing something blatantly wrong
01:52:13 Quit actionshrimp ("a bird in the bush is worth two in your house")
01:58:52***Saving seen data "./dancer.seen"
02:05:27 Join toolmanwill [0] (
02:05:56 Quit toolmanwill (Client Quit)
02:11:45 Quit preglow ("leaving")
02:24:53 Join ashridah [0] (
02:39:53 Quit ]RowaN[ ()
02:56:54 Quit thegeek (Read error: 104 (Connection reset by peer))
02:59:27 Join thegeek [0] (
03:58:53***Saving seen data "./dancer.seen"
04:04:55 Join Vlad0man [0] (
04:05:37 Join QT [0] (i=as@madwifi/users/area51)
04:12:46-->"LogBot Annoy tempest_" received from Jaywalker (
04:16:07 Quit QT_ (Read error: 113 (No route to host))
04:22:27 Quit _Vladoman (Read error: 110 (Connection timed out))
05:12:45 Join Lost-ash [0] (
05:13:05 Quit ashridah (Nick collision from services.)
05:13:11 Nick Lost-ash is now known as ashridah (
05:58:56***Saving seen data "./dancer.seen"
06:19:57 Join XShocK [0] (
06:21:26 Quit XShocK (Client Quit)
06:26:38 Join XShocK [0] (
06:38:04 Join Webguest [0] (
06:39:59 Nick Webguest is now known as user109 (
06:43:49zeanybody know of an archive of cool 1-bit images in either vector format or 600x600+ bitmap ?
06:47:08 Part user109
06:56:48 Join amiconn_ [0] (
07:03:38 Join BBub_ [0] (
07:09:39 Join paulproteus [0] (
07:09:45 Part paulproteus
07:10:26 Part XavierGr
07:11:37 Join XavierGr [0] (
07:14:23 Quit amiconn (Read error: 110 (Connection timed out))
07:14:24 Nick amiconn_ is now known as amiconn (
07:21:34 Quit BBub (Read error: 110 (Connection timed out))
07:21:34 Nick BBub_ is now known as BBub (
07:59:00***Saving seen data "./dancer.seen"
08:06:12 Join ender` [0] (i=ychat@
08:17:50 Join goa [0] (
08:19:28 Join linuxstb_ [0] (
08:26:37 Quit webguest97 ("CGI:IRC (Ping timeout)")
08:26:37 Quit t0mas (Read error: 104 (Connection reset by peer))
08:26:37 Join ender_ [0] (i=ychat@
08:26:37 Quit ender_ (Read error: 104 (Connection reset by peer))
08:26:37 Join t0mas [0] (n=Tomas@unaffiliated/t0mas)
08:34:34 Quit linuxstb (Read error: 110 (Connection timed out))
08:50:06 Quit gromit` (Read error: 104 (Connection reset by peer))
08:52:03 Join gromit` [0] (n=gromit`
09:08:57 Nick Lynx_awy is now known as Lynx_ (
09:09:46 Quit tvelocity ("Leaving")
09:11:31 Join markun [0] (
09:13:52HCldoes anyone know how hard it would be to play an mp3 stream out of an .avi ?
09:14:19HClwith rockbox
09:22:40markunHCl: I think it's not too difficult to write a demuxer:
09:25:25SlasheriHCl: should be quite easy with linux and with some demuxer/encoding software
09:25:48Slasheriffmpeg/mencoder/transcode could be useful too
09:33:40 Join webguest18 [0] (
09:34:56 Quit webguest18 (Client Quit)
09:59:02***Saving seen data "./dancer.seen"
10:18:07HClSlasheri: i kind of meant on rockbox itself, not extracting the mp3 stream
10:18:16HClSlasheri: i have some music clips in avi with an underlying mp3 stream
10:32:21 Join tvelocity [0] (n=tony@
11:10:30 Quit DMJC-L (Read error: 104 (Connection reset by peer))
11:33:08SlasheriHCl: oh, i see :)
11:33:46Slasheribut that should be easy too
11:35:03 Join Marcus1 [0] (
11:35:50Marcus1Slasheri: yesterday I asked you about the anti skip buffer
11:36:31Marcus1you explained me your current iriver implementation, but why it didn't like archoses?
11:37:18Marcus1like amiconn said, it's a different way, and the value 0 is possible
11:37:54 Quit ashridah (Read error: 110 (Connection timed out))
11:38:23Marcus1slasheri: you there?
11:39:46SlasheriMarcus1: yes, i think that will be fixed for iriver
11:40:13Slasherisomething like 10s should be a good safety margin
11:40:36Marcus1ok, thanks man
11:40:36Slasheriand then we could add the anti skip buffer setting to that value
11:40:59Marcus1archos way?
11:41:20amiconnSlasheri: The archos code measures the actual disk spinup time...
11:41:33Slasheriamiconn: oh, then it's pretty accurate.. :D
11:41:49 Quit lImbus (Read error: 110 (Connection timed out))
11:41:51Slasherii think we shouldn't make iriver that accurate but instead add some good value..
11:42:00Slasheribecause the buffer space is not so limited
11:43:25 Join ashridah [0] (
11:44:11Marcus1Slasheri: how goes the iriver recording thing? :)
11:44:12Slasheriand the spinup time will be never constant while moving.. at least with the iriver's 1.8" disk. Maybe it's more accurate with larger disks
11:44:24SlasheriMarcus1: i haven't had time to look into that
11:44:28Slasherimaybe in few weeks..
11:46:45Slasheriamiconn: Hmm, or maybe we could save some statistics of the player usage over a longer period (including average disk spinup times) and use that to automatically adjust the buffer margin
11:48:32Slasheriin fact we couldn't use average but the highest disk spinup time from some period
11:56:03 Join Zagor [0] (
11:59:03***Saving seen data "./dancer.seen"
11:59:48 Quit Marcus1 ("CGI:IRC (Ping timeout)")
12:33:21 Join thegeek_ [0] (
12:33:21 Quit thegeek (Read error: 104 (Connection reset by peer))
12:45:34 Join cYmen [0] (
12:50:45 Join cYmen_ [0] (
12:55:58 Join cYmen__ [0] (
13:01:11 Join cYmen___ [0] (
13:01:11***Alert Mode level 1
13:01:11DBUGEnqueued KICK cYmen
13:01:11DBUGEnqueued KICK cYmen_
13:01:11***Alert Mode level 2
13:01:11DBUGEnqueued KICK cYmen__
13:01:11DBUGEnqueued KICK cYmen___
13:01:11***Alert Mode level 3
13:03:27 Quit cYmen (Connection timed out)
13:06:21 Join cYmen [0] (
13:06:21***Alert Mode level 4
13:06:21***Alert Mode level 5
13:06:21DBUGEnqueued KICK cYmen
13:06:21***Alert Mode level 6
13:06:21***Alert Mode level 7
13:06:21***Alert Mode level 8
13:08:27 Quit cYmen_ (Read error: 110 (Connection timed out))
13:11:27 Join cYmen_ [0] (
13:11:27***Alert Mode level 9
13:11:27***Alert Mode level 10
13:11:27DBUGEnqueued KICK cYmen_
13:11:27***Alert Mode level 11
13:11:27***Alert Mode level 12
13:11:27***Alert Mode level 13
13:13:34 Quit cYmen__ (Connection timed out)
13:16:39 Join cYmen__ [0] (
13:16:39***Alert Mode level 14
13:16:39***Alert Mode level 15
13:16:39DBUGEnqueued KICK cYmen__
13:16:39***Alert Mode level 16
13:16:39***Alert Mode level 17
13:16:39***Alert Mode level 18
13:18:41 Quit cYmen___ (Connection timed out)
13:21:52 Join cYmen___ [0] (
13:21:52***Alert Mode level 19
13:21:52***Alert Mode level 20
13:21:52DBUGEnqueued KICK cYmen___
13:21:52***Alert Mode level 21
13:21:52***Alert Mode level 22
13:21:52***Alert Mode level 23
13:24:14 Quit cYmen (Connection timed out)
13:25:26 Join preglow [0] (
13:27:02 Join cYmen [0] (
13:27:02***Alert Mode level 24
13:27:02***Alert Mode level 25
13:27:02DBUGEnqueued KICK cYmen
13:27:02***Alert Mode level 26
13:27:02***Alert Mode level 27
13:27:02***Alert Mode level 28
13:28:55 Quit tvelocity ("Leaving")
13:29:25 Quit cYmen_ (Connection timed out)
13:32:08 Join cYmen_ [0] (
13:32:08***Alert Mode level 29
13:32:08***Alert Mode level 30
13:32:08DBUGEnqueued KICK cYmen_
13:32:08***Alert Mode level 31
13:32:08***Alert Mode level 32
13:32:08***Alert Mode level 33
13:32:27 Join actionshrimp [0] (
13:34:17 Quit cYmen__ (Connection timed out)
13:37:17 Join cYmen__ [0] (
13:37:17***Alert Mode level 34
13:37:17***Alert Mode level 35
13:37:17DBUGEnqueued KICK cYmen__
13:37:17***Alert Mode level 36
13:37:17***Alert Mode level 37
13:37:17***Alert Mode level 38
13:39:22 Quit cYmen___ (Connection timed out)
13:39:52preglowSlasheri: yo, your mpc overflow fix halves the volume of the output, so something's clearly not right
13:42:30 Join cYmen___ [0] (
13:42:30***Alert Mode level 39
13:42:30***Alert Mode level 40
13:42:30DBUGEnqueued KICK cYmen___
13:42:30***Alert Mode level 41
13:42:30***Alert Mode level 42
13:42:30***Alert Mode level 43
13:45:11 Quit cYmen (Connection timed out)
13:46:55markuncYmen___: something wrong with your IRC client?
13:47:37 Quit ashridah ("Leaving")
13:47:40 Join cYmen [0] (
13:47:40***Alert Mode level 44
13:47:40***Alert Mode level 45
13:47:40DBUGEnqueued KICK cYmen
13:47:40***Alert Mode level 46
13:47:40***Alert Mode level 47
13:47:40***Alert Mode level 48
13:49:50 Quit cYmen_ (Connection timed out)
13:54:53 Quit cYmen__ (Connection timed out)
13:55:18 Join ashridah [0] (
13:57:41***Alert Mode OFF
13:59:05***Saving seen data "./dancer.seen"
13:59:44 Join webguest76 [0] (
14:00:00 Quit cYmen___ (Connection timed out)
14:00:44 Quit webguest76 (Client Quit)
14:06:46 Join webguest81 [0] (
14:07:20 Quit webguest81 (Client Quit)
14:26:46 Join webguest82 [0] (
14:28:19 Quit webguest82 (Client Quit)
14:37:01 Join sanguis [0] (
14:42:34 Quit sanguis ("CGI:IRC")
15:05:09 Join RiverFish [0] (
15:37:51 Quit Nibbler ("life is like a rental car, you fuck it up, and give it back.")
15:50:44 Quit ashridah ("Leaving")
15:59:06***Saving seen data "./dancer.seen"
16:09:12 Quit ghode|afk (Read error: 104 (Connection reset by peer))
16:13:39 Join ghode|afk [0] (
16:21:33 Quit RiverFish ("CGI:IRC")
16:23:19 Join FishFace [0] (
16:25:03 Join _FireFly_ [0] (
16:29:27FishFaceHi all. Just got my cygwin build system running and wondering what to do next.
16:29:43FishFaceI'm an iRiver H120 user
16:29:55FishFaceRockbox is already way better than the iRiver firmware
16:30:49FishFaceWhere's a good place to look for the current status of iRiver port work? I'd like to help out
16:41:11FishFaceIs there any software method of increasing sound output from the H120 without clipping? Could do with a bit more power for driving my Etys.
17:04:30Slasheriyes, disabling short circuit protection on uda could give more power output but there is a great risk the chip can be damaged by doing that
17:07:23 Join dpassen1 [0] (n=dpassen1@resnet-233-61.resnet.UMBC.EDU)
17:07:35amiconnSlasheri: Really? Did you try that?
17:08:14FishFacesounds risky
17:11:32 Join linuxstb [0] (n=linuxstb@
17:12:09amiconnhi mr. set top box :)
17:12:34 Quit Zagor ("Client exiting")
17:13:06FishFaceWhat's the status of encoding? Is anyone trying to get Vorbis encoding working or is it likely to be too cpu intensive?
17:13:21linuxstbHello all.
17:13:58linuxstbFishFace: You could also look at for a list of areas of Rockbox that still need work.
17:14:20FishFaceta again
17:14:36markunFishFace: You think RB is way better that iRiver's firmware? Some people think it sounds like a bag of shit ;)
17:16:05FishFaceTo me RockBox sounds cleaner but maybe that's just the placebo effect :)
17:16:19FishFaceI remember reading something about it having different treble and bass boost behaviour
17:16:48FishFaceWhat do you think markun?
17:18:28FishFaceIs there any reason to think that RockBox should sound noticeably different from the iRiver firmware?
17:18:28 Quit solexx (Read error: 104 (Connection reset by peer))
17:18:32 Join DangerousDan [0] (
17:23:32 Join solexx [0] (
17:24:18 Join webguest36 [0] (
17:24:57webguest36Hello people
17:25:13webguest36I've got a quick question please
17:25:56webguest36the folders systeme volume informations are needed for Rockbox, or can I delete them?
17:26:14webguest36directories I mean
17:26:20_FireFly_you can delete is only windows relevant
17:26:37_FireFly_it belongs only to windows
17:27:03webguest36oh ok thanks, not reapear when I'll go to windows?
17:27:48webguest36with the dircache, it's stupid to have unnecesary dirs :)
17:27:57webguest36it's why I asked
17:28:29webguest36Firefly: it will not reapear?
17:28:46Slasheriamiconn: not so much that i could hear a difference. But in theory it might be possible
17:28:58_FireFly_it could reapear
17:29:01Slasheristill better not to try that :)
17:29:34webguest36ok so delete is stupid if i will use windows
17:29:45webguest36it's this?
17:30:08Slasherijust delete it, you don't need that
17:30:34webguest36ok, just want to know if they reapear when I will go under windows
17:30:45Slasheriit is possible, you will see..
17:31:07webguest36it's possible it eat me dircaching time :)
17:31:40webguest36I'll check when I'll back home, thanks guys for informations
17:31:50Slasherievery directory will eat it a little
17:32:15webguest36yes it's why I make clean in my dirs
17:32:17amiconnSlasheri: A propos dirchache - does the dircaching code remember the last used cache size even if it does a transparent rebuild?
17:32:33webguest36bye all
17:32:34 Quit webguest36 ("CGI:IRC")
17:32:36Slasheriamiconn: yes of course
17:32:47amiconnI'm thinking of the situation the something is deleted from the box while the cache is active and valid
17:33:00amiconnWould it reserve less RAM at the next boot?
17:33:05Slasheriyep, on next boot the reserved cache should be smaller
17:33:12Slasheriit should
17:33:36 Join Philip_0729 [0] (
17:33:37amiconnOne could try ... with that dreaded System Volume Information folder
17:34:34amiconnIt tends to have many files and sub folders if you don't delete it regularly (or somehow convince windows you really don't want system restore on the drive)
17:34:43Slasheriin fact deleting wont lower the cache size..
17:35:06amiconnCertainly not live, no. But I mean after a reboot
17:35:24Slasheriah, yes. You need two reboots for that
17:35:49amiconn(or even a second reboot, in case the cache code only computes the necessary size when rebuilding the cache)
17:43:53 Quit FishFace ("Miranda IM! Smaller, Faster, Easier.")
17:45:42markun_FireFly_: To me it doesn't sound any different. But I don't use SRS WOW and maybe my headphones are not so good.
17:45:48markun_FireFly_: Ehh..
17:45:53 Quit ghode|afk (Read error: 110 (Connection timed out))
17:46:41_FireFly_markun: he is gone :)
17:46:49markunyes, I see :)
17:55:54 Join lamed [0] (
17:57:51lamedHello. how do I check the cvs server to update my sourcefiles? i've just 'cvs -z3 co rockbox-devel, there wasn't any real update made & i can see that mandelbrot and som stuff are newer.
17:59:02_FireFly_lamed: you don't need to run this line every time you want to update your source
17:59:10***Saving seen data "./dancer.seen"
17:59:12ender`cvs up -dP
17:59:38_FireFly_only on checkout and then you can easily make a cvs update in the dir where the sources are in
18:01:02lamed~>cvs up -dP
18:01:06lamedcvs update: No CVSROOT specified! Please use the `-d' option
18:01:10lamedcvs [update aborted]: or set the CVSROOT environment variable.
18:01:16_FireFly_go into rockbox-devel
18:01:20 Quit novimon (Read error: 110 (Connection timed out))
18:01:23_FireFly_and then run cvs up -dP
18:01:34lamed(i damb)
18:05:33lamedI'm l00king for documetation about it. what does this exactly do? makes my source folder same as the one in the cvs? -if i changed something will it get overwritten?
18:08:49CtcpIgnored 14 channel CTCP requests in 18 hours and 33 minutes at the last flood
18:08:49*lamed swears to god he has been over that 100 million times, but this time it's there!
18:10:12lamed_firefly_: thanks
18:15:34preglowSlasheri: you read my comment about your mpc overflow fix?
18:15:54 Quit _FireFly_ ("Leaving")
18:28:34 Join webguest93 [0] (
18:37:58lamedI'd love to know what kde are you people using. i'm working with dev-c , to debug or to build I have to go into a cygwin bash shell and use plain syntax commands. I hate debugging code when I can't place a breakpoint / make a watchpoint table that will stay visible. also, dev-cpp doesn't treats rockbox like it's a project, so i always have to go looking where functions are located, headers, etc. advice, please!
18:39:31 Part linuxstb ("Leaving")
18:44:13lamedwhoops... IDE. but of course you already knew that.
18:44:46lamedeveryone using dev-cpp? wow.
18:44:48 Join mirak [0] (
18:45:16mirakther is no bootloader for H300 ?
18:45:29amiconnlamed: No IDE, just cygwin bash and a decent editor
18:46:48ender`emacs? :)
18:47:50lamedamiconn, are you serious..? I'm blasted. don't you rather things in a nice application, cute watchpoints window. ender': can i define emacs to build and debug from it?
18:48:31amiconnlamed: There's gdb if you need to debug something deeply. I never missed an IDE
18:50:15amiconn(and yes, I am serious.)
18:50:16ender`lamed: i don't think there's a thing emacs doesn't do. it's got everything but the kitchensink
18:50:44mirakthere is still no rockbox for the H3x0 ?
18:50:48mirakI guess no
18:50:52mirakit's sad :(
18:52:02amiconnmirak: Nope, but I am pondering to do some lcd driver work (leaving out the actual hardware access), win32 simulator and bitmap conversion tool, based on an assumed pixel format
18:52:14lamedamiconn: I am using gdb, it does gives me the _feeling_ i know what i'm doing (writing to plain ascii screen is a whole-lot cooler) but truth to be told it's not comfterble and sometimes I miss the reason why i cannot do something (-say setting a breakpoint that meets only one time when I excpect it to be met again..)
18:52:43mirakamiconn ok
18:53:00amiconnlamed: In fact I use gdb very rarely
18:53:32lamedso what _do_ you use? i sometimes has to debug something only to deccipher a function...
18:54:04mirakamiconn so if rockbox is avialalbe onH3x0 we will profit of all the advancement ?
18:54:22mirakof rockbox
18:54:26amiconnI'm often verifying functions by writing a testset (e.g. as a plugin)
18:54:28mirakor somethings will not work ?
18:54:33amiconnI don't use the sim often
18:54:41lamedender': what's a "kitchensink"? is this some kind of a new operation system?
18:55:21 Join _FireFly_ [0] (
18:56:00amiconn...and on target there is either no gdb support yet (iriver), or it requires a hardware mod that I don't want to do on my real-use boxes (archos)
18:56:04lamedamiconn: but i'm not quite there yet
18:56:33lamedamiconn: and the sim is very well enough for my needs and studies.
18:57:38amiconnAfair I never worked with breakpoints or whatever in gdb. I only used it for finding the point where the sim crashed some times
18:58:14amiconnEven then it wasn't that helpful
18:58:26lamedamiconn: is xemacs better then emacs? (in what "strong ide" you or everyone are using?) can i define emacs to build & debug code over rockbox, as well as finding function placements for me..?
18:59:29amiconnlamed: I never used emacs so I can't comment on it
18:59:30mirakI don't wan't to troll, but was just wondering if linux would hae a been a so wrong solution ?
18:59:51mirakjust curious
18:59:53lamedit's a great thing.
19:00:12_FireFly_mirak: why do you think so ?
19:00:24mirak_FireFly_: that I am trolling ? :D
19:00:25Slasheripreglow: yes, the volume is indeed lower..
19:00:44mirak_FireFly_: don't know, just that linux have a good code base
19:00:47lamedamiconn: so, what, do you, or most developers, use?
19:00:54mirak_FireFly_: that's why I wonder
19:01:11mirak_FireFly_: I guess linux have to much overhead
19:01:22mirak_FireFly_: or something
19:01:30_FireFly_mirak: ???
19:01:47amiconnI can't speak for most developers, only for me. I am developing under windows, so I have to use cygwin, and I use the ConTEXT editor
19:01:54_FireFly_mirak: i don't understand what do you want to say to me
19:02:08lamedthanks, now that's a stright answer.
19:02:13preglowSlasheri: it's lower by exactly half
19:02:24Slasheripreglow: yep
19:02:28preglowSlasheri: which is also half as loud as it should be
19:02:59Slasheribut the scaler setting seems to be correct if we take a look for example the mpa codec
19:03:29lamedamiconn: "developed mainly to serve as secondary tool for software developers"
19:04:26mirak_FireFly_ the question was why starting with rockbox when there is linux
19:04:56preglowSlasheri: yeah, i too think it looks correct, but it obviously isn't
19:05:28Slasherimirak: i think with linux there would have been much more work to do than without it
19:05:40Slasheripreglow: yes..
19:07:28amiconnmirak: Imho linux would use too much overhead, and I plain don't see the advantage
19:08:20mirakok that's what I wondering
19:08:20amiconnThe only linux thing that would be usable straight away would be the kernel itself. Everything hardware driver as well as every part of the application would still need to be developed
19:09:13amiconn...and the hardware drivers might be even more complex because of more functions, pre-emptive multitasking would add complexity etc etc
19:09:15_FireFly_and it might be bigger then rbx is
19:09:20mirakbut a kind of merge with ipodlinux could have make sens ? I dont talk about rockbox, just if rockbox was linux based
19:11:51lamedwould someone else be kind enough to let me know of some good ide's for building and debugging it through the ide?
19:13:28mirakhem the bootloader for H100 need to be adapted for H300 or redone from scratch ?
19:13:42mirakbeside Linus nobody have the skills to do that ?
19:14:08amiconnIt's not about skills, it's about debugging hardware
19:14:31mirakI though hardware was similar to some point ?
19:14:33amiconnWithout a BDM wiggler, there is no chance to regain access to the box if the first shot doesn't work
19:14:51mirakwhat is it ?
19:15:04miraka thing connected to the motherboard ?
19:15:07amiconnIt's a tool to command the CPU from the outside, from a PC
19:15:27mirakpaypal ?
19:15:56amiconnLinus has such a beast available
19:16:00mirakwell I am sure we could ventilate the price easily with donation system ?
19:16:16mirakand he is busy
19:18:17mirakI want rockbox features they are nice :p
19:18:57 Quit webguest93 ("CGI:IRC")
19:19:13 Join Moos [0] (
19:20:02lamedThere's a concert starting in ten minuts, is it safe to record a long file to rockbox by line in? i forgot, no mp3 real encoder yet :( nm
19:20:50 Part mirak
19:21:45 Join Lear [0] (
19:21:52crwli've successfully recorded to wav
19:27:18 Join ghode|afk [0] (
19:29:43Learpreglow: is codecs/codec.h obsolete now?
19:32:17 Join DrMoos [0] (
19:36:52preglowLear: no
19:37:09preglowLear: there needs to be done more on that part, and i haven't had the time to think about it yet
19:37:16preglowLear: however, xxx2wav.* should be obsolete
19:37:45Learbut what codec includes it?
19:37:55preglowall of them
19:38:04preglowthe libraries themselves
19:38:08preglowthe codec plugins use only codeclib.h
19:38:27preglowbut the codec libs can't do that, 'cause they'll complain about all the mem* and str* duplicates
19:39:24 Join linuxstb [0] (
19:40:13LearI see. I was wondering about why there were malloc-related defines duplicated there...
19:40:30LearI'm about to get rid of codec_alloca, since it isn't needed, after all.
19:42:41preglowhow come?
19:42:43LearHm... Does codeclib.h really need the defines for malloc etc? Shouldn't, IMHO...
19:42:51preglowLear: no, afaik, they shouldn't
19:42:58linuxstbamiconn: Are you planning a full 18-bit driver for the H300's LCD? I'm wondering if it could be configured as 16-bit, to simplify things a little and same memory for the framebuffer.
19:43:01preglowLear: if you want to fix this issue, feel free, i wont have time for several days
19:43:13LearThe reasone was a declaration of alloca() int tremor/misc.h... For some reason that caused problems for the simulator only...
19:43:56LearOkay, I'll try removing the defines then, now that I know better how the files are used.
19:48:54 Join Propri [0] (
19:53:04 Quit Moos (Read error: 110 (Connection timed out))
19:53:58amiconnlinuxstb: Why should we restrict ourselves to 16 bit when 18 bit are possible?
19:55:54amiconnIf we restrict the core to 16 bits and then a plugin wants to use the full 18 bits (e.g. jpeg.rock) we would have to switch modes
19:56:06linuxstbI'm not saying we should, I'm just asking the question. Do you know how noticable the increase of two further bits of colour give us?
19:56:54ze2 bits?
19:56:56amiconnI think the colur steps will be clearly visible even with 18bit, unless we do dithering
19:57:01ze4x as many colors?
19:57:11amiconn...and that will increase CPU load for sure
19:57:15zeor no wait
19:57:19zei dunno
19:57:23linuxstbI know the mathematics, but what will your eyes see?
19:57:27amiconnze: you're right.
19:57:35zeyour eyes will see banding on 16bit gradients
19:57:45zemaybe less banding on 18bit
19:57:58ender`linuxstb: try displaying a red gradient with 32 and 64 steps and then decide
19:58:05amiconn16 bits even cause colour banding in grey gradients
19:58:25amiconn...because the number of bits isn't equal for all pixels
19:58:41preglowuse the full 18 bits
19:58:42amiconn(or we would need to restrict even further and use 15 bits)
19:58:53amiconnender`: r5g6b5 it is
19:59:08ender`of course
19:59:13amiconnThe eye is most susceptible to green, so that's what gets most bits
19:59:14***Saving seen data "./dancer.seen"
19:59:19ender`was too quick at typing the numbers :p
19:59:32amiconnI've already seen r6g6b4 as well
19:59:47amiconn(but the lcd uses r5g6b5 in 16 bit mode)
19:59:54amiconn*h300 lcd
20:01:16amiconnThe bad thing is that I don't know the lcd hookup in the h300, so I don't know the exact format to use
20:01:39amiconnThe lcd allows serial, 8bit, 9bit, 16bit and 18bit bus width
20:02:27amiconnI don't expect serial or 18bit though
20:02:36 Nick Vlad0man is now known as Vladoman (
20:03:21amiconn16bit mode only allows 16bit colour, so I expect it's either 9bit or 8bit
20:05:59XavierGrsay: the other day I was discussing about programming with a friend, he had to choose between 2 languages, fortran or C (she doesn't know anything of programming yet) as a college lesson (Physics department). Though I don't know anything about fortran I advice her to choose C because (maybe I am mistaken) fortran is a programming fossil. Is there any reason to use fortran against C, and in which aspect fortran could be better for programm
20:07:20_FireFly_afaik fortran is mostly used in scientific ares
20:07:47XavierGryes but why it is better than C nowadays.
20:07:48_FireFly_in super computers which calculate the weather
20:07:49 Quit DrMoos (Connection timed out)
20:08:16XavierGris it more fast managing calculations?
20:08:27_FireFly_afaik it has an mathematik syntax
20:09:01_FireFly_afaik cobol is also used in these areas
20:09:39linuxstbI've never used Fortran, but I've always thought of it as a language for mathematical applications. C is a general-purpose language. It's a case of the right tool for a particular problem.
20:09:48SlasheriI would recommend C as the language to start with (and the one to study well). After that it should be very easy to study other languages as well
20:09:59preglowavoid fortran
20:10:29_FireFly_preglow: why ??
20:10:32crwljust use qbasic
20:10:50preglowon the grounds of it being an old, crappy and ugly language
20:10:54 Quit actionshrimp (Read error: 110 (Connection timed out))
20:11:29XavierGrI said to her that maybe once upon a time fortran could outmatch C in speed or something specific, but this lesson exists in this department as a fossil of older students who learnt fortran in the old days.
20:11:58SlasheriI think also that C is one of the hardest languages to begin with but it will give you a very good idea of high level languages in general
20:12:23amiconnC is very flexible and powerful, but imho not the best language to start with
20:12:24 Nick Propri is now known as Moos (
20:12:47SlasheriI think that it's best to start with the hardest one..
20:12:48zei started with batch files, then moved to bash scripts, and eventually got around to C
20:12:51linuxstbSlasheri: I disagree - C will give you an understanding of low-level languages and how computers actually do things.
20:12:54zebut i still do most stuff with bash scripts
20:13:07Slasherilinuxstb: yep, and that can't be a bad thing :)
20:13:27crwli like python because with it you can actually do something useful with about an hour of exposition to it
20:13:30linuxstbSlasheri. Exactly. But it's not the same as most high-level languages.
20:13:31SlasheriI started with C and studied later C++ and other more object oriented languages
20:13:40XavierGrI started with VB when I got the programming logic I swithced to C
20:13:55preglowi started with basic and assembler
20:14:09ender`qb->vb->delphi->c++->perl :)
20:14:19XavierGrOne time I tried to learn assemly.... my god it IS a BEAST!
20:14:24SlasheriI don't like that today most people start programming with java.. They don't need to worry anything about pointers, memory allocations etc. And then it's a full mess if you some day want to program with C
20:14:47Slasheriand qb&vb are real badness
20:15:01ender`perl uber alles >:)
20:15:06preglowender`: agreed
20:15:11Slasherithey can give the worse programming habits of using gotos everywhere..
20:15:12XavierGrSlasheri: But a handy quick solution.
20:15:13preglownicest language i know of
20:15:43SlasheriXavierGr: i doubt that.. in fact i found qb harder than C.. Fortunately, i barely used qb after all
20:16:02amiconnImho the most important thing is to understand the concepts of programming, and how computers work
20:16:12amiconnThen one language more or less doesn't matter
20:16:14*ender` once wrote a graphic editor in QB :)
20:16:32preglowamiconn: unless you start trying to use haskell...
20:16:39XavierGryes but programming a small program for Windows I think VB is better. Except if you are going for fast execution and optimization e.t.c
20:16:53_FireFly_or delphi
20:16:58ender`imho, delphi is much easier to use than VB
20:17:00Slasheriender`: me too, but it saved the graphics as .bas file that was excecuted to redraw the graphics ;)
20:17:16ender`nah, mine could work with BMP PCX and GIF files :)
20:17:22Slasherihehe :D
20:17:23amiconn_FireFly_: Afaik, delphi is the successor of turbo pascal?
20:17:28ender`amiconn: yes
20:17:46Slasheriyep, delphi language is in fact object pascal
20:17:51Slasheriit's really good with gui programming
20:18:33ender`delphi's only real downside is that VCL is huge - empty form in D6 makes a 600k exe iirc
20:19:09 Join muesli- [0] (
20:19:14XavierGrwow 600k?
20:19:35ender`yeah, because it pulls the complete GUI library in
20:20:02XavierGrmuesli-: high :P
20:20:45 Quit Philip_0729 ("bye everybody")
20:20:50zemuesli-: are you a kind of cereal?
20:21:15XavierGrkiller maybe, lol
20:21:20ender`if you make a console project, you can get under 10k (or if you use windows API for creating GUI, but that kinda defeats delphi's purpose)
20:21:36muesli-ze si ;)
20:21:57zemuesli-: i recently made my own cereal
20:22:08zekinda like muesli i think
20:22:53zewheat flakes, rolled oats, wheat germ (i'll leave that out next time i think though), walnuts, almonds, hazelnuts, dates, raisins, cinnamon, nutmeg, brown sugar, honey, and rice milk
20:23:01ze(and toasted till dry and crispy)
20:23:03 Quit fubar_ ("leaving")
20:23:18zeand i just realized i forgot to write down how much brown sugar i put in
20:24:10XavierGramiconn: ConTEXT editor is a beauty indeed! Have you used notepad2 (though this is a notepad replacement rather than programming editor (with limited syntax highlighting))
20:24:57_FireFly_metapad is also a good notepad replacement :)
20:25:13dpassen1ive always liked ultraedit for coding
20:25:27*ender` uses a console-based file manager for about 80% of the things he does with computer
20:25:41XavierGrI will have to disagree on Ultra Edit
20:25:50amiconnNo. Before discovering ConTEXT I used phase5. But phase5 is tailored for HTML editing, and has resource problems if kept open for extended times
20:25:53_FireFly_ender`: mc ?
20:26:00amiconn(and isn't developed any more afaik)
20:26:04ender`on linux, yes, but on windows i use FAR
20:26:40XavierGrdpassen1: I remember it was a little heavy...
20:26:58dpassen1still is, i recently tried UEStudio, its gotten heavier
20:27:03zei dunno what makes muesli muesli though
20:27:11zebbl, gotta get ready for school
20:27:12dpassen1ill give ConTEXT a try
20:27:36amiconndpassen1: I didn't try UltraEdit, mainly for one reason: it costs money
20:27:38muesli-ze i am unique ^^
20:27:43amiconnConTEXT is free
20:27:53amiconn(not as in speech, but as in beer ;) )
20:27:56dpassen1i must try it then
20:29:09XavierGrwhat's the output console on ConTEXT for?
20:30:13muesli-btw benchmarked ~25h of runtime
20:30:23dpassen1can you fold and expand functions in ConTEXT?
20:30:38XavierGrthen less battery time?
20:30:55XavierGrlink for test results?
20:31:09muesli-XavierGr just a second
20:34:27muesli-here we go
20:35:21XavierGrso it died on the 25th hour?
20:35:50XavierGror it continues to work?
20:35:51muesli-something like this
20:36:08XavierGrfirst or second?
20:36:09muesli-no, it shut down after 25hours
20:36:33muesli-<XavierGr> so it died on the 25th hour? ->yes
20:36:44XavierGrhmmm that is 4 whole hours, where those went?
20:36:58XavierGrI couldn't expect that
20:37:03_FireFly_on the high bass boost
20:37:17_FireFly_or had you the same bass boost on iriver fw ??
20:37:39XavierGrI think it could be better to run the test with flat eq.
20:37:42_FireFly_was denn nu ?? :)
20:37:46muesli-i had to set the bass @24 to get same results
20:37:58muesli-XavierGr true
20:38:13muesli-but those lab tests suck
20:38:29muesli-i'd loved to get a useful result
20:38:36XavierGrI never use eq! :)
20:38:55_FireFly_"robin hood helden in strumpfhosen" is currently on tv :)
20:39:30muesli-am just looking who wants to be a millionaire
20:41:23 Quit phaedrus961 ("Leaving")
20:44:19 Join actionshrimp [0] (
20:44:44 Join phaedrus961 [0] (
20:54:43lamedwhat's a good program to split mp3 file and having it gaplessed?
21:16:14 Join webguest92 [0] (
21:16:22muesli-_FireFly_ "behind bars" @rtl :D
21:17:06_FireFly_better numb3rs
21:18:04 Quit webguest92 (Client Quit)
21:18:06muesli-uuuh l33t stuff ;)
21:18:21_FireFly_so is the original title :)
21:18:48muesli-but prefer report @ard
21:28:27Slasheribtw, the musepack sounds clearly superiour to vorbis/mp3.. maybe i should start encoding with that
21:29:08crwli have yet to ABX -q4 ogg vorbis
21:29:11dpassen1musepack is superior @ 160 kbps and up
21:29:15Slasherihehe, yes. Drums etc. are very clear
21:29:24crwlthough, i haven't tried with my "new" headphones yet
21:30:54crwlSlasheri, at what bitrates is that?
21:31:30dpassen1Slasheri: if that will inspire you to optimize libmusepack, maybe you should ;-)
21:31:55Slashericrwl: Hmm, something over 200. But i need to do more testing..
21:32:05crwlSlasheri, well, i'm quite sceptical on that
21:32:19SlasheriI think amiconn already has great ideas doing that
21:32:45dpassen1crwl: musepack −−standard is comparable to lame mp3 −−aps and ogg -q6 (i believe) but yields a lower bitrate
21:32:51crwli once thought i could hear all kinds of differences with even highish-bitrate files (>160 kbps VBR mp3's/oggs), but when i tried ABXing them, i couldn't
21:33:53dpassen1theres only one song ive ever had, of all my −−aps mp3s where i could hear something that sounded weird and then could abx regularly
21:34:15crwldpassen1, yes... but most people (i guess over 99%) can't hear differences between ~200 kbps files and the original with any modern encoder, regardless of the format
21:34:26crwlof course i'm not claiming that somebody can't
21:34:34dpassen1i think i support that, only one song out of nearly 2500
21:35:40crwlsomebody could ABX the gaps between mp3 files, though :P
21:36:39crwlanyway, i always encode to FLAC now
21:36:47crwland transcode to -q4 oggs to my iriver
21:36:57crwland i think that's more than enough when i'm on the road
21:37:11crwl...because last time i tried, i couldn't reliably ABX those from the original
21:37:15crwli should try again though :)
21:37:25dpassen1when i get a new HDD, im thinking of reripping everything to WavPack Hybrid, putting the .wv files on my H120, and when i listen on main system, .wv plus .wvc
21:38:33crwli'm not really delighted of musepack, monkey's audio or wavpack... mostly because of the lacking support on any other platform besides win32
21:38:39crwlthough wavpack is somewhat better, i think?
21:38:52crwli haven't seen a single file in that format yet, though :)
21:39:02dpassen1i know wavpack isn't as tied as monkey's audio
21:39:15crwlyes, they seem to have some player plugins at least
21:39:24crwland at least a linux decoder
21:40:14crwlthat's good
21:40:33crwlit's always possible to convert lossless files
21:41:18Slasherimusepack and wavpack works great on linux at least
21:41:23crwli ran to some .ape files the other day... and couldn't even decode them
21:41:31crwlSlasheri, do they have gstreamer/xine plugins to use with amaroK?
21:41:45Slashericrwl: yes, amarok plays musepack great
21:41:49Slasherihaven't tried wavpack
21:41:54crwli found a linux ape tool from frank klemm's site, but it just gave some odd error
21:42:09crwlwell, wavpacks are convertable to flac so that's no big problem
21:42:17Slasheribut now sleep :) ->
21:42:20crwlbye :)
21:43:19_FireFly_curl i use this apev2 tag tool
21:44:21_FireFly_by the way the wavpack encoder has under linux the possibility to tag the files self
21:44:25solexxis there any mass tagging tool with flac supprt that doesn't suck?
21:44:32solexxpreferably for linux
21:44:48solexx_FireFly_: using that right now
21:44:50thegeek_foobar2000 <3
21:44:59 Join JoeBorn [0] (
21:44:59solexxit doesn't handle umlauts very well
21:45:08_FireFly_which version ??
21:45:26crwli haven't had umlaut problems with easytag, but crashing problems... yes
21:45:57solexxmaybe my umlauts are broken. don't even know whether they are in iso ur utf
21:46:12solexxis abcde known to be broken in this regard?
21:46:16crwlno idea
21:46:31solexxi am using it for ripping my whole cd ccollection to flac atm
21:46:34crwljack was my favourite text mode ripping tool
21:47:11solexxhm, doesn't do flac
21:47:14crwlbut for some reason i stopped using it, and used grip for a while... which had some really strange utf-8 problems and crasing stuff at some point, so i switched to kaudiocreator which just seems to do the job and shut up
21:47:18crwldoesn't it?
21:47:36solexxnot according to debian's package description
21:47:44crwlwell, according to it does
21:48:13solexxdamnit. thanks
21:48:17crwli can't remember, though. maybe i only encoded to vorbis when i used it...
21:48:53solexxi found abcde pretty cool when i re-discovered it some days ago
21:49:06crwldoes it encode and rip at the same time already?
21:49:07 Quit _FireFly_ ("Leaving")
21:49:12solexxbut you have to go through shell-hell when you want to influence file-naming
21:49:23solexxcrwl: abcde? it does,.
21:49:41crwlsolexx, hmm, maybe i just remember wrong or then it didn't do that some years ago :)
21:51:30solexxthis whole undertaking is a little exhausting because weird characters screw up every other tool i am using
21:51:57solexxstrangely, some iso-8859-1 characters make it through to the filename *and* the tag and some don't
21:52:20solexxi should have replaced all of them in the first place
21:52:23crwlluckily i haven't run to anything like that :x
21:52:36solexxor have my system switched to utf8 completely
21:54:14solexxbut then again, file names could still be broken on their way to the fat filesystem of my iriver...
21:55:19solexxcrwl: is there any way to tell easytag to use only lowercase characters for filenames?
21:58:06crwlsolexx, no, i'm not sure there is...
21:59:17***Saving seen data "./dancer.seen"
21:59:57solexxthanks. will use qmv then
22:00:36*solexx needs a new harddisk, or two
22:02:13 Join Philip_0729 [0] (
22:12:24 Join arkascha [0] (
22:18:03 Quit Lear ("Chatzilla [Firefox 1.4.1/undefined]")
22:27:07 Quit Philip_0729 ("bye everybody")
22:27:56 Quit arkascha ("Konversation terminated!")
22:31:29 Quit linuxstb ("CGI:IRC")
22:35:29lamedis firefly remote patch on the cvs? where?
22:36:11t0masnot afaik
22:37:18lamedah. off of that, i think "load firmware file" should remamber what was your last move. don't you agree?
22:38:53lamedquote Re: Remote lcd support(testing)
22:38:56lamed the fix was commited to cvs
22:39:01lamednd i have only updated my copy through cvs and rebuild it.
22:39:12lamedunquote firefly
22:45:55 Quit lamed ("CGI:IRC (EOF)")
22:53:31 Join pike [0] (
23:00:40 Join muesli__ [0] (
23:10:59 Quit dpassen1 ()
23:19:09 Quit muesli- (Read error: 110 (Connection timed out))
23:20:59 Quit ender` (Read error: 104 (Connection reset by peer))
23:33:27 Join InnerPhalanx [0] (
23:39:57 Quit muesli__ (Read error: 113 (No route to host))

Previous day | Next day