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 2004-08-24

00:03:21amiconn(to check the i2c timing of course. Maybe the mas wants to wait in some more places by holding the clock pin low... perfectly valid according to i2c specs
00:20:42LinusNamiconn: i see some really suspicious stuff in init_recording
00:21:12amiconnLinusN: Yes?
00:21:38 Join webguest72 [0] (
00:21:55amiconnWhat is it?
00:22:21LinusNthe second 7f1 write doesn't wait for validation
00:22:39LinusNafter the sleep
00:23:07LinusNin fact, we should probably wait for validation after the app start as well
00:23:25amiconnReally - I didn't see that.
00:24:12amiconnWe don't have to wait for validation of the first one - there the apps isn't running yet, and after starting it we wait until it started
00:27:18amiconnI wonder why waiting for validation within start_prerecording (line 2211 ff.) uses val - shouldn't it use shadow_7f1 instead?
00:28:01LinusNwe don't use the shadow when reading
00:28:07amiconnOr do you do this because you want to "conserve" bit 0 as set
00:29:01LinusNstill, the second 7f1 write should wait for validation
00:29:38amiconnThe question is: should it wait before or after draining the dma buffer?
00:30:38*amiconn is trying that
00:37:17 Quit webguest72 ("CGI:IRC (Ping timeout)")
00:37:47amiconnThat's it !!! :))
00:38:51 Quit zeekoe (Read error: 113 (No route to host))
00:39:04LinusNamiconn: really?
00:39:37amiconnYup. I did 20 tries after adding the wait for validation, running from rom of course. All worked ok...
00:40:13amiconnI just have to quick-check if it records ok, then I'll commit
00:40:45*LinusN is pleased to have found that nasty little bug
00:41:05amiconnLinusN is a genius :)
00:41:30*LinusN is a genius :-)
00:41:52*LinusN wrote that code
00:42:16amiconnThere is still that .playlist_control problem ;)
00:42:41*amiconn will do some 1 min test recordings
00:44:04LinusNi introduced that bug when i added the prerecording feature
00:44:43amiconnYou took out the wait then?
00:44:58LinusNno, i moved some code around
00:45:06LinusNand forgot to add a wait
00:45:15amiconnAh ok.
00:45:44amiconnInterestingly, this triggered only that often when running from rom...
00:46:10LinusNmust be a timing thing
00:47:02amiconnIndeed, running from rom may be a bit faster in some places, despite the rom is only 8 bit wide
00:47:03LinusNi'll see if i can do some analysis in the next few days
00:47:23amiconn(no refresh and no page changes)
00:48:08lImbusi'm following the irc now since a few hours. it's better than a movie !
00:48:23LinusNlImbus: get a life :-)
00:48:57amiconnLinusN: I wonder if I should try using soft reset again (it's a bit cleaner imho)
00:49:04LinusNamiconn: can you reproduce the .playlist_control bug?
00:49:14LinusNamiconn: sure, why not?
00:49:16amiconnYes, easily
00:49:17lImbusI planned going to bed more than an hour ago, but doing some small work (that needed to be done, of course) while reading irc was better ;-)
00:50:09LinusNand there was a happy ending too, just like any good movie :-)
00:50:56lImbusmhmm. usually I hate happy ends. maybe it's the fact I know of a few other bugs ,-)
00:51:01lImbusok guys, I'm really of for bed now. congrats and good night.
00:51:08amiconnnite lImbus
00:51:31 Quit lImbus (" HydraIRC -> <- *I* use it, so it must be good!")
00:53:08amiconnBecause lImbus mentioned it - I know of at least 3 other bugs :(
00:55:27LinusNbtw, how's the main stack usage today?
00:55:49amiconnMain stack is big enough
00:55:59amiconnThe 3 bugs I'm thinking of
00:56:12LinusNi just added another 360 bytes in the rename function
00:56:16amiconn(1) .playlist_control corruptions - easy to reproduce
00:56:39amiconn(2) ata -31 error when booting rockbox from the archos charging screen
00:57:01amiconn(3) loading
00:57:17amiconn.. .cfg stopping playback, but not properly quitting wps
00:57:44LinusNtell me how to reproduce 1)
00:58:14amiconn[19:58:22] <amiconn> It's actually quite easy to trigger the bug:
00:58:15amiconn[19:58:52] <amiconn> (1) Set the cursor over a directory containing music. Do not enter it.
00:58:15amiconn[19:59:15] <amiconn> (2) Press On+Play and select playlist->add. The music will start playing
00:58:15DBUGEnqueued KICK amiconn
00:58:15amiconn[19:59:37] <amiconn> (3) Hit OFF to stop, the hit ON again -> corrupt playlist control file
00:59:03amiconnIt does _not_ occur if you play a file... you have to add a dir from Stop state
00:59:18amiconnSee also:
00:59:57amiconn[21:13:28] <amiconn> What I did understand so far: (1) If a playlist is generated the way I described (via On+Play) on the dir), it is much longer than otherwise - every single file is listed on a separate line
00:59:57amiconn[21:14:15] <amiconn> (2) The corrupt files just end prematurely, but not always at the same position (and the position is not sector aligned)
01:00:23amiconn[22:30:48] <amiconn> Bagder: It found out some more: If I create a playlist that way, DON'T stop the playback but plug in USB and look into the .playlist_control, it looks alright
01:00:23amiconn[22:31:25] <amiconn> After unplugging USB and re-plugging it, it is chopped the same way as it is when I hit OFF to stop
01:01:15LinusNplaylist->add? all i see is "insert"
01:01:30amiconnJust a moment...
01:01:36 Quit _aLF (Remote closed the connection)
01:02:14LinusNi can repro it now
01:02:23amiconn...yes, it's called insert
01:02:58amiconn(didn't check before, had it switched to german as usual)
01:03:31amiconnBtw, I did 4 test recordings, each ~2 min. All worked perfectly...
01:08:08CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
01:08:08*amiconn edits the RomBox wiki page...
01:12:33*amiconn has to rebuild the rombox packages
01:15:04LinusNgotta sleep now, cu around
01:15:13amiconnnite LinusN
01:15:23 Part LinusN
01:27:04***Saving seen data "./dancer.seen"
01:30:30 Join bagawk [0] (
01:34:47 Quit bagawk (Client Quit)
02:28:30 Part amiconn
02:42:45 Quit mecraw ("Trillian (")
02:46:52 Nick midk|gone is now known as midk (
03:01:50 Join Smooth [0] (
03:17:45 Quit AciD (Read error: 104 (Connection reset by peer))
03:27:06***Saving seen data "./dancer.seen"
03:52:07 Quit edx ()
03:52:09 Quit Smooth (Read error: 104 (Connection reset by peer))
03:59:24 Nick midk is now known as midk|bbs (
04:16:38 Quit Nibbler (Read error: 54 (Connection reset by peer))
04:28:04 Nick midk|bbs is now known as midk (
04:36:42 Nick StrathAFK is now known as Strath (
04:38:14 Join AciD [0] (
04:41:03 Part Strath
05:13:47 Quit AciD (Connection timed out)
05:27:10***Saving seen data "./dancer.seen"
05:37:56 Quit Ka_ (
05:37:56 Quit midk (
05:37:56 Quit maikeul (
05:38:26NJoinmidk [0] (
05:38:26NJoinmaikeul [0] (
05:38:26NJoinKa_ [0] (
06:00:09 Join Nibbler [0] (
06:14:07 Quit midk (Remote closed the connection)
06:17:53 Join midk [0] (
06:34:18 Quit Nibbler (Read error: 54 (Connection reset by peer))
07:27:13***Saving seen data "./dancer.seen"
07:57:54 Join Bagder [241] (
08:17:59 Join Nibbler [0] (
08:26:43 Quit midk (Read error: 104 (Connection reset by peer))
08:26:56 Join midk [0] (
08:30:22 Join midkay [0] (
08:30:29 Quit midk (Read error: 104 (Connection reset by peer))
08:48:01 Nick midkay is now known as midk (
08:49:55 Join amiconn [0] (
08:50:08Bagderhi amiconn, congrats on that recording bug squash!
08:51:22amiconnYeaj; I have to thank Linus as well.
08:55:52 Join amiconn_ [0] (
08:56:55 Quit amiconn (Nick collision from services.)
08:56:55 Nick amiconn_ is now known as amiconn (
09:02:18 Quit midk ("just STOP it arspy")
09:05:38 Join midk [0] (
09:09:01 Join Zagor [242] (
09:27:15***Saving seen data "./dancer.seen"
09:44:53 Quit Nibbler (Read error: 54 (Connection reset by peer))
09:50:07 Join kurzhaarrocker [0] (
09:51:18 Join [IDC]Dragon [0] (
09:51:35[IDC]Dragon'morning guys!
09:52:18[IDC]Dragonamiconn: Another congratulation to you and Linus for the recording fix!
09:53:08*kurzhaarrocker just read about that happily
10:13:50[IDC]Dragonamiconn: are you there?
10:23:00 Nick midk is now known as midk|sleep (
10:24:42amiconn[IDC]Dragon: Now I'm here
10:28:13[IDC]Dragonhi there
10:28:29[IDC]DragonI just posted about ROMbox on FM
10:30:24kurzhaarrockerlol! "downflashing the FM into a V2"!
10:31:18[IDC]Dragonwhy not?
10:31:54kurzhaarrockerIt's an idea that was just to weired to pop up in my mind. But on the other hand I have a V1 so I wouldn't bother anyway.
10:32:28[IDC]Dragonamiconn: does you current ROMbox build have the "ROM watchdog"?
10:37:12amiconn[IDC]Dragon: You could put the FM-to-V2 downflash description into the RomBox wiki article. I don't have full access to my mail atm
10:37:48[IDC]DragonOK, I'll do that.
10:38:16amiconnAnd no, the current rombox builds don't activate the "ROM watchdog", because ordinary rombox doesn't deactivate it again. If I would have done that, you'd always have to f1-boot in order to flash...
10:38:53[IDC]Dragon(you can review the mailing list vai haxx as well... ;-)
10:39:45amiconnI know
10:40:07kurzhaarrockerFAQ Q: Ok, I flushed the jukebox down, but now my earphones are all wet. Why?
10:40:07kurzhaarrockerA: You _flash_ your jukebox down, not _flush_ you moron!
10:40:21[IDC]Dragonjust for that mail excuse... ;-)
10:52:00amiconn[IDC]Dragon: I already mentioned downflashing to v2 within the article...
10:56:34[IDC]Dragonyes, I know. I just extended that sentence.
11:01:30 Join GoBeG4820 [0] (
11:01:51 Part kurzhaarrocker
11:04:12amiconn[IDC]Dragon: Does it really work that simple way? I thought the flash plugin would prevent that
11:05:22[IDC]Dragonthe FM and V2 are so similar that my checks don't tell them apart
11:06:07[IDC]Dragonbut the plugin itself does check it
11:06:31[IDC]Dragonthat's why you have to use the V2 build of it, the FM one would complain
11:09:07 Join Nibbler [0] (
11:09:38 Join webguest41 [0] (
11:10:55 Quit webguest41 (Client Quit)
11:17:30 Quit GoBeG4820 ("CGI:IRC (Ping timeout)")
11:17:40[IDC]DragonBagder: are you there? Yesterday we briefly chatted about the font format
11:17:58BagderI'm here
11:18:29[IDC]Dragonmy question was: would a compatibility break of the fonts be acceptable?
11:19:02[IDC]Dragonor should I try to runtime convert them while reading, in case I change the internal format?
11:19:02Bagdersure, if there are benefits that outweights the pains
11:19:21Bagdermost people get the full zip anyway
11:19:22[IDC]Dragonhard to tell
11:20:01[IDC]Dragonthe offset table is in longs, where shorts would be enough
11:20:21[IDC]Dragonand we could store the fonts already rotated
11:20:54[IDC]Dragonso, the font buffer can be smaller for the same effect
11:21:17Bagderyeah, that font rotation would be nice to get remove
11:21:20[IDC]Dragonthe short offset would save size on the internal font
11:21:24Bagderwould save us a little code space too
11:21:44[IDC]Dragonrather RO data
11:22:13BagderRO data for the internal font, yes, code for the not-having-to-rotate on load
11:22:22[IDC]DragonI'm not sure how fonts with height > 8 pixels are stored, but can look it up
11:22:48[IDC]Dragonstored handled, that is
11:22:57[IDC]Dragonstored & handled
11:23:51BagderI don't recall those details
11:26:00amiconn[IDC]Dragon: The font bitmaps are in rockbox native bitmap format after rotation
11:26:21Bagderah, right
11:26:34Bagderso that they are quickly lcd_bitmap()ed
11:26:46[IDC]Dragonso, row-wise stored
11:26:49amiconnBagder, [IDC]Dragon: It will even save some bss space, because the font buffer size can be lowered
11:27:19***Saving seen data "./dancer.seen"
11:27:45[IDC]Dragonprobably the moset difficult part is to change the bdf_conv tool for that general case
11:28:20[IDC]Dragon(any height, proportional width)
11:28:23*Bagder now spots that some fonts in the zip already are larger than 9K
11:29:06[IDC]Dragonthere's information in the file format which we never use, like the name
11:35:42amiconnBagder: If someone changes/extends a tool within the tools/ directory, does this get recompiled automatically (for the automated builds at
12:42:35 Join uski [0] (
12:56:13 Join lImbus [0] (
13:27:20***No seen item changed, no save performed.
13:37:32 Join edx [0] (
13:57:08 Join webguest74 [0] (
13:57:32 Part webguest74
14:50:12 Join AciD [0] (
15:08:40 Join dionoea [0] (
15:08:51dionoeahello every one
15:08:59Bagderhey ho
15:09:12dionoeaI was wondering how the simulator accessed the files on disk ?
15:09:24Bagderit assumes that a dir named 'archos' is the root dir
15:09:45dionoeadoes it use read and write functions ?
15:09:55dionoeaor is it using some rockbox specific fnuctions
15:10:30Bagderit simulates the functions rockbox provides
15:11:01dionoeaso it uses the fat functions provided by rockbox ?
15:11:18Bagderit works on any file system
15:11:26Bagderit simulates the read/write functions
15:11:41dionoeaso you have no way of testing fat functions other than on the real thing ?
15:11:47Bagdercheck rockbox/uisimulator/common/io.c
15:11:58Bagderdionoea: we have special fat test code
15:17:10dionoeastill considering hard drive / partition access : would it be possible to support multiple partitions ?
15:17:18dionoeawithout having to change the whole code
15:17:27Bagderwould be possible, sure
15:17:42Bagderbut the archos firmware only supports one
15:17:59dionoeaBagder: well rockbox isn't the archos firmware :)
15:18:14Zagorwhat would the point be?
15:18:16Bagderbut if you run non-flashed, you still need it
15:18:26dionoeawell, i was considerinf ext2 support
15:18:32Zagorthat won't happen :)
15:18:35dionoeaso you need a fat part to boot
15:18:44dionoeaand then you can use ext2
15:18:47Bagderyou'd live in your own fork ;-)
15:19:26Bagderthere's no point in doing ext2
15:19:31Bagderit'll only be more code
15:19:33Zagorthere is no rational reason for rockbox to use more than one file system type
15:22:31Zagorumm, I didn't mean to kill the conversion...
15:22:46dionoeait's true that it isn't really useful
15:23:18dionoeabut well, i was thinking that it might be fun to implement
15:23:58Bagderwe could have is a configure-time option 1) FAT 2)ext2 ;-)
15:24:22Bagder3) ext3 4)reiserfs4
15:24:23dionoeayep, be you would need a flashed box
15:24:51dionoea(and my box doens't flash ... :( )
15:25:01Zagoryeah it could be fun, but personally i find it even more fun to write useful code ;)
15:27:21***Saving seen data "./dancer.seen"
15:31:25Bagderlike pong
15:31:29*Bagder grins
15:50:09 Join GhUl [0] (
15:52:46 Part GhUl ("Leaving")
15:53:51 Join GhUl_ [0] (
15:55:07GhUl_hi @ all
15:55:21*Bagder waves
15:56:28GhUl_Do you guys have experience with recording using an external mic?
15:56:47BagderI don't
15:57:25[IDC]Dragonamiconn: I'd like to change bitmap_t into unsigned char, this will affect your grayscale fonts as well
15:59:00GhUl_other problem: i got my replaced recorder 20 last week, but the flash-info says its unknown ...
15:59:14GhUl_i thought only old models were unflashable
16:01:06amiconn[IDC]Dragon: Iirc this only affects gray_putsxy(), which you can easily adapt yourself. Just change the local variable bits to "unsigned char *", and remove the cast within the gray_drawbitmap() call
16:02:56lImbusGhUl_: it looks like it depends which chips were avaiable for fabrication.
16:03:07lImbusBTW: there are ONLY old archos ;-)
16:04:08lImbusI ordered 4 units at a eBay-store (mainly for 4 coworkers). 3 of them were flasheable, 1 was not.
16:04:56GhUl_hmm ok its such a big problem anyway
16:05:01GhUl_hmm ok its not such a big problem anyway
16:05:50GhUl_mostly i just pause it
16:09:01[IDC]Dragonamiconn: I'll try. This font change ripples through some stuff...
16:11:48amiconn[IDC]Dragon: It's only the poiter type itself that changes, the data format stays the same
16:19:37amiconnThere is still no plugin using grayscale fonts...
16:20:23[IDC]Dragonnot even your demo?
16:22:36amiconnNo, for the lack of screen space
16:22:55Bagdermaybe "bounce" should be improved
16:49:45 Quit GhUl_ (Read error: 110 (Connection timed out))
16:55:40 Join bagawk [0] (
16:58:53 Part Zagor
17:03:39bagawkBagder: there are some problems with the, in general
17:03:47Bagderlike what?
17:04:17bagawkBagder: you cannot be in a level higher than the rockbox directory when using the script
17:04:39bagawki had another one at one time... let me look
17:06:10bagawkalso, iwas thinking it might be good to just move the viewer rocks out of the main rocks directory
17:06:29bagawkjust have the script look at the the viewers.cofnig file to do it
17:06:40Bagderit does that
17:06:48bagawkit does?
17:06:50Bagder`mv .rockbox/rocks/$1 .rockbox/viewers`;
17:07:01bagawklet me try again, seems i always have them in the rocks dir also
17:07:20BagderI'll fix the bad "root dir" assumption
17:09:02bagawkheh silly me
17:13:22bagawki wonder what the highest capacity batteries are now
17:13:39bagawki just got 2300 mah yesterday
17:16:50Bagderfixed the path problem with builzip now
17:18:01 Join mecraw [0] (~lmarlow@
17:18:41Bagderthere seem to be at least 2500 mah AA batteries these days
17:22:48lImbusi've seen 3 Ah batteries in the size for the archos (can't remember if its AA or TRIPLE-A)
17:23:35lImbusI think battery constructors cheat more on these data than hardrives on their capacity :-/
17:23:54BagderI suspect so too
17:24:43Bagderthe last ones I bought had a fine print saying something like (these are guaranteed to provide NNN mah) where NNN were like 200 mah less than the big fat numbers they advertised
17:27:22***Saving seen data "./dancer.seen"
17:33:34 Part Bagder
18:25:43 Quit bagawk ("umount /dev/brain")
18:26:02 Join bagawk [0] (
18:47:56 Quit AciD (Read error: 110 (Connection timed out))
18:54:50 Join AxiD [0] (
18:59:10 Quit bagawk (Read error: 110 (Connection timed out))
19:14:02[IDC]Dragonamiconn: r u there?
19:21:04 Quit AxiD ()
19:22:29amiconn[IDC]Dragon: I'm here
19:24:53[IDC]Dragonah, I just sent you an email
19:25:37[IDC]Dragonwith a reduced and const'ed font
19:26:10 Join zeekoe_ [0] (
19:26:59[IDC]Dragonshould save ~2688 bytes RO and 4480 bytes bss
19:27:26***Saving seen data "./dancer.seen"
19:27:58amiconnWill try it soon, just a few qns:
19:29:03amiconn(1) It seems that while bits is not yet defined as unsigned char *, the data is already byte-packed?
19:29:46[IDC]Dragonyes, but only because of an even width
19:30:12 Nick zeekoe_ is now known as zeekoe (
19:30:35[IDC]Dragonand it may be subject of endian ordering, which a redefinition to bytes overcomes
19:31:39amiconn(There is no (2), just found the answer myself)
19:31:40[IDC]Dragonnote: of course you need to take out the rotate call in font_init()
19:32:16amiconnIt looks like there is indeed an endian problem...
19:32:32[IDC]Dragoncolumns intertwined?
19:32:54[IDC]Dragonhave you tested already on the target?
19:33:09[IDC]Dragonor just from looking at it?
19:33:16amiconnLook at the "-" - it reads 0x0800, 0x0808, 0x0808 but should read 0x0008, 0x0808, 0x0808 - sh1 is big endian
19:33:48amiconnDidn't try it yet, I have to do some shopping before the shops close
19:34:42[IDC]Dragonchanging the bitmap definition to bytes is a bigger job
19:35:46amiconn(away now)
19:42:16 Join _aLF [0] (
19:57:22zeekoethat's better
19:57:32zeekoei had numlock on :P
19:58:08zeekoeooh... you all have to try this:
19:58:11zeekoeit's soo fun
19:58:39zeekoeyou'll need a browser with speech support though, like opera 7.60p1 ->
19:59:59zeekoe_aLF: have you been here before?
20:00:11_aLFyes, sometimes
20:01:59zeekoecant remember seeing you :)
20:02:16_aLFa made a patch for radio one year ago
20:06:26_aLFinterresting your link, I will try
20:08:21 Quit Ka_ (
20:09:02zeekoei dont have radio
20:09:09zeekoejust v1 recorder
20:09:16zeekoeradio can be quite nice
20:10:20zeekoejust go to, press scroll lock or the microphone, and say "i would like twenty large pizzas with extra cheese, Olives, Mushrooms, Onions, Peppers, Bacon, Chicken, Ham, Meatball, Sausage and Pepperoni". Or anything else you'd like. It rocks :-P
20:10:23 Nick midk|sleep is now known as miek (
20:10:26 Nick miek is now known as midk (
20:10:35zeekoestill sleeping a bit, miek? :-P
20:11:09*zeekoe waits for midk saying "die die die" or something similar
20:11:16midkdie nerd
20:11:58zeekoethat should include you
20:13:11NJoinKa_ [0] (
20:16:40[IDC]Dragonleaving you guys...
20:16:55 Quit [IDC]Dragon ("CGI:IRC")
20:26:35 Quit maikeul (Read error: 110 (Connection timed out))
20:26:50 Join maikeul [0] (
20:45:12 Join bagawk [0] (
20:55:54*zeekoe is away: tea
21:27:23 Join [IDC]Dragon [0] (
21:27:29***Saving seen data "./dancer.seen"
21:27:57amiconnre [IDC]Dragon
21:27:58[IDC]Dragonhi again, Jens
21:28:06amiconnGot your mail...
21:28:34[IDC]Dragonyep, byte-font worked surprisingly well
21:28:38*zeekoe is away: I'm busy
21:28:48*zeekoe is back (gone 00:00:03)
21:29:19amiconn[IDC]Dragon: The size info is a bit weird - it tells how many words there are, but the real data is in bytes.
21:29:33[IDC]Dragonit's ignored anyway
21:29:51[IDC]DragonI could remove it from the struct
21:29:54amiconnAh ok
21:30:34amiconnYou talked about changing the loadable font format earlier today...
21:33:32bagawkbtw, i think the disk info is really screwed it says for me: Disk: 14.0gb Free: 14.1
21:33:37bagawkhow could that be?
21:34:03zeekoemy disk also has 15 gb free all the time
21:34:09zeekoeout of 18 gb
21:34:20[IDC]Dragonload format, yes?
21:36:04amiconnbagawk: This is most likely caused by Windows almost never updating the disk free info. Go to debug->View disk info, free space screen; then press play. This should fix it
21:36:07amiconn[IDC]Dragon: yes
21:36:28amiconnbagawk: Beware, that may take a minute or so
21:36:39bagawkahh i forget about that from when i had no jukebox
21:37:34[IDC]Dragonamiconn: what about the font file?
21:37:54*[IDC]Dragon doesn't know what amiconn wants to tell...
21:40:00amiconnI'm just asking if you would go for it and change the loadable format as well.
21:40:31[IDC]Dragonyes, otherwise I need to write an on-the-fly converter
21:42:24amiconnI think this shouldn't cause severe problems. There is a version number in the font files, so if you bump that for the new format, it would simply refuse to load old .fnt files
21:42:44[IDC]Dragonyes, I saw that
21:43:07[IDC]Dragonthe worst part is to change bdf_conv for the general case
21:43:36amiconnWhy is that? It seemed to me like you have done that already...
21:43:52[IDC]Dragonfor a special case, and for C output only
21:44:15amiconn(font format) You could toss away font name and copyright info too
21:44:18[IDC]Dragonspecial=fixed width, height=8
21:44:30[IDC]Dragonalready did that
21:44:37[IDC]Dragon(for the C version)
21:45:48amiconnIsn't it possible to simply reuse the rotleft() function?
21:46:46[IDC]Dragonmaybe, I have to look into it
21:49:17amiconnI had to change that function a bit when I changed lcd_bitmap() to use the rockbox native bitmap format...
21:52:27amiconnThat just remembered me that I wanted to add a stride parameter to lcd_bitmap() - this would simplify the implementation of lcd_putsxyofs(), saving some code
21:52:49amiconnIdeas ideas
22:01:04bagawkjust want to make sure i have the wiki thing ready before i submit my doc
22:01:18midkwiki thing ready?
22:02:56bagawkdo i just submit that as it is there
22:03:30midki spot three mispellings and countless grammar errors
22:03:45bagawki bet
22:05:49zeekoe−−-++ VU Meter
22:05:49zeekoe_A wonderfull set of VU meters!_
22:05:49DBUGEnqueued KICK zeekoe
22:05:49zeekoe| *Key* | *Function* |
22:05:50***Alert Mode level 1
22:05:50zeekoe| Off | Save settings and quit |
22:05:52zeekoe| On | Help |
22:05:54zeekoe| F1 | Settings |
22:05:56zeekoe| Up | Raise Volume |
22:05:58zeekoe| Down | Lower Volume |
22:06:00zeekoedon't use notepad++
22:06:04zeekoeoops... big paste :( sorry
22:06:17bagawkwhy would that not work?
22:06:22midkthat's fine
22:06:24midkit's wiki format
22:06:29zeekoeis it?
22:06:30bagawkthe twiki code formats it
22:06:42zeekoei'll keep my mouth shut
22:06:54midkyou better ;)
22:06:56CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
22:06:56*zeekoe like wee kee
22:07:21zeekoeor is it wookie :P
22:07:38midkthe official pronounciation is "wicki" or "weekee"
22:07:52midki don't see how people can pronounce it wicky. it just sounds/looks wrong
22:08:21zeekoeagreed :)
22:08:29zeekoewas just kidding
22:08:43midkwas just mentioning
22:09:01zeekoei think .gifs would be better than .bmps
22:09:26midk.pngs are suggested
22:09:36 Part lImbus
22:09:45zeekoethat's fine too
22:09:47bagawki have gifs
22:09:57zeekoei hate bmps
22:10:15midkpngs - my clock pngs are <1kb apparently
22:10:36bagawkmy bmps are very small
22:10:37midkSize: 0.66 KB (678 bytes)
22:10:49midkhaha... sure... 10kb perhaps
22:11:13midkthat's an extra .4kb!
22:11:25bagawkwhatever,i just use dthe screenshot thing
22:11:28bagawkit was came out
22:11:32bagawkdoes it really matter?
22:12:05amiconnBmp is badness within www. It's bulky, and not a real standard outside the Windows world
22:12:28midki told you!
22:12:37midkdidn't you see me tell him?
22:12:59bagawkmy god midk
22:13:48bagawkmine are 385 bytes in png
22:13:58midkWOA! OMFG.
22:15:45*zeekoe agrees with amiconn
22:15:51***Alert Mode OFF
22:16:06zeekoeyou _can_ use them, since even opera supports them now
22:16:09zeekoebut it's not nice
22:16:20zeekoefor standard's sake
22:16:33midkopera supported them even back when i used to use it, when i was ecstatic about v6.11 or whatever
22:16:53 Join Bagder [0] (
22:17:04 Quit mecraw (Read error: 104 (Connection reset by peer))
22:17:28 Join mecraw [0] (~lmarlow@
22:20:53zeekoeuh... ok
22:21:05zeekoei used opera a bit in the 5.12 times
22:21:30zeekoebut it became usable for most browsing in the later 6, early 7 series
22:22:16midkso.. not later 7?
22:22:45midki'm just saying.. "<zeekoe> you _can_ use them, since even opera supports them now"
22:22:51midkopera has supported them for a long long time
22:23:08amiconnzeekoe: It's not about Opera or [insert your favorite browser here] supporting it, but about web standards
22:23:09midkprobably back in 3.0
22:24:08zeekoeamiconn: ok, you're right about that :)
22:26:03zeekoei tend to forget that things when [my favorite browser] supports it, but i shouldnt
22:27:41amiconnIt is rumoured that there are platform with no Opera available. ;) What should their users do if they want to view such pages?
22:29:26zeekoeyep, you're right :)
22:31:32*zeekoe is away: I'm busy
22:33:04 Quit [IDC]Dragon ("CGI:IRC (EOF)")
22:38:36bagawkamiconn: i will be fixing those things, and adding those heatures we talked about wheni get odne with the checkers game
22:44:56*zeekoe is away: I'm busy
22:45:05*zeekoe is back (gone 00:00:02)
22:45:38zeekoemidk: die
22:45:52zeekoethou shalt not ctcp version me
22:45:58midki have a bigger version number than you
22:46:04zeekoei have a longer
22:46:07zeekoenever mind
22:46:43midkzeekoe, attention span?
22:46:55zeekoeyeah, that's what i was referring to
22:47:00zeekoei have a bigger OS version number than you
22:47:29midkwindows 5.1 < mandrake 10.0
22:48:01bagawkhis wondows kernel versoin is higher than your linux one
22:48:07zeekoeVERSION xchat 2.4.0 Linux 2.6.3-7mdk [i686/2.66GHz]
22:48:17midkok, let's count.
22:48:27zeekoei even have mandrake 10.0rc1, that's even bigger
22:48:32zeekoebut not on this pc
22:48:38midknt=1, 95=2, 98=3, me=4, 2k=5, xp=6
22:48:44midkxo.... xp < mandrake 10.
22:48:52midkeither way, it's less
22:49:13midkdo you know what this menas bagawk?
22:49:14zeekoewrong... you forgot windows 1.0, 2.0, 3.0, 3.11
22:49:18midkthey rebuild the kernel too much.
22:49:31zeekoeor they exist longer
22:49:34midkthey are less efficient at bug stomping.
22:49:42zeekoeor they upper version numbers more often
22:49:44midkthey rebuild often only to find the bug is still there
22:49:55midkzeekoe, that means i win if that's the case
22:49:59bagawki am talking about a kernel version, not windows verison
22:50:05midkwhy the hell are we arguing about who has a bigger kernel number
22:50:11midkbagawk, so were we.
22:50:12zeekoeyou started
22:50:18midkyou continued
22:50:37zeekoebut you started
22:53:40 Quit bagawk ("umount /dev/brain")
22:56:53 Join bagawk [0] (
22:57:10bagawkwo xchat 2.4 has a totaly differnt server screen
22:59:36 Nick zeekoe is now known as zeekoe|toothbrsh (
22:59:36DBUGEnqueued KICK zeekoe|toothbrsh
23:04:06 Nick zeekoe|toothbrsh is now known as zeekoe (
23:04:06DBUGEnqueued KICK zeekoe
23:15:08 Quit midk ("just STOP it arspy")
23:17:43 Quit zeekoe ("quit...")
23:18:10 Join cpu60 [0] (
23:20:29 Join zeekoe [0] (
23:23:13 Quit _aLF ("Leaving")
23:23:23CtcpIgnored 2 channel CTCP requests in 4 minutes and 30 seconds at the last flood
23:23:23*zeekoe is away: startrekTNG_S07-22
23:23:31 Join _aLF [0] (
23:24:46 Join Smooth [0] (
23:26:08Smoothhello bag
23:27:22 Quit cpu60 ()
23:27:24Smoothwhats up?
23:27:30***Saving seen data "./dancer.seen"
23:28:31 Join amiconn_ [0] (
23:28:44 Quit amiconn_ (Remote closed the connection)
23:28:56 Join amiconn_ [0] (
23:29:18 Part amiconn
23:29:40 Nick amiconn_ is now known as amiconn (
23:32:02Smoothi don't think archo's are made of metal anymore
23:32:04Smoothall plastic
23:35:59bagawkhi Smooth
23:42:56 Quit zeekoe ("quit...")
23:56:30 Quit Bagder ("Leaving")

Previous day | Next day