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 2004-07-20

00:05:48midkamiconn: rombox?
00:06:18midkah, running entirely from rom?
00:06:25amiconnYup. Rockbox running directly from flash rom, leaving more ram for the mp3 buffer
00:07:17***Saving seen data "./dancer.seen"
00:08:59midklots more? or a bit more?
00:09:32midkbbl all
00:09:33 Quit midk ("just STOP it arspy")
00:32:15 Join BC|coding [0] (
00:32:15 Quit Nibbler (Read error: 104 (Connection reset by peer))
00:54:40 Join tpelliott [0] (
01:09:40 Join amiconn_ [0] (
01:09:54 Part tpelliott
01:10:11 Quit amiconn (Nick collision from services.)
01:10:11 Nick amiconn_ is now known as amiconn (
01:25:57amiconnNite all
01:26:12 Part amiconn
01:29:00 Part BC|coding
01:33:00 Quit lImbus (" HydraIRC -> <- The dawn of a new age")
01:34:17kaboofahar har
01:34:24kaboofairssi is a dawn of a new age
01:40:21 Join gromit`` [0] (
02:07:18***Saving seen data "./dancer.seen"
02:07:57 Quit mecraw__ ("Trillian (")
02:10:12 Join Nibbler [0] (
02:48:49 Join scott666_ [0] (
02:48:49 Quit scott666 (Read error: 54 (Connection reset by peer))
02:55:53 Quit AciD (Read error: 54 (Connection reset by peer))
03:28:56 Quit Nibbler (Read error: 104 (Connection reset by peer))
03:53:50 Join dstar5 [0] (
04:07:19***Saving seen data "./dancer.seen"
04:20:28 Quit scott666_ ("i'll be back...eventually...")
04:30:55 Join Strath [0] (
04:38:47 Join scott666_ [0] (
04:39:54 Quit Ka_ (
04:39:54 Quit elinenbe (
04:41:14NJoinKa_ [0] (
04:41:14NJoinelinenbe [0] (
04:55:12 Join AciD [0] (
05:01:06 Quit dstar5 ("Leaving")
05:11:57 Join Nibbler [0] (
05:17:39 Part Strath
05:18:06 Quit Ka_ (
05:18:06 Quit elinenbe (
05:26:27NJoinKa_ [0] (
05:26:27NJoinelinenbe [0] (
05:39:32 Quit AciD ("resistance is futile. you will be ass-stimulated")
06:07:21***Saving seen data "./dancer.seen"
06:18:42 Quit Nibbler (Read error: 104 (Connection reset by peer))
06:31:28 Join midk2k3 [0] (
06:31:46 Nick midk2k3 is now known as midk (
06:43:12 Join LinusN [200] (
06:45:21 Quit scott666_ ("i'll be back...eventually...")
07:07:54 Quit midk ("Going... BYE!")
07:08:58 Join midk [0] (
07:50:13 Join Nibbler [0] (
08:00:17 Join Naked [0] (
08:04:18 Quit Hadaka (Read error: 111 (Connection refused))
08:04:19 Nick Naked is now known as Hadaka (
08:07:22***Saving seen data "./dancer.seen"
08:24:14 Quit Nibbler (Read error: 104 (Connection reset by peer))
08:40:26 Join midk2k3 [0] (
08:41:51 Quit midk (Read error: 54 (Connection reset by peer))
08:56:29 Join W_Sherman [0] (
08:58:08 Quit W_Sherman (Client Quit)
09:02:31 Nick midk2k3 is now known as midk (
09:12:09 Join amiconn [0] (
09:12:44midkhi amiconn
09:12:44amiconnhi all
09:27:43amiconnLinusN: While I const'ed the icons, I wondered about the variable slider_bar[]. This seems to be used nowhere, however the compiler doesn't warn about an unused variable.
09:30:56 Join Strath [0] (
09:31:07Strathhey guys
09:31:44 Join [IDC]Dragon [0] (
09:31:58*midk points at strath, yelling, "HE PETS ME IN WRONG WAYS!"
09:32:03StrathI uploaded gromit's pictures of his disassembled gmini 220
09:32:43Strathi didn't "pet" you... i did "pat" you though :P
09:32:53midkok, he pats me in wrong ways then
09:33:02midkand.. in wrong.. directions.
09:33:26Strathlike one would a puppy when it begs for a treat
09:33:53midkformat my hard drives, ma'am?
09:33:53amiconn[IDC]Dragon: Your flash id function change works, however it seems to me that calling "View HW info" is slower now?
09:34:04[IDC]Dragonhi there
09:34:31[IDC]Dragonmaybe because of the memcmp()
09:34:58[IDC]Dragonit compares the boot ROM region with the flash
09:35:17[IDC]Dragonto check if they are the same, which means no boot ROM
09:35:53amiconnAnd running from rom itself makes it slower?
09:36:29[IDC]Dragonthe flash has 8 bit bus, versud the DRAM has 16
09:37:05[IDC]Dragonon the other hand it has no CAS delay or whatever
09:37:22amiconnIt seems that some more functions could be moved to iram
09:37:56amiconnI'd just like to know how much space should be left free for the debug builds
09:39:12[IDC]DragonStrath: this "half" HD looks funny
09:39:53Strathi know...
09:40:16Strathit's like sidways or something
09:40:17midkthat'd be a hitachi?
09:40:25midkused in the karma..
09:41:47amiconn[IDC]Dragon: Moving the string handling functions (those that are actually used) to iram should speed up operation from RAM too. For ROM operation moving anything that runs for some time would get faster in IRAM.
09:42:18[IDC]Dragonamiconn: yes
09:42:46[IDC]DragonI'd like to know what the flash ROM execution penalty really is...
09:43:47amiconnI have a funny battery effect: The battery set I intended to use for the runtime tests (1200 mAh Varta) doesn't work in the recorder - it looks as they are empty, even if they are fully charged.
09:43:48[IDC]Dragonamiconn: how did you handle the .lds file?
09:44:10amiconnHowever, they work perfectly in other devices
09:44:15[IDC]DragonI just tried, it gets overwritten
09:44:28[IDC]DragonI think this wasn't the case back in time
09:44:42[IDC]Dragonso you changed the template?
09:45:09amiconnI put the into the build dir, with its original name. This way it doesn't get deleted when I do "make clean"
09:45:19amiconnThen I use the following script for building:
09:45:49amiconnrm ajbrec.ajz
09:45:49DBUGEnqueued KICK amiconn
09:45:49amiconnuclpack >/dev/null −−none rockbox.bin rockbox.ucl
09:46:27[IDC]Dragonthe make restores my original .lds
09:46:48amiconnThe cp command sets the time stamp of the destination file to the current time and date, so it doesn't get overwritten by the following make
09:46:51[IDC]Dragonand if I make it write only to stop this, the build fails
09:47:23amiconnmake decides whether to regenerate a file based on the time stamp
09:47:44[IDC]DragonI know (or, I should)
09:47:48amiconnBefore I made the script I used touch every time
09:47:58[IDC]Dragonwasn't thinking about the file date
09:48:59[IDC]Dragonthe map file tells us names only for the exported symbols
09:49:19[IDC]Dragonso the size for menus, etc. remains unknown
09:49:37[IDC]Dragonwe only know the .data consumption of the whole module
09:50:19[IDC]Dragonwell, a very few KB to be gained, maybe
09:51:10[IDC]Dragonthe plugin API struct is not onst, yet
09:51:36[IDC]DragonI'd like to use that pointer to check within the flash plugins
09:51:49[IDC]Dragonand bail out if we're running from ROM
09:52:14Strathok... time for bed, g'night guys
09:52:20midkpat pat
09:52:22midkgood boy!
09:52:24amiconnThe default lang string do already go into .rodata; the array itself has to stay in .data because it gets overwritten when you load a .lng file
09:52:28 Nick midk is now known as midk|quakeIII (
09:52:31 Quit Strath ("Client closed")
09:52:46[IDC]Dragonthe strings stay?
09:54:02[IDC]Dragonin ROM, they have to, inorder to have init values
09:54:27[IDC]Dragonbut in RAM, i'd expect them to make room for the new language
09:54:59[IDC]Dragonmaybe this can be improved
09:55:56amiconnThe loaded strings go into another ram area.
09:56:20[IDC]Dragonwould be better to use the same
09:56:38[IDC]Dragonfor normal RAM execution, this would save space
09:56:59amiconnYes, while it won't improve when executing from rom
09:56:59[IDC]Dragonfor ROM, it doesn't make a difference
09:57:44[IDC]DragonI need to get to work
09:57:54[IDC]Dragoncu you in 30 min
09:58:03 Quit [IDC]Dragon ()
10:05:01 Join Nibbler [0] (
10:07:26***Saving seen data "./dancer.seen"
10:08:39 Quit midk|quakeIII (Read error: 104 (Connection reset by peer))
10:20:34 Join midk [0] (
10:24:00 Quit midk (Read error: 104 (Connection reset by peer))
10:25:30 Join [IDC]Dragon [0] (
10:25:44amiconnre [IDC]Dragon
10:28:27[IDC]Dragonhi again
10:29:18 Join midk [0] (
10:29:29amiconnIf we want to put the default lang string into the memory area that is used for the loaded strings as well, we must somehow force the compiler to put the strings there.
10:29:43[IDC]Dragonyes, define a region
10:30:20 Nick midk is now known as midk|quakeIII (
10:31:01amiconnThe size of this region has to be adjustable independent of the size of the default string, because the loaded strings may be larger (and this is the case with deutsch.lang)
10:31:17amiconndeutsch.lng even
10:31:54[IDC]Dragonit needs to have the size we currently use for the lng loading, yes
10:32:08LinusNi'm a little bit rusty on the lang system, but aren't the english default strings supposed to be used when the string isn't defined in the .lng file?
10:32:40[IDC]Dragonoops, yes
10:34:00amiconnLinusN: Did you get my question concerning slider_bar[] ? It seems that it is no longer used...
10:39:15LinusNquestion about slider_bar?
10:39:51LinusNah, in icons.c?
10:40:17LinusNkill it
10:41:14amiconnWhere did it belong to? It seems that it wasn't used even before you killed some of the various bars
10:41:29LinusNvery old, obsolete stuff
10:42:08 Join midk [0] (
10:45:34amiconnKilled. Still compiles without warnings, saves some bytes. I can't test on the target atm, because a runtime test is running
10:53:29 Join Guest1 [0] (~jirc@
11:07:48 Quit midk|quakeIII (Read error: 110 (Connection timed out))
11:08:01 Quit Guest1 ("Leaving")
11:14:44LinusNamiconn: bad news!
11:17:16amiconnTell me
11:20:44LinusNthis could also be related to the ATA optimizations:
11:23:09amiconnHmm. The latter report does not state the exact version that causes the problems, and also not the disk type.
11:23:37LinusNno, i'm just guessing, i'll send him the same test version as well
11:23:49amiconnI must say i hate those %&$§ Hitachi disks ;(
11:25:47amiconnCould you prepare additional test versions with only optimized reading/ writing disabled?
11:27:04amiconnIf we decide to simply disable the optimizations for Hitachi DK23CA & DK23DA this would impact users whose hw combination doesn't exhibit problems with it
11:27:47amiconnAdditionally, I'd like to see the other disk info (Firmware version, access times etc) and the hw mask of the problematic boxes
11:29:40amiconnI wonder if this is a 0x0102 masked recorder...
11:36:31 Nick midk is now known as midk|sleepy-ing (
12:05:02 Quit Nibbler (Read error: 104 (Connection reset by peer))
12:07:29***Saving seen data "./dancer.seen"
12:18:15 Quit midknight2k3 (Read error: 104 (Connection reset by peer))
12:18:22 Join midk2k3 [0] (
12:24:43amiconn[IDC]Dragon: In settings.c, it looks like rtc_bits[] and hd_bits[] could also be declared const, or are they written to somewhere? If I am correct, these tables only define the layout, defaults and names of the settings
12:50:10[IDC]Dragonamiconn: yes, I mentioned that yesterday
12:50:26[IDC]Dragon(just got back from lunch)
12:51:09 Join Zxcvb [0] (
12:53:57amiconnThe mp3 genres are already declared const
12:54:26[IDC]Dragonyes, I saw and mentioned that
12:54:43[IDC]Dragon(don't you read what others type? ;-)
12:55:15Zxcvbcould rockbox support the flac format?
12:55:44amiconn[IDC]Dragon: For the latter one I seem to have missed that.
12:56:02[IDC]Dragonamiconn: no pb
12:57:02amiconnFor the former, if you refer to "13.37.38 # <[IDC]Dragon> the settings tables", this wasn't precise enough for my limited imagination ;-)
12:57:38amiconnZxcvb: No, see
12:57:52[IDC]DragonI'll be more descriptive next time
12:59:00[IDC]Dragonare you currently const-policing?
13:00:44amiconnNot much atm. I want to have at least some preliminary runtime results before tackling the more complex cases.
13:01:23amiconnPerhaps the runtime doesn't increase at all, in case reading the rom all the time makes it draw more power
13:01:50amiconn(Complex cases mean e.g. sysfont)
13:01:57[IDC]Dragonif you use old cells, their runtime may get better with each charging cycle
13:02:09Zxcvbso the DSP would have to be figured out?
13:03:01amiconn[IDC]Dragon: My old 1200 mAh cells don't seem to work for the recorder (as reported earlier), so I have to use the default 1500 mAh ones. These are still in good shape
13:03:40LinusNZxcvb: yes
13:03:52[IDC]DragonI meant, if you haven't used them for a while, they may improve with "training"
13:04:17[IDC]Dragonso better test like A,B,A
13:04:44Zxcvbsounds hard without any docs
13:04:47[IDC]Dragonand see how much the two A results differ, before trusting B
13:04:52LinusNZxcvb: we have programming docs for the MAS3507D, but it's a bitch to program, and the memory is very small in the DSP
13:05:27Zxcvbso which formats could the DSP be used to handle?
13:05:44LinusNhard to tell
13:05:59amiconn[IDC]Dragon: I didn't put them away for too long, only about a month. Of course I'll do at least two complete runs for each rom and ram based, probably best in the order ABBA
13:06:01LinusNin theory, we could handle anything, but there are other limitations
13:06:20LinusNthe memory size, for instance, rules out OGG
13:06:33[IDC]Dragoni liked ABBA, as a teenager
13:06:58LinusNthe slow serial connection to the MAS rules out >20kHz WAV
13:07:08ZxcvbLinusN: flac?
13:07:15LinusNi dunno
13:07:23amiconnLinusN: Some time ago I had a look at the MAS3587F programming docs. Although the command set is described there, it gives no info about accessing the ports :(
13:07:48LinusNamiconn: where did you get them from?
13:08:46LinusNi have only the 3507 docs
13:10:09amiconnErr, I confused them. Still, in the MAS3507D docs available, port access isn't described.
13:11:35Zxcvbvideo apparently works ok
13:11:43amiconnIf we are able to get hold of the MAS3587F docs, wav may be possible in full quality if the sh1 serial is able to handle it
13:12:19LinusNZxcvb: the video isn't done by the DSP
13:12:37LinusNamiconn: but the serial cable isn't fast enough
13:12:38Zxcvbbetter than videonow from what I have seem
13:12:49[IDC]Dragonnot even by the SH, to be precise
13:13:09[IDC]Dragonit's pre-rendered on a PC
13:13:42Zxcvbisn't that what videonow does, though it just uses pressed discs?
13:14:06[IDC]DragonI don't know about that
13:14:41CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
13:14:41*[IDC]Dragon browses
13:15:18amiconnThe MAS' serial can go up to 6.144 Mbits/s, so this is sufficient for 44.1 kHz 16 bit stereo uncompressed wav (~1.76 Mbits/s including start&stop)
13:15:45Zxcvbvideonow is a 80x80 b&w video player intended for kids
13:18:08amiconnsh1 serial can go up to 2 Mbits/s for 12 MHz clock, or ~1.84 Mbits/s for 11 MHz
13:18:56amiconnSo 48 kHz / 16 bit wouldn't be possible on the recorder
13:26:21amiconnIt seems that only the MAS3587F can handle such a high serial bitrate, the MAS3507D only handles 1 Mbits/s
13:37:20 Join Nibbler [0] (
13:42:40 Quit midk|sleepy-ing (Read error: 104 (Connection reset by peer))
13:56:41 Join _lImbus [0] (
13:57:04 Nick _lImbus is now known as lImbus (
13:57:06lImbushi all
14:06:22 Join AciD [0] (
14:07:30***Saving seen data "./dancer.seen"
14:15:03 Join ka__ [0] (~tkirk@
14:44:50 Quit Nibbler (Read error: 104 (Connection reset by peer))
15:44:50 Quit lImbus (Remote closed the connection)
15:47:40 Part LinusN
16:00:17 Quit Zxcvb ("Leaving")
16:00:36 Join hernest [0] (
16:03:05 Join lImbus [0] (
16:04:46 Join stripwax [0] (~slowdownc@
16:04:55stripwaxGood afternoon
16:07:32***Saving seen data "./dancer.seen"
16:29:12 Join Nibbler [0] (
16:41:27 Quit hernest (Client Quit)
16:52:12 Part stripwax
16:52:47 Join dstar5 [0] (
17:10:42 Quit Nibbler (Read error: 104 (Connection reset by peer))
17:17:28dstar5my machine is no where close to eing able to run doom 3 :(
17:26:46 Quit [IDC]Dragon ("no fate but what we make")
18:07:35***Saving seen data "./dancer.seen"
18:17:39 Part lImbus
18:22:43kaboofawho needs doom 3 when you have history's mysteries
18:22:58kaboofaHistory channel = <3
18:36:40 Join zeekoe [0] (
18:53:29 Join Nibbler [0] (
19:09:27 Quit Nibbler (Read error: 104 (Connection reset by peer))
19:18:00 Quit zeekoe (Read error: 60 (Operation timed out))
19:24:14 Join mecraw__ [0] (~lmarlow@
19:29:37 Quit AciD (Connection timed out)
19:35:56 Join midk [0] (
19:36:38 Join zeekoe [0] (
19:38:41zeekoewhy are you logged in twice?
19:39:42midki'm special
19:40:29zeekoei understand
19:40:42zeekoebtw, anyone tried ROMbox yet?
19:40:53zeekoei may wanna try it
19:41:26midki wasn't aware it was available yet
19:41:49zeekoeit is, at the mailinglist
19:41:51amiconnzeekoe: :)
19:41:56zeekoejust vor recorder v1
19:42:20zeekoeamiconn: do you have a v1 recorder?
19:43:42amiconnOf course. How else would I be able to test rombox?
19:49:39zeekoeso you are testing rombox?
19:49:46*zeekoe keeps quiet
19:50:16amiconnI'm currently doing runtime tests. See and who posted about rombox...
19:50:16midkamiconn created it/
19:50:28midkamiconn, didn't you create rombox?
19:50:50zeekoeyeah, i already knew you were jens arnold
19:51:07zeekoei once said "i like the jensarnold one" about the WPS things
19:51:19zeekoeand then you said, "hey, that's me" or something like that
19:51:21midkHA HAHA
19:51:26midkand i said, that's amiconn
19:51:28midkno i did
19:51:30amiconnmidk: I didn't create it on my own, actually the idea and much work came from Jörg aka. [IDC]Dragon. I "merely" found a stubborn bug and optimized it a bit - see my latest commits
19:51:33zeekoeyou said
19:52:06midkah, i thought it was your idea - i was wrong then
19:53:18*zeekoe having to watch LOTR3
19:53:25zeekoe(again :-P)
19:53:28zeekoecya l8er
19:53:28midkoh, what TORTURE
19:53:33midklater zeekoe.
19:53:36zeekoeit's a good movie
19:53:36amiconnzeekoe: I have that as rvf..
19:53:44zeekoeyou're kidding
19:53:53zeekoewell, bye :)
19:54:56midkbye sir.
20:07:39***Saving seen data "./dancer.seen"
20:10:48kaboofawork in 2 hrs
20:13:09 Quit zeekoe (Read error: 110 (Connection timed out))
20:15:10dstar5me here :)
20:15:17midkME TO :D
20:16:01dstar5rombox? what?
20:16:13midkum, rockbox running in ROM
20:16:41dstar5does it not already do that..?
20:17:04*dstar5 checks email
20:17:10midkrunning in from
20:17:13midknot running out of
20:17:21dstar5is it slower?
20:17:23midkmore mp3 buffer room.
20:17:28midki'm not sure
20:22:15 Join scott666_ [0] (
20:23:33 Join Strath [0] (
20:25:07scott666_and the spelling award goes to...diddy
20:25:11 Nick scott666_ is now known as scott666 (
20:28:27midkcome here, come here! that's it! sit! GOOOOD BOYY!!
20:28:38*midk pats strath on the head
20:28:43midkthe head, see... :]
20:28:54kaboofathat's kinky!
20:28:58midkwtf, stfu
20:29:11*midk pats strath's fur
20:29:22kaboofadamn 'furry' people :P
20:29:26*kaboofa walks off
20:29:33midkgoing so soon?
20:29:36*kaboofa means, by walking off, back to code
20:29:43kaboofabecause code > human interaction.
20:29:53kaboofabut i have to depart for work in an hour :(
20:30:01*midk pats kaboofa's little "mouth"
20:30:06midkok sounds fun
20:30:09kaboofaAnd listening to Skazi
20:30:22midkopen wider
20:30:33kaboofalast time someone told me that 'bad things' happened.
20:30:50midkit's just your mouth
20:30:54midkopen up
20:30:58midki'll give you a treat
20:31:06midkwatch the teeth, too.">
20:31:12kaboofathere you go
20:31:29kaboofawhen you bring a 0 upto an 80 in an AP CompSci class.
20:31:36kaboofaTHAT IS AS WIDE AS IT GOES!
20:31:47midkit won't fit. it's too big.
20:31:49kaboofa<@Matthew−−> actually, writing a nes emu in .net would be great.. no one could run it
20:31:53kaboofa<@Sloat> matt, they already have those in real C++ though
20:32:08midkdon't you want a treat?
20:32:29kaboofaif by 'treat' you mean 'poison candy' then yes, i want a treat
20:32:39midkhm, close!
20:32:53kaboofawhat's up?
20:32:57midkyou could define it as poison candy i guess
20:33:04midki was just messing with my dog treat..
20:33:07kaboofai don't even want to kn..
20:33:08midkhaha NO
20:33:10midki was not.
20:33:17midki don't even want to knok.
20:33:31kaboofatime for making macarono asdfjashkldfj asdflkjh
20:33:41midkgo put some good ingredients in
20:33:42kaboofai hit some weird key combo and it registered as the 'break' key.
20:33:48kaboofaand i had to type fg
20:34:15kaboofai don't exactly know what the name of fg is
20:34:23kaboofaall i know is that it brins the program back when you hit break
20:34:29kaboofa[1]+ Stopped irssi
20:34:36kaboofaBOOM! irssi is back
20:34:39kaboofaand i continue talking on irc.
20:34:58midkso go make the food
20:38:44kaboofaI AM!
20:38:57midkam what
20:54:12 Join Nibbler [0] (
21:00:17 Quit mecraw__ (Read error: 54 (Connection reset by peer))
21:00:37 Join mecraw__ [0] (~lmarlow@
21:01:40*dstar5 wonders why bash and uname.exe would ever need internet acess...
21:01:56 Quit dstar5 ("Leaving")
21:12:03 Join AciD [0] (
21:16:31 Join midknight2k3 [0] (
21:18:10 Quit midk (Nick collision from services.)
21:18:12 Nick midknight2k3 is now known as midk (
21:24:51 Join midknight2k3 [0] (
21:26:14kaboofai got a 4 on the AP exam :D
21:26:20kaboofaout of 5
21:26:34kaboofaand i got a 5 in Java
21:26:36kaboofakick ass.
21:28:38midknight2k3gj kaboofer.
21:29:09kaboofa4 in calc
21:29:18 Nick kaboofa is now known as BoofaWork (
21:29:20midknight2k3you mean "woo"
21:29:24midknight2k3ooh, where.
21:29:34BoofaWorkgrocery chain
21:29:58midknight2k3GO GET EM
21:37:10 Join Smooth [0] (
21:39:42 Quit midk (Read error: 110 (Connection timed out))
21:40:13 Nick midknight2k3 is now known as midk (
21:45:09 Join flemse [0] (
21:48:33 Join dstar5 [0] (
22:01:14 Quit ka__ ("Leaving")
22:07:42***Saving seen data "./dancer.seen"
22:09:14 Join StrathAFK [0] (
22:14:49 Quit Smooth (Read error: 104 (Connection reset by peer))
22:24:22 Quit Strath (Read error: 110 (Connection timed out))
22:27:04 Quit flemse ()
22:37:34 Join maikeul [0] (
22:43:26 Quit dstar5 ("Leaving")
22:43:49 Join dstar5 [0] (
22:44:04 Nick dstar5 is now known as dids (
22:44:21 Quit dids (Client Quit)
22:44:40 Join dstar5 [0] (
22:44:48 Nick dstar5 is now known as dids (
22:46:35 Join [IDC]Dragon [0] (
22:46:52amiconnhi [IDC]Dragon
22:47:53amiconnAs my first runtime round is now over and the batteries have to be charged anyway, I did some more experimentation with declaring things const
22:48:31amiconnI wonder why you said that the plugin api struct is not really const - as far as I can see, it is changed nowhere
22:49:18[IDC]Dragonno, I think it is const
22:49:29[IDC]Dragondid I say different?
22:49:58 Join da4walker [0] (
22:50:20da4walkerhi folks
22:50:23da4walkeranyone here?
22:51:02 Quit midk ("just STOP it arspy")
22:51:25da4walkerI am searching a jukebox for my car
22:51:35da4walkerI am a totally newb concerning that thing
22:51:53amiconn09.51.10 # <[IDC]Dragon> the plugin API struct is not onst, yet
22:51:53amiconn09.51.36 # <[IDC]Dragon> I'd like to use that pointer to check within the flash plugins
22:52:02amiconnMaybe I got this wrong...
22:52:14da4walkerat first I want to let the original car radio in and connect the cd line with a jukebox
22:52:22da4walkerbut is this possible?
22:52:27[IDC]Dragonit is not declared as const yet, I meant
22:52:45da4walkerdo I need a lot of things or just the jukebox and a simple cable?
22:52:46[IDC]Dragonbut should be
22:53:02[IDC]Dragonda4walker: depends on your stereo
22:53:23da4walkerhmm, Ok, what must the stereo have so that it works?
22:53:41[IDC]Dragona line in
22:53:45amiconn[IDC]Dragon: Ah ok. I already checked this - this requires to touch _every_ plugin, as the declaration of plugins of plugin_start has to be changed
22:54:08da4walkerok, the stereo has a button for cd, so I can put a cd changer somewhere in car
22:54:18da4walkerand control it with the car stereo
22:54:27[IDC]Dragonamiconn: too bad, many peopke have private plugins
22:54:29da4walkerso it must have an line in or not?
22:55:05[IDC]Dragonda4walker: the radio must be willing to allow the input to be used without a changer
22:55:14[IDC]Dragonwhat brand is it?
22:55:42da4walkerit's the original car radio which was in the VW Passat Variant (newer one)
22:55:58amiconn[IDC]Dragon: The change isn't difficult
22:56:02amiconnenum plugin_status plugin_start(struct plugin_api* api, void* parameter)
22:56:07amiconnhas to be changed into
22:56:07da4walkerand I dunno want to change it cause its blue colour in night is fantastic :)
22:56:13amiconnenum plugin_status plugin_start(const struct plugin_api* api, void* parameter)
22:56:20 Quit gromit`` (Read error: 110 (Connection timed out))
22:56:32didsda4walker: those do look cool :)
22:56:46amiconnOtherwise the plugin wouldn't compile
22:56:51da4walkerOk, I read on the rockbox page that I can also control via remote control, is this right?
22:57:01[IDC]Dragonamiconn: can't we cast it in the call?
22:57:45[IDC]Dragonda4walker: there is some effort going on to connect VW stereos via a little microcontroller
22:58:03[IDC]Dragonsee the VWCDPIC project
22:58:33da4walkerhmm, what u mean by microcontroler? a hardware part which has to be between jukebox and VW stereo?
22:58:42 Join wake [0] (~wake@
22:58:52da4walkerwill look
22:59:01amiconn[IDC]Dragon: Of course we could cast within the call, but this is ugly - see the open() problem
22:59:24[IDC]Dragonwell, but nobody is gonna modify the struct
22:59:47[IDC]Dragonand the open() code had no cast, this is no guarantee
23:00:48[IDC]Dragonamiconn: we can prepare a V2 recorder ROM build equally well
23:01:16[IDC]Dragonit has the same remaining space, same start address
23:01:28 Join Smooth [0] (
23:01:32 Part Smooth
23:01:36 Join Smooth [0] (
23:01:38amiconnThe open code did two casts from (const char *) to (char *)
23:02:05[IDC]DragonI thought the strrchr() was doing this indirectly
23:02:24[IDC]Dragontaking a const char*, but returning a char*
23:02:29da4walkerOk, I understood it like that, I need also that microcontroller because of the special signal VW has implemented, and if I use Archos Jukebox MP3 player with Rockbox I also can control the jukebox with original buttons on my car stereo (also need another special cable for that) , is this right?
23:02:49[IDC]Dragonda4walker: I think so, yes
23:03:42da4walkerOK, buying the jukebox and flashing it with rockbox shouldn't be a problem
23:03:48amiconn[IDC]Dragon: I wonder why it will fit for the v2 while it won't for the fm. I think rockbox has the radio code in for both fm and v2?
23:03:52da4walkerbut u have a link where I can buy these cables I need?
23:06:29amiconnAh oops - I didn't take into account the archos fw may be different. The start address is the same for v2 as for v1 you said?
23:06:58[IDC]Dragonamiconn: yes
23:07:54[IDC]Dragonda4walker: the plug can be bought from Reichelt (Germany)
23:08:08[IDC]Dragonapart from that, I don't know
23:08:11da4walkeryeah, cool
23:08:14da4walkerdas brauch ich
23:08:50da4walkerdragon r u german?
23:09:05[IDC]Dragonamiconn: oops, you are correct about the Rockbox part
23:09:20[IDC]DragonI forgot Rockbox is so much larger
23:09:32amiconnJust checking out a compile...
23:09:35[IDC]Dragoncan you check if it fits?
23:09:47[IDC]Dragonda4walker: yes
23:10:11da4walkercool me too :)
23:10:43da4walkeru know the name of the plug on reichelt?
23:11:07da4walkeror the category?
23:11:09[IDC]Dragonda4walker: are we a bit lazy? ;-)
23:11:47da4walkerI am searching, but cannot find it on reichelt
23:11:55da4walkerof course everyone is lazy ;)
23:13:14amiconn[IDC]Dragon: It does fit, but this one is _very_ tight: while for the v1 recorder there are 6812 bytes left, for the v2 there are only 2636 bytes free
23:13:45amiconnDo you have a v2 to test?
23:14:36[IDC]Dragonamiconn: I have an FM, could use that
23:14:57[IDC]Dragonbut I'd have to force the V2 image into it first
23:15:15[IDC]Dragonwhich the plugins normally prevent
23:16:25[IDC]Dragonda4walker: part "KS K4-35" or "KS K4G-35"
23:16:28amiconnBtw, Saturn still sells v2 recorders (in Berlin)
23:16:34[IDC]Dragonhere too
23:17:57da4walkerthnx dragon
23:18:07amiconnMy first runtime test (running from rom) yielded 10:03 h of runtime - not bad, especially as the average bitrate of the mp3s used was 227 kbps and I used stock 1500 mAh cells
23:19:00da4walkerah, yeah but thats just the Klinkstecker, I thought of a place where I can buy the VWCDPIC thing... :(
23:19:26[IDC]Dragonda4walker: the author sells that
23:20:15 Join Doehni_ [0] (
23:20:17 Join flemse [0] (
23:20:28Doehni_hi there
23:20:46[IDC]Dragonamiconn: can't tell, but at least the ROM activity does not completely mess up the runtime
23:22:05Doehni_someone here who could help me? my rockbox won't start anymore
23:22:36[IDC]Dragontell us some details
23:22:53dids[IDC]Dragon: with rockbox running in rom.. Wouldn't stuff be writting in there a loot, and the flash chip is only meant to be written to so much right?
23:22:59Doehni_i heard music and shut the rockbox down. after a few minutes i tried to restart rockbox, but i only see the rockbox-logo
23:23:03da4walkerno he doesn't sell that
23:23:11Doehni_i only heard music before this
23:23:14da4walkerhe has stopped because he has problems with patents
23:23:41[IDC]Dragondids: only the code is in ROM, everything that's written has to stay in DRAM
23:24:11amiconnBah, my client is f***ed up
23:24:31 Part amiconn
23:24:46 Join amiconn [0] (
23:27:07da4walkernoone has a link for buying VWCDPic?
23:27:12[IDC]Dragonda4walker: drop me an email, a friend of mine is working on that, too
23:27:34Doehni_the only thing i can make now with rockbox is to plug it on usb and browse trough the folders by my computer... is someone there who know how to solve this problem?
23:27:51flemseerh... is there a limit on how many patches you can have on recorders?
23:29:40da4walkerok, just give me your mail dragon (query)... I will write u!!
23:30:28[IDC]Dragonnewer seen me in the group? well, idc-dragon at gmx dot de
23:31:03da4walkersorry, first time I am in this channel
23:31:49da4walkershould I write email on english or german :)?
23:31:54da4walkerwhich one do u prefer?=
23:32:40[IDC]Dragongerman of course
23:35:00amiconn[IDC]Dragon: Plugin api const'ed, as well as the irqname array (system.c)
23:35:20amiconnData section is now down to slightly more than 8 KB
23:35:56[IDC]Dragonshould I do the menus?
23:38:20da4walkeremail is out to
23:38:44da4walkerwould be really great if u say me when ur friend has success
23:38:50da4walkerhave to go now
23:38:51amiconnI wonder if the menus will make a big difference (are they defined all over the place?)
23:38:56da4walkerthnx for the infos so far folks
23:39:08 Quit da4walker ()
23:40:19amiconnFrom looking at the map file, the biggest remaining blocks of initialized data are (apart from the sysfont and lang.c, which must remain as it is) screens.c and mp3_playback.c
23:42:06amiconnAh - screens.c contains the usb logo, which surely can be const'ed
23:42:43[IDC]Dragonoh, I thought you did that one, with icons and rockbox splash
23:43:12amiconnI wonder why the usb logo is in screens.c, while the rockbox logo is in icons.c
23:43:41[IDC]Dragonmaybe because we have a USB screen
23:44:17amiconnBut you could say we have a splash screen as well...
23:44:50[IDC]Dragonat that time the application layer is not relly there, but anyway
23:44:56*dids likes the idea of a splash screen for usb
23:45:46 Part Doehni_
23:53:08amiconn[IDC]Dragon: from looking at mp3_playback.c, it seems all the lookup tables at the beginning (units, numdecimals etc.) could be declared const - please correct me if I am wrong
23:53:51[IDC]DragonI think you are right
23:54:10[IDC]Dragon(ahem , and I mentioned these as well ;-)
23:54:29amiconnGrr, I should read more careful

Previous day | Next day