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 2006-03-15

00:00:27Paul_The_Nerdvoltagex: I'm not really sure, honestly.
00:00:28obothe current libscrobbler takes a unix epoch timestamp
00:00:46 Quit tinodeleste ("Client exiting")
00:01:05ohrnanyone around that can help me to get profiling working on a plugin?
00:01:45webguest69obo: that could be handled by the script that imports the logfile, simply tell it which timezone the timestamps are in
00:02:07webguest69leaving all the hard work off the target
00:02:41oboyeah, guess so
00:03:11 Quit midkay_ (Read error: 104 (Connection reset by peer))
00:04:37webguest69how about making the log-keeping a tsr plugin instead? Won't work when other plugins are loaded of course, but won't increase code-size either
00:05:38oboI thought about that - but if I was going to use it, I'd want it on all the time
00:05:46 Part LinusN
00:06:30voltagexPaul_The_Nerd: running the player with the AC adaptor plugged in seems to boost the CPU speed as rockboy is going faster.
00:07:22webguest69obo: how often are you using a regular plugin while listening to music though? Probably won't add much code-size to jam it into the core though, I guess..
00:08:28 Quit petur ("here today, gone tomorrow")
00:08:42Paul_The_Nerdvoltagex: That's... odd.
00:09:54obowell, since I just got a DAP, quite a bit at the moment - playing :)
00:10:18 Join Mong [0] (
00:11:01voltagexoh and I also found a program called CrossTool which builds crosscompilers automatically
00:14:50 Quit ender` (" Eighty percent of all people consider themselves to be above-average drivers.")
00:15:27 Quit webguest69 ("CGI:IRC")
00:17:06***Saving seen data "./dancer.seen"
00:18:02 Quit lImbus (" HydraIRC -> <- The future of IRC")
00:18:38 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
00:33:23 Quit matsl (Remote closed the connection)
00:37:10 Part amiconn
00:37:59 Quit youngcereal ("... und tschüß")
00:38:36 Part Paul_The_Nerd
00:43:47 Quit Moos ("Glory to Rockbox !!!")
00:45:01 Join kernel_sensei [0] (n=boris@
00:45:43 Quit kernelsensei (Nick collision from services.)
00:46:21 Nick kernel_sensei is now known as kernelsensei (n=boris@
00:52:48 Join JdGordon [0] (
00:59:34 Join midkay [0] (n=midkay@rockbox/developer/midkay)
01:02:35 Join stripwax [0] (
01:03:57 Quit voltagex (Read error: 104 (Connection reset by peer))
01:14:02 Quit obo ("bye")
01:21:40 Join damaki [0] (
01:24:40midkayyay, sharpe.
01:27:51 Part stripwax
01:37:20 Join nnod [0] (
01:39:26 Quit Mong ()
01:44:51nnodis there any rockbox port that currently runs on the gmini XS200?
01:45:14 Quit muesli__ (Read error: 110 (Connection timed out))
01:50:25 Join Rob2222 [0] (
01:50:51sharpei'm bored... midkay, i'm going to get to work on the packaged wps idea thingie, this weekend hopefully...
01:52:00JdGordonif your bored, why not strt it now?
01:52:33 Join linuxstb_ [0] (
01:53:08 Quit linuxstb (Nick collision from services.)
01:53:09 Nick linuxstb_ is now known as linuxstb (
01:53:39sharpei've previous matters to attend to...
01:54:15sharpepretty much this week
02:07:26 Quit Rob2222_ (Read error: 110 (Connection timed out))
02:17:09***Saving seen data "./dancer.seen"
02:22:19 Quit xmixahlx ("blah blah blah")
02:23:39 Join imphasing_ [0] (
02:31:57 Part radsaq
02:35:03midkaysharpe, that's cool
02:42:07 Quit imphasing_ (" webchat CGI:IRC 0.5.4 (2005)")
02:56:12 Join fiftyfour123 [0] (
02:56:48 Quit fiftyfour123 (Client Quit)
03:01:18 Join tvelocity [0] (
03:07:13 Quit tvelocity ("Ex-Chat")
03:11:26 Quit midkay (Read error: 104 (Connection reset by peer))
03:11:29 Join midkay_ [0] (
03:15:30 Quit paugh ("Leaving")
03:17:11 Join nave7693 [0] (
03:17:21 Join Spida [0] (
03:19:30 Part nave7693
03:27:35 Quit Spida_ (Read error: 110 (Connection timed out))
03:41:52 Part fantomas
03:49:26 Join voltagex [0] (
04:02:50 Quit thegeek (Read error: 104 (Connection reset by peer))
04:02:51 Join thegeek_ [0] (
04:02:51 Quit midkay_ (Read error: 104 (Connection reset by peer))
04:03:32 Join midkay_ [0] (
04:08:33 Quit midkay_ (Read error: 104 (Connection reset by peer))
04:09:23 Join midkay_ [0] (
04:11:12 Quit Daishi ("Client exiting...")
04:17:10***Saving seen data "./dancer.seen"
04:19:44 Quit nnod ("Leaving")
04:19:45 Quit midkay_ (Read error: 104 (Connection reset by peer))
04:20:21 Join midkay_ [0] (
04:29:57 Join nave7693 [0] (
04:36:45 Nick Kilohert is now known as kilohurt (
04:52:19 Join virtualball2 [0] (
05:06:09 Join Kreator666 [0] (
05:07:12Kreator666I think that you guys might want to have a look at this :)
05:12:38midkay_Kreator666, nice.. one thing first off, though, it doesn't seem to check for errors.. i extracted just the batch file and ran it, and although i got a few "file not found errors" the script continues asking me which model i have.. :)
05:12:53 Part nave7693
05:13:07midkay_The system cannot find the file specified.
05:13:07midkay_Bootloader copied, all files in place!
05:13:07DBUGEnqueued KICK midkay_
05:13:07midkay_Press any key to continue . . .
05:13:22Kreator666what ipod, what file?
05:13:28midkay_no ipod, no files.
05:13:54midkay_it just tries to do what it thinks it should and says success even without the appropriate files present..
05:14:56Kreator666that's the thing with bat files, there really is only 'if exist' to check files
05:15:17midkay_are you using that or what? it doesn't appear so..
05:15:22Kreator666have you extracted everything to the same folder? that's what it relies on
05:15:26midkay_i don't know much about batch files anyways.
05:15:28midkay_Kreator666, no, that's my point.
05:15:41Kreator666"IF EXIST C:\Rockbox\bootloader-video.bin ECHO Bootloader copied successfully!"
05:15:42midkay_i purposefully did not connect my ipod and did not copy all the files.
05:15:51Kreator666yeah well
05:15:53midkay_Kreator666, ah, that'd make sense.
05:16:09midkay_i do have a few files in c:\rockbox from my last install.
05:16:43Kreator666you need to extract everything in the zip to the same folder, as these are the references that it runs off
05:16:52Kreator666before changing with the cd command
05:17:11midkay_Kreator666, never mind..
05:17:46Kreator666never mind what? i can see that you are purposefully putting things in the wrong place..
05:17:54Kreator666Mr. Trivial
05:18:05midkay_Kreator666, yes, that, was, my, point.
05:18:42midkay_i said the batch file was not clever because even with a lack of files it claims "success". you explained that it checked in c:\rockbox for a certain file. i said ah, i do have that certain file from when i installed manually.
05:19:30Kreator666ok then
05:19:58Kreator666batch files are really basic anyway, but i wouldve thought that they would be good enough for this type of thing
05:20:18Kreator666definitely wasnt expecting C++ quality coding from this
05:20:26Kreator666extremely simple
05:20:33midkay_Kreator666, i'm not saying they aren't. i was under the impression there were no file checks.
05:20:56Kreator666there are, just not at the start
05:21:15Kreator666ill put them in then, to check for the bootloader/fw files at the beginning
05:21:22midkay_Kreator666, do what you wish
05:26:21 Join Leperkawn [0] (
05:28:28 Quit Leperkawn (Client Quit)
05:30:41 Part Kreator666
05:46:08 Quit virtualball2 ("virtualball2 has no reason")
06:17:13***Saving seen data "./dancer.seen"
06:21:34 Quit imphasing (Read error: 110 (Connection timed out))
06:23:54 Quit RotAtoR ("zzzzzz")
06:28:33 Join Leperkawn [0] (
06:28:40LeperkawnHello all.
06:29:23LeperkawnIs anyone active at the moment?
06:31:03 Quit phaedrus961 (Read error: 110 (Connection timed out))
06:31:21midkay_Leperkawn, what's up?
06:31:31LeperkawnNot too much.
06:31:35LeperkawnI just have a question.
06:31:44LeperkawnHow would I install a WPS on my 5G?
06:32:01midkay_the wps file itself goes in .rockbox/wps
06:32:11LeperkawnWhat about the other files?
06:32:17midkay_if you download a zip file from the WpsGallery it generally has the file structure already set up.
06:32:34midkay_if it has some folders in the zip.. ie backdrops, wps, themes..
06:32:39LeperkawnSo I just drop the "brushed_metal" folder into WPS?
06:32:42midkay_then you just need to copy that all to .rockbox.
06:32:45midkay_haha. that's mine. :)
06:32:58LeperkawnYou created it?
06:32:58midkay_go inside the folder..
06:33:12midkay_yes, i posted it a night or two ago :)
06:33:15LeperkawnIt's nice.
06:33:19LeperkawnAnyways, I go inside wps
06:33:28midkay_thanks. go inside brushed_metal, what's in there?
06:33:38Leperkawnbackdrops, fonts, themes, wps
06:33:49midkay_yes, copy all four of those folders to .rockbox
06:34:04LeperkawnInside \wps correct?
06:34:06midkay_that's generally what's included with a file from the WpsGallery.
06:34:09midkay_no, just .rockbox
06:35:05LeperkawnNow I shall boot up rockbox and test
06:35:17midkay_yes. "browse themes" from the menu, and choose brushed_metal.
06:37:02LeperkawnNot showing up.
06:37:23midkay_brushed_metal in the Browse Themes menu?
06:37:53LeperkawnI think i know what I did wrong
06:38:06LeperkawnWas I supposed to copy those 4 folders, or brushed_metal?
06:38:57midkay_all four folders.
06:39:01midkay_they correspond to folders in .rockbox.
06:39:07LeperkawnOh, I see.
06:39:58midkay_the way it works, is.. the WPS file goes in .rockbox/wps; if the WPS has bitmaps associated with it, they go in /wps and a subfolder with that WPS' name. a theme config file goes in .rockbox/themes which tells rockbox to load that WPS along with set specific colors or a backdrop, etc.
06:40:34LeperkawnI see.
06:41:17LeperkawnHow can I reset my ipod while it's attached in USB mode?
06:42:17midkay_you can only reset settings outside of rockbox by flipping the hold switch on during boot..
06:42:33Leperkawnthat would explain alot
06:42:40LeperkawnWhy my settings randomly erased.
06:42:48midkay_yes, it's quite annoying..
06:42:55LeperkawnReset, I mean, as in reboot
06:43:02midkay_..unplug usb?
06:43:10LeperkawnI guess that would work
06:43:16LeperkawnI was wondering if there was a key combo.
06:43:29midkay_oh, no.
06:43:57LeperkawnNiice theme
06:44:06CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
06:44:06*Leperkawn gives midkay thumbs up
06:44:35LeperkawnIs it just me, or is navigation a bit annoying in the 5G?
06:45:03midkay_thanks. :) annoying?
06:45:14LeperkawnAnd why does it take quite a bit of time for my compy to see my ipod once I plug it in?
06:45:21LeperkawnMaybe it's just becasue I'm new
06:45:24midkay_fine for me. may be a bit of time before you get used to it once you've used the ipod firmware a while..
06:45:31Leperkawnbut if I want to view a playlist I have to go though all those menus
06:45:34midkay_only thing that bugs me is >> "accepting" settings in the menu.
06:45:48midkay_Leperkawn, during playback.. hold SELECT, "playlist", "view current playlist".
06:46:20midkay_everything gets a bit clearer as you use it more..
06:46:29LeperkawnI'm just a wee little noob :P
06:48:14LeperkawnGOod ol' windows
06:48:32Leperkawnit says "I don't think Josh wan't control of his taskbar. I'll just freeze it!"
06:48:52Leperkawn"and whatever commands he told it to do, I shall do roughly 5 minutes later, severely pissing him off!"
06:49:00midkay_yeah, that's basically it :)
06:49:05*Leperkawn pats windows on the back
06:49:12LeperkawnWith a knife.
06:49:13midkay_partially related to the hardware disk mode, i think..
06:49:56*Leperkawn watches my windows appear and dissapear roughly 10 times as "Show Desktop" is ran 10 times in a row, about 5 minutes after I hit it
06:50:13midkay_Leperkawn, haha, that's all too familiar :)
06:50:38LeperkawnTime to put some rage, a7x and nin on my pod.
06:50:49Leperkawnthe other 10k songs can wait.
06:50:57 Quit linuxstb (Read error: 110 (Connection timed out))
06:51:06midkay_Leperkawn, haha :)
06:51:30 Join linuxstb [0] (
06:51:36LeperkawnAnd the other 20k unorganized songs from torrents can REALLY wait
06:52:02LeperkawnThey Might Be Giants discography someone decided they didn't want to organize or tag properly, I'm talking to you.
06:52:10Leperkawn(and the rest of the torrents)
06:52:27LeperkawnVehl, its about time I turned in.
06:52:35midkay_hahaha.. if you're going to download that much, at least have the.. you know.
06:52:42midkay_.. patience to tag it all manually!
06:52:44LeperkawnI do
06:52:47LeperkawnJust not
06:52:50LeperkawnI dont :P
06:53:05LeperkawnI can do it in small chunks.
06:53:22Leperkawnlike sit down for a few hours, bang my head off of some concrete blocks, take a nap, repeat
06:53:33midkay_Leperkawn, that's what it makes me want to do..
06:53:34Leperkawn(sit down = sit down at computer and tag songs)
06:53:37Leperkawnto clear that up.
06:54:08midkay_after 5 minutes of tagging i go slowly insane and would rather slam my head against the sharp jagged edge of my brick fireplace.. ah, right on the corner.. with the point facing in towards my eye socket..
06:54:08Leperkawnanyways, 0600 comes early
06:54:33midkay_that'd own though.
06:54:42midkay_accidentally stabbing yourself in the eye with the corner of a fireplace..
06:54:45LeperkawnErm, except for the person who would have to clean up the mess
06:55:00LeperkawnWhoops I tripped... into the corner of the fireplace... with my eye...
06:55:07Leperkawnyeah... can we get a doctor?
06:55:13*Leperkawn bleeds.
06:55:29LeperkawnAnyways, thanks for the help
06:55:34Leperkawngotta get up for 0600
06:55:38midkay_no problem. :) alright. laters!
06:55:59*Leperkawn disappears
06:56:01 Quit Leperkawn ("Chatzilla 0.9.71 [Firefox]")
06:56:37qwmBg3r: i'm here to relieve you from your grief.
06:59:59 Join needleboy [0] (
07:10:58 Quit voltagex (Read error: 104 (Connection reset by peer))
07:13:37MikachuPaprica: that was 1:34 in the night :)
07:24:21 Join savagenator [0] (
07:25:17savagenatori need help registering at rockbox forums
07:25:30savagenatorit wont send me the email to activate it
07:25:41 Join banan_ [0] (
07:28:22 Quit [TCK] ("well, if you say so.")
07:28:27ScoTTie_try another account
07:28:32ScoTTie_(email account)
07:28:53savagenatori like the have the screen name savagenator....but i'll try it
07:28:53 Quit Sinbios (Connection timed out)
07:29:14 Join Sinbios [0] (
07:30:15savagenatornow i remember why i needed help here for real (i'm tired) i only have one email adress, and i already registered
07:30:19savagenatorbut it wont send the email
07:30:48ScoTTie_guess you'll have to wait for whoever runs the site to show up
07:30:59ScoTTie_not sure who it is, so i cant help you there.
07:31:12savagenatori'll come back tommorow then
07:31:48savagenatorif anyone has an acound at MR that can help me my screen name is savagenator....thanks scott
07:32:16 Quit Ashex ()
07:32:17 Quit savagenator ("CGI:IRC (EOF)")
07:38:29 Quit sharpe (Read error: 110 (Connection timed out))
07:42:11 Quit merbanan (Read error: 113 (No route to host))
07:45:10 Join Paul_The_Nerd [0] (
07:45:49 Join ender` [0] (i=ychat@
07:52:20 Quit needleboy ()
08:11:55 Join phaedrus961 [0] (
08:11:59 Nick banan_ is now known as merbanan (
08:15:59 Join B4gd3r [0] (
08:17:15***Saving seen data "./dancer.seen"
08:20:42 Join Rob2222_ [0] (
08:20:49 Join needleboy [0] (
08:28:20 Quit midkay_ (Read error: 104 (Connection reset by peer))
08:29:11 Join midkay [0] (n=midkay@rockbox/developer/midkay)
08:30:09 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
08:30:15 Part LinusN
08:34:18 Quit B4gder (Read error: 113 (No route to host))
08:36:42 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
08:37:12amiconngood morning
08:37:47Bg3rmorning, ami
08:38:56 Quit Rob2222 (Read error: 110 (Connection timed out))
08:39:48Paul_The_NerdTechnically it's morning here too by this point BHSPitLappy
08:39:55 Quit B4gd3r (Read error: 113 (No route to host))
08:42:41 Join LinusN [0] (
08:44:37amiconnJdGordon: I've completed my build tests.
08:46:03amiconnUnlike I thought before, compiling the same cvs code with the same gcc crosscompiler version on different host systems produces identical binaries.
08:47:22amiconnI tested with binutils 2.16.1 + gcc 3.3.6 + newlib 1.14.0 on debian-testing-amd64, debian-testing-i686 and cygwin.
08:49:10amiconnRecorder v1 builds from cvs state 13 Mar 2006 18:00 UTC produced an apps/rockbox.bin of 212568 bytes
08:52:46Paul_The_Nerdamiconn: I also got the sizes you reported, on my Cygwin builds of them
08:53:15Paul_The_NerdFrom when you were talking with JDGordon about it a couple days back
08:54:11amiconnThe builds from my patched tree are smaller, because it contains my (unfinished) language cleanup
08:55:05 Quit wenko (Read error: 113 (No route to host))
09:06:40 Join Zagor [0] (n=bjst@rockbox/developer/Zagor)
09:11:18 Quit Sinbios (Connection timed out)
09:11:46 Join Sinbios [0] (
09:17:31 Join petur [0] (
09:18:54JdGordoncheers amiconn
09:19:59 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
09:20:30 Part Paul_The_Nerd
09:25:31 Quit needleboy ()
09:28:55 Join muesli__ [0] (n=muesli_t@
09:31:22 Join Ashex [0] (
09:32:53 Quit midkay (Read error: 104 (Connection reset by peer))
09:33:00 Join midkay_ [0] (
09:41:50 Join needleboy [0] (
09:52:12muesli__amiconn r u arround?`
09:55:48 Quit needleboy (Read error: 104 (Connection reset by peer))
09:55:49 Join warewolf [0] (
09:56:00warewolflinus around?
09:57:29warewolfI founded a bug
09:57:40warewolfI think it relates to the h3xx remote detection code
09:58:12warewolfif you boot rockbox with the remote's hold button on, and boot by pressing play on the unit itself, the remote doesn't work correctly. I think it maps the buttons to the other kind of remote
09:59:00Bg3rwarewolf yes, this is known bug
09:59:02muesli__warewolf using the h3xx lcd remote?
09:59:21muesli__that babe is a pain in the ass :o
10:00:24Bg3rafaik the problem is that when the remote hold is on, the adc which is used for remote type detection starts to return very different values
10:01:35warewolflooks like it
10:01:50warewolfI like being able to change the default bg/fg colors now
10:01:51warewolfthat's nice
10:02:01warewolfI reversed the colors white text on black so I could read it easier
10:02:47muesli__warewolf does the remote works properly with official build? (normal booting without hold on(just using the experimental)
10:02:59warewolfI used today's build
10:03:03warewolfnot CVS or bleeding edge
10:03:37muesli__but the rest works fine besides the issue youve mentioned?
10:04:23warewolfaslong as the unit isn't booted with the remote hold switch on
10:04:41warewolfturning hold on after booting works as expected
10:04:48warewolfturning it back off also works as expected
10:06:24warewolfer, rewind
10:06:30warewolfI'm not using the official H3xx remote
10:06:35warewolfI'm using a H1xx remote on a H3xx
10:06:48warewolf'cause the H3xx remote wasn't available for FUCKING FOREVER in the US
10:06:48muesli__mk :-/
10:07:11warewolfiirc the official h3xx remote doesn't have a LCD neither
10:07:17muesli__just copied files, will test it myself ;-)
10:07:24warewolfs/official h3xx/official US h3xx/
10:07:41muesli__yepp, that one quite useless
10:08:55muesli__keys are mixed up while using the bleeding edge
10:09:19warewolfI stopped using the bleeding edge code when it did unexpected things :)
10:09:44muesli__dont tell the devs ;)
10:10:01 Join Lynx_ [0] (
10:10:26warewolfsomeone please *please* explain wtf "Car Adapter Mode" is?!
10:10:54muesli__something concerning the charger
10:11:16*warewolf confused
10:11:21Lynx_has to do with disk spin up time a skipping
10:11:33Lynx_but i'm sure it's in the manual
10:11:45 Quit tenzing (Read error: 110 (Connection timed out))
10:13:23warewolfok I found it in the wikimanual
10:14:08Lynx_ok, seems like i was wrong.
10:15:39 Join Jungti1234 [0] (n=jungti12@
10:16:33 Join tenzing [0] (
10:16:49muesli__hi Jungti1234
10:16:56muesli__dont like strings?
10:17:18***Saving seen data "./dancer.seen"
10:18:42Jungti1234I have question
10:18:46Jungti1234ATA error: -32
10:18:46Jungti1234Insert USB cable and press
10:18:46Jungti1234a button
10:19:02Jungti1234What's meaning?
10:19:27LinusNbad hard drive copnnection or too low battery
10:20:00Jungti1234But, files(folders) see at USB connection.
10:20:04 Quit damaki (Read error: 110 (Connection timed out))
10:20:31LinusNwhen did this start to happen?
10:22:42Jungti1234I don't understand
10:23:24Bg3rJungti1234 when did this happen for the first time ?
10:24:44Jungti1234After put a song to H300, disconnect USB and booted.
10:28:05 Join Paul_The_Nerd [0] (
10:32:59 Quit nudel ("At Argon, we're working to keep your money.")
10:37:13 Join nudel [0] (
10:37:45Bg3rJungti1234 did you try to stop/start it ?
10:39:08Jungti1234It doesn't work.
10:41:02 Quit midkay_ (Read error: 104 (Connection reset by peer))
10:41:28Bg3rJungti1234 have you dropped your unit to the floor ?
10:42:15Bg3rdid you do a chkdsk <your_iriver_drive> /f /r
10:42:40Jungti1234I didnt drop it.
10:43:41Bg3rokay, then i suggest you to connect it to your PC again, run cmd (start->run->cmd)
10:43:48 Quit nudel ("At Argon, we're working to keep your money.")
10:43:48Bg3rand there run the chkdsk
10:43:58 Join nudel [0] (
10:44:23Bg3ralso, have you tried to boot the iriver fw ?
10:44:59nudelYay, found out how to make mIRC automatically connect to two networks at startup, without me having to click 1000 things.
10:45:27Jungti1234Bg3r: yes
10:45:44Jungti1234It marks '0000'.
10:45:59 Quit crashd ("leaving")
10:46:09Jungti1234But, file seem if connect USB.
10:46:14 Join midkay_ [0] (
10:46:34 Join ashridah [0] (
10:49:00 Join crashd [0] (
10:51:05Mikachuare there (on the headphone cable) remotes for ipods?
10:51:44nudellooks like it:
10:51:45 Quit midkay_ (Read error: 104 (Connection reset by peer))
10:52:00 Join midkay_ [0] (
10:52:57Mikachui just see a radio remote
10:53:38nudelWith the wired remote, you can control slideshow, video and music playback — adjust volume, skip forward/back within a playlist −− plus switch between radio stations and turn the radio on or off, even if your iPod is in your pocket or backpack.
10:55:05nudelalso plugs in the bottom of the device and has no screen so is a bit rubbish if you ask me :(
10:55:27amiconnmuesli__: Now I am
10:57:12 Join needleboy [0] (
10:58:00 Quit quobl (Remote closed the connection)
11:00:21 Join Kohlrabi [0] (
11:03:55PapricaMikachu, ready to check tetrox on nano?
11:05:32Mikachusorry, not today, at least not until i'm done with this assignment
11:06:43 Join youngcereal [0] (
11:07:33Mikachuand about die(), i didn't just move the code to it's own function, i also added the else {} thing around the rest
11:07:43Mikachuthat fixes something but i don't remember what :P
11:07:51Mikachuyou can do that in the original code without separating it
11:08:19Papricai made some cleanups to the code, i will commit them and the patch at the night
11:10:29Papricatetrox on nano
11:11:07Mikachuhow does it look with a black background?
11:11:31Papricawhy black? =\
11:12:23Mikachuthe tetris clone i used to play was black :)
11:14:52 Quit warewolf ("leaving")
11:15:30muesli__amiconn ive noticed that a-b isnt used in the jpeg-viewer. could we use that key for displaying exif-data?
11:15:39linuxstbMikachu: I agree, a black background looks better.
11:15:50Mikachuyeah i think so
11:15:53Paul_The_Nerdmuesli__: How would you display it on other targets?
11:16:01Mikachubut maybe the pieces bitmaps need some adjusting then
11:16:11*Paul_The_Nerd votes for black as well.
11:16:26linuxstbAlso, what was the size (in squares) of the original tetris? That screenshot doesn't seem to be in the correct proportions.
11:16:41Mikachui'm pretty sure it's 10 wide
11:17:11muesli__Paul_The_Nerd how do you mean that?
11:17:14dwihno10 wide
11:17:19dwihnoheight, I don't know
11:17:41Mikachuabout 20-30 maybe, but i think it's different on all clones
11:17:54Papricatetrox screen is 10 wide too
11:18:15Papricaon all the targets, also at archos
11:20:38amiconnmuesli__: A-B *is* used in the jpeg viewer, it's "zoom out"
11:20:53Paul_The_Nerdmuesli__: The JPEG viewer also works on iPods. Which don't have an A-B button.
11:21:15amiconn..and I have no idea how exif works (that can be changed), and where/how we could display it
11:21:30Mikachuscroll wheel is zoom on ipods
11:23:13linuxstbLooking here, Tetris seems to be about 20 high, and 10 wide, with a few rows of space at the top for the new shape to appear in:
11:23:19 Join mirak [0] (
11:25:29Paul_The_NerdAnd Tetrox is 10*12ish vs the 10*20 of the arcade?
11:26:07Papricasomthinng like that =\
11:26:38Paprica10*17 on h300
11:26:48Paprica10*13 nano
11:26:56Paprica10*12 h100
11:26:56Paul_The_NerdIn my opinion, it should at least be the same height in boxes on all targets...
11:27:09Paul_The_NerdOtherwise the gameplay experience is rather different
11:27:12*linuxstb agrees
11:28:24linuxstbUnless you rotate the screen (like rockblox), it seems that there will have to be a lot of wasted screen space.
11:29:26linuxstbLooking at the arcade screenshot, it seems that 10x24 is needed (to give space for the new block above the 20 actual rows)
11:29:31 Quit midkay_ (Read error: 104 (Connection reset by peer))
11:30:02 Quit youngcereal (Remote closed the connection)
11:32:07 Join midkay [0] (n=midkay@rockbox/developer/midkay)
11:32:49Mikachumaybe would make sense to flip the display on some targets?
11:33:04Mikachuand by flip i mean rotate
11:33:19Mikachubut i guess some players may be hard to operate then
11:34:39Papricai'll decrease the block size
11:35:45linuxstbIf you go for a 10x24 playing area, and don't rotate, then you'll be using less than one third the width of the screen. e.g. 220x176 would need 7x7 squares, which is 70x168
11:36:38linuxstbBut rotating it would allow for 9x9, which is 216x90 - more than half the height of the screen.
11:37:05Papricanot 24
11:37:13linuxstbWhere do the new blocks appear?
11:37:27Mikachuso they are visible when they appear
11:38:07linuxstbIIUC, the actual playing area is 10x20, but you need space at the top for the new block.
11:38:16linuxstbWhich can be up to 4 row high.
11:39:07Mikachui think those 4 lines are included in the 20
11:39:37Paul_The_NerdHmm... Looks like the GB version was a little less high than 20
11:40:42Mikachuthe text mentions it was an exception with 10x18 lines, being a bit harder than the nes version
11:40:53Paul_The_NerdYeah, basically the less lines the harder.
11:41:03muesli__Paul_The_Nerd amiconn sorry, got 2 go..but will talk to you l8er
11:41:08Paul_The_NerdI think the the 4 extra lines can be considered offscreen.
11:41:38Paul_The_NerdAll you'd have to do is put ^, ^^, ^^^, or ^^^^ markers along the top row to show where the next piece will fall
11:41:56Mikachunew blocks should appear completely inside the playing field
11:43:08linuxstbI would prefer rotating the screen and having the four rows above the playing field...
11:43:16Mikachuat least, that's my preference
11:43:24Mikachutetris worlds for gba appears them above..
11:43:27Paul_The_NerdI definitely vote for rotating the screen.
11:44:06amiconnWhy not just make rockblox work properly on the other targets?
11:44:47Paul_The_NerdOr that. That works too
11:45:11Mikachuwhat does it have over tetrox?
11:46:28Paul_The_NerdIt's been around for a while, it has a proper sized playing field, and behaviour will be exactly the same across all bitmap screen models?
11:49:13amiconnI think so. The only missing thing is proper scaled gfx per target, and perhaps colour
11:49:23amiconnWhy reinvent the wheel?
11:49:57Paul_The_NerdIt feels a wee bit fast on the h100, I don't know if that's just because it's so small, or if it's just a bit quicker than I'm used to for beginning a game of tetris
11:52:12 Quit midkay (Read error: 104 (Connection reset by peer))
11:52:13 Quit needleboy ()
11:52:14 Join Moos [0] (
11:52:15 Join midkay_ [0] (
11:56:48linuxstbamiconn: I agree, if it was me, I would have just improved Rockblox. But IMO, as long as the end result is a game that is at least as good as Rockblox, and works well on all platforms, then it's up to the person doing the work how to achieve it.
12:01:47 Join quobl [0] (n=quobl@tor/session/x-52293daf81b7ce29)
12:03:47 Quit petur ("overloaded")
12:10:23 Quit quobl (Remote closed the connection)
12:10:28 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
12:10:58 Join quobl [0] (n=quobl@tor/session/x-264c0038452d2a4f)
12:11:17preglownothing like waking up and feeling you've got a cold coming on
12:11:56ashridahyay, vitamin C time :P)
12:12:02*tucoz hands preglow a large cup of tea (with a dash of rum in it)
12:12:13ashridahooh ooh. alcamahol!
12:12:47Mikachuvitamin java
12:12:54tucozOr pirateoil as they called in an old swedish translation of Tintin
12:13:10Mikachuheh, they're airing tintin episodes on tv here (sweden)
12:14:41ashridahaah. tintin, that brings back memories
12:14:46LinusNi like tintin
12:14:53ashridahyeah, i did too.
12:15:00ashridahthat an asterix :)
12:15:13LinusNand gaston!
12:15:23Paul_The_NerdMan, I used to love Tintin and Asterix.
12:15:33Paul_The_NerdI don't see those very often over here.
12:17:21***Saving seen data "./dancer.seen"
12:17:33 Join damaki [0] (
12:18:17Mooswee for the french cartoons ;-P
12:20:08ashridahI've got to say, my favorite tintin of all time was the one where the professor made capt. haddock 'immune' to alcohol :)
12:20:22ashridahkind of thing i'd love to do to some people sometimes :)
12:20:30 Join Membrillo [0] (n=sam_kill@
12:22:05tucozashridah, which one was that?
12:22:34ashridahah, sketchy on the details, unfortunately.
12:22:38ashridahbeen a long time
12:22:44ashridahi forget if it was the comic or the tv series
12:23:12tucozhehe, I think it was when he added something to the whisky that made him not being able to dring it.
12:24:03tucozno, now I recall. It is the 'picaros' episode. He adds a pill to the food.
12:25:47 Quit Membrillo ()
12:28:16XavierGramiconn: I made a little patch about the zoom thingy we were talking about the other day. Do you want to take a look. Probably it needs more testing
12:28:47 Quit Jungti1234 ()
12:35:57 Quit muesli__ (Read error: 110 (Connection timed out))
12:37:40 Part tucoz ("Leaving")
12:40:11JdGordonwell sports fans... let the games begin!
12:43:48*ashridah starts playing armchair apathy
12:44:17XavierGrnice game
12:45:24JdGordonis the manual actually compiled ever? or do u need to compile it yourself if u want to look at it?
12:45:59BagderJdGordon: my plan is to start offering daily manual builds "soon"
12:46:29Bagderuntil then, building your own is the way
12:46:43XavierGrdaily manual buids? Wow
12:47:10XavierGrI thought that the new manual would be used on the next release
12:47:32Paul_The_NerdAre you going to require .tex files with new plugin patches?
12:48:23BagderI wish that was possible
12:49:34amiconnWhat about bootbox builds?
12:49:42Bagderthat should be added
12:49:44amiconnXavierGr: I can test tonight
12:49:58XavierGrokay I will submit a patch then
12:50:41amiconnBagder: I think we don't need to build all 6 variants. 4 should cover everything that could go wrong
12:50:55 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
12:51:02Bagderyes, I think so too
12:51:14amiconni.e. fm/v2 and ondiosp/ondiofm bootbox have the exact same functionality
12:52:23JdGordondoes any1 know why un/pluggin in usb stops the dircahce thread?
12:53:53JdGordonmy uneducated guess is that the usb kicks off the dircache thread and it doesnt get put back on whichever que the threads are on.. on something like that
12:54:13 Join sharpe [0] (
12:54:39Paul_The_NerdIf I recall it was mentioned that dircache thread stops on a lot of things, as a sort of precaution of some sort.
12:54:57JdGordonos its by design
12:55:05Paul_The_NerdThough I don't really remember. I just remember the USB insert thing being brought up, probably months ago now.
12:56:01Paul_The_NerdWell, the only time in which the dircache should be built on a USB insert is if you're going to charge, right? IE, you're not entering USB mode?
12:56:38JdGordonye, but its not rebuilt after the unplug.. iirc
12:56:59*JdGordon testing...
12:57:09JdGordonhold rec while pluggin to not goto usb mode right?
12:57:27Paul_The_NerdDunno, I don't have an H300
12:57:30 Join KN|stiff [0] (
12:58:02Paul_The_NerdWell, dircache predates the H300 port, so it's probably just not expecting that situation.
12:58:07*Paul_The_Nerd wouldn't really know
12:58:25amiconnThere's something wrong anyway with usb & threads
12:59:03JdGordonhmm.. just did a test.. works fine.. (pluggin in stops the thread, unplguggin restarted it)
12:59:47amiconnIf voice is enabled and you plug/unplug usb, neither voice nor playback work afterwards
13:00:14JdGordonplugging it in so it goes to charging mode does stop the dircache and not restart it on unplug...
13:00:14amiconnIf you try to start playback, rockbox will hang you you'll need the (in)famous paper clip...
13:00:14Paul_The_NerdThat may be a voice issue though
13:01:02Paul_The_NerdOn iPod, if the voice file is loaded at all, playback doesn't work. And it seems to load the file even if voice is set to disabled, because playback won't work until the file is actually removed from the player, so voice has a few problems there
13:01:45amiconnI'm talking 'bot iriver here. Voice is working except some quirks and said problem
13:01:54amiconnOn archos no such problem exists
13:02:10Paul_The_NerdI'm just wondering if the problem's somewhere in Voice rather than in USB by chance.
13:02:47amiconnIf so, then the problem must be in the voice codec thread
13:02:51*JdGordon rememebred the bug report wrong.. the problem is plugging in ac, not usb
13:05:42amiconnIs there a standard name for a mem swap function?
13:08:47 Quit JdGordon (Read error: 104 (Connection reset by peer))
13:14:20 Quit ashridah ("Leaving")
13:14:43 Quit Aditya ("Trillian (")
13:14:47 Join Aditya [0] (
13:17:11 Quit midkay_ (Read error: 104 (Connection reset by peer))
13:17:56 Join midkay [0] (n=midkay@rockbox/developer/midkay)
13:22:02 Join kernel_sensei [0] (n=boris@
13:22:33preglowBagder: in all fairness, bluechip has contributed a fixer for broken mp3s from archos recorders that might be seen as quite useful
13:22:49BagderI forgot that
13:22:57Bagderbut that is true
13:23:10preglownever tested it, though, not owning anything archos
13:23:38BagderI don't use windows and I never record
13:24:04preglowno source code??
13:24:09Bagdernot that I know
13:24:14preglowhow nice
13:24:35preglowi can't imagine a single portability problem to be found in a program like that
13:27:02Bagderonly for >2GB files
13:27:24preglowtrue enough
13:31:53cs_weaselwhups wrong window
13:32:05cs_weaseler. it's offtopic but you still might enjoy
13:33:48linuxstbamiconn: The only standard memory swapping function I know is "swab - swap adjacent bytes"
13:33:54 Quit Rob2222_ ()
13:34:28 Join needleboy [0] (
13:35:40linuxstbPaul_The_Nerd: If you're interested in trying to get voice to work on the iPods, one obvious (I think) bug is the use of #ifdef SIMULATOR instead of #ifdef ROCKBOX_LITTLE_ENDIAN in a few places in apps/talk.c - the voice files are big-endian, and I'm assuming talk.c was written before we had the endian macros.
13:37:03preglowoh, which makes me remember my arm endian swapping routine
13:37:12preglowshamelessly stolen from an arm book
13:38:05Paul_The_Nerdlinuxstb: Thanks
13:39:17 Join damaki_ [0] (
13:40:04 Quit kernelsensei (Connection timed out)
13:40:54 Nick kernel_sensei is now known as kernelsensei (n=boris@gentoo/developer/kernelsensei)
13:41:07 Nick kernelsensei is now known as kernel_sensei (n=boris@gentoo/developer/kernelsensei)
13:41:35 Nick kernel_sensei is now known as kernelsensei (n=boris@gentoo/developer/kernelsensei)
13:41:45 Join damaki__ [0] (
13:46:25tucoz...and to chip in on the naming issue. BC does his fair deal of user support on the mailing list.
13:47:30tucozalthough, I don't understand the paranoia in the thread on the dev-ml
13:50:21amiconnlinuxstb: I mean for swapping memory areas...
13:51:06Bagdertucoz: none of us do
13:52:03pregloware there any standard functions for endian swapping large memory areas?
13:52:58BagderI can't recall any
13:53:05amiconnI don't want to endian swap, but to swap contents
13:53:24amiconn...for swapping msuic & voice codec w/o intermediate buffer
13:53:35 Quit damaki (Read error: 110 (Connection timed out))
13:56:13Paul_The_NerdWhat is .text.UIE ?
13:56:37 Quit damaki_ (Read error: 110 (Connection timed out))
14:00:49 Part tucoz ("Leaving")
14:02:20 Quit moozooh (Read error: 113 (No route to host))
14:04:06 Quit Zagor (Remote closed the connection)
14:06:51 Quit midkay (Read error: 104 (Connection reset by peer))
14:11:48 Join moozooh [0] (n=moozooh@
14:14:07 Join Zagor [0] (n=bjst@rockbox/developer/Zagor)
14:16:03linuxstbamiconn: I would just call it memswap
14:17:22***Saving seen data "./dancer.seen"
14:20:07linuxstbPaul_The_Nerd: For the ARM, we tell gcc to put every function in its own section - .text is where normal (not IRAM) functions go, so .text.UIE is the subsection containing the UIE function.
14:20:25 Part XavierGr
14:20:56Paul_The_Nerdlinuxstb: What does UIE stand for then?
14:21:41Mikachusomething in crt0.S it seems
14:21:52Mikachuor system.c
14:22:07Paul_The_Nerdsystem.c it seems.
14:22:11Paul_The_NerdOr yeah
14:22:16Paul_The_NerdAnyway, that's where the error with voice is occurring.
14:23:17linuxstbWhat's the error?
14:24:05linuxstbUIE is (I think) that deals with the unexpected interrupts - like data abort (unaligned memory accesses) and prefetch aborts (reading from invalid memory locations).
14:24:18linuxstb^the function that deals...
14:24:53LinusNUIE is an unhandled exception, be it an unaligned access or a timer interrupt or whatever
14:28:19Paul_The_Nerdlinuxstb: Ah, well then it's not so helpful since it's just a data abort still
14:29:05linuxstbI had that before - a data abort in the data abort handler....
14:29:14 Join muesli- [0] (n=muesli_t@
14:29:22linuxstbBut I forget what the problem was.
14:31:11linuxstbThe places to look would be the SWCODEC specific parts of the voice UI - the generic parts should be safe because they work on Archos. Only the Coldfire allows unaligned memory accesses.
14:33:32 Join _FireFly_ [0] (
14:33:54linuxstbAt what point do you get the data abort?
14:35:15 Join SereR0KR [0] (
14:36:00Paul_The_NerdThe instant you try to playback music
14:36:50Paul_The_NerdThere's no voice at all with the voice file present, but whether voice is enabled or not, if a .voice exists for your current language, trying to start playback causes a data abort
14:37:23linuxstbSo it doesn't crash if you just browse the menus, it just doesn't talk?
14:38:54amiconnPaul_The_Nerd:UIE == Unexpected Interrupt or Exception
14:40:32Paul_The_Nerdlinuxstb: Exactly.
14:42:32 Quit _FireFly_ (Remote closed the connection)
14:48:01 Join Mong [0] (
14:49:53Monghello? im having diffucluties with the cygwin dev thing
14:51:48LinusNwhat difficulties?
14:52:26Mongwell :D, i dont know what to do
14:52:48Mongim installing it at C;\rockbox\
14:53:21Mongwell .. just did
14:54:08Monganyway when i goto start it(double clicking on the cygwin icon) all i get is a flash a box
14:56:26Mong*floash of the box
14:57:05 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
14:57:20Mongill try again
14:57:48Lynx_mong: do you have c:\rockbox\cygwin.bat ?
14:58:07Lynx_Mong: or c:\rockbox\cygwin\cygwin.bat
14:58:47Lynx_is the icon shortcut correct?
14:58:58Lynx_and if not, can you start the bat file directly?
14:59:59Mongyes shortcut is correct and ive tried it directly
15:00:26Lynx_what happens if you open a command prompt and start c:\rockbox\cygwin\bin\bash.exe ?
15:00:27Paul_The_NerdTry running it from the command prompt, and see if there's any useful output before it terminates.
15:00:45muesli-Paul_The_Nerd to come back to my exif suggestion...
15:01:46 Join webguest84 [0] (
15:02:22Paul_The_Nerdmuesli-: My only concern is button choice, since the ipods only have the wheel (zoom), Up / Down / Left / Right equivalents (scrolling around the zoomed picture) and Select
15:03:07MongLynx; theres nuthinngd in the cygwin\bin\ directory
15:03:16muesli-Paul_The_Nerd sorry, dont know ipods keys. but even if theres no key for ipods left cant be the reason for implementation??
15:03:47Lynx_Mong: then there's something wrong with the cygwin install
15:04:18*Mong reinstalls
15:04:27 Nick Mong is now known as Mongey (
15:04:38Paul_The_Nerdmuesli-: I do think it's a useful function
15:05:52*amiconn is trying microsoft sfu (aka interix) for building rockbox
15:06:15*muesli- feels ignored by amiconn
15:07:02linuxstbPaul_The_Nerd: We could just add a menu to the jpeg viewer for targets with very few keys such as the ipod.
15:07:30Mongeywhats the local package ?
15:08:17amiconnBah, trying sdl sim: can't find SDL-config. Trying X11 sim: Unsupported system: interix, fix configure and retry...
15:08:49Paul_The_Nerdlinuxstb: That's pretty much what I suggested doing earlier. A menu for all targets, then any targets with "extra" buttons, bind them as shortcuts to whatever's most likely to be used quickly
15:09:01*amiconn would like to know how to fix configure...
15:09:22*amiconn remember ender`did some interix work..
15:09:28webguest84Wow, the fwpatcher sure acts strangely. If you minimize it right after starting, then restore and select the input field, it tries to patch
15:10:44Mongeyah i see what i did wrong
15:11:18linuxstbamiconn: What is "uname" displaying for sfu?
15:11:32linuxstbAh, "interix"...
15:11:57webguest84sfu is such a bad name. sfw more like.
15:12:07webguest84usfw, maybe.
15:12:35*amiconn will try to build crosscompilers next
15:12:38Mongeydo i skip zip and unzip ?
15:12:56linuxstbI would try duplicating the Linux entry starting at line 124 of tools/configure and take it from there...
15:13:54LinusNMongey: why would you skip those?
15:13:59Paul_The_NerdMongey: Doesn't it say that you need them on the Cygwin wiki page?
15:15:06*amiconn has high hopes that interix will be significantly faster than cygwin and also offer better integration than colinux
15:16:06*webguest84 eyes line 417-421 of tools/fwpatcher/main.c
15:16:38webguest84Might that cause the patcher to attempt to patch for a number of various actions?
15:20:11 Join alphakiller [0] (n=bohas@
15:20:38linuxstb Does anyone have a picture of the H300 remote?
15:21:37alphakillerhey ... long time
15:21:58webguest84linuxstb: googling gives a result on MR, but that page has mooved
15:22:00Bagderthe lcd version
15:22:01Mongeylinuxstb; ill take one, but i can find one if you want
15:22:14Bg3rlinuxstb see the google link
15:23:17linuxstbThanks all. So how many iriver remotes are there? The h100 lcd remote, the h300 lcd remote, ???
15:23:21Papricawhat do you think now?
15:23:28Bg3rlinuxstb and one non-lcd
15:23:38Bg3r(that comes with the international/eu versions)
15:23:42Bagderyes, the non-lcd h300 one
15:23:47Bagderdoes that work with h100?
15:23:58Bg3rand has only "play/pause/on/off"/ff/rew/vol+/vol- buttons
15:24:04Bg3rBagder should work
15:24:18Mongeyh300 lcd and non lcd
15:24:25tucozWe also need pictures for the manual of some targets. I.e. h300, ipodnano, ipodcolor, ipodvideo
15:24:29tucozsimilar to this
15:24:44Bg3r <= here is the non-lcd remote
15:25:27webguest84that sure looks horribe
15:26:04Bg3rwebguest84 which ?
15:26:20alphakillerLast time I went here, you guys were discussing on how to implement the Arm Co processor support ... has it been done ?
15:27:25webguest84Bg3r: the non-lcd remote. Huge for what it does and very limited functionality
15:28:10Bg3rwebguest84 it's not huge at all
15:28:23Mongeycygwin thing is working
15:30:14amiconnBg3r: My H340 came with no remote at all
15:30:15webguest84linuxstb: has a bunch of pics of the lcd remote
15:30:44Bg3ramiconn iiuc, there are 2 "EU" versions - "normal" and "SE" (aka "Lite")
15:30:47Mongeyamiconn; was it the lite version?
15:30:54linuxstbalphakiller: No - no-one's looked at the cop at all.
15:30:55amiconnSeems so
15:31:10Bg3rnormal comes with dock,remote ... external battery adapter ...
15:31:26alphakillerlinuxstb: do you need someone hacking there ?
15:31:27MongeyEU, is normal :D US is abnormal
15:32:19*alphakiller finished my microkernel project :P and I feel I had a "debt" with you guys ... I really want to contribute ... My mk worked fine on iPod graciously booted by your bootloader :)
15:32:30Bg3rmic as "external mic"
15:32:41linuxstbalphakiller: Definitely. For normal Rockbox use, we seem to be managing quite well with just one processor. IMO, it would be useful to be able to run some of the more advanced plugins on the COP, so we can have simultaneous gaming and music.
15:33:09alphakillerthat's what I thought :)
15:34:04alphakillerI'm updating my CVS
15:34:23alphakillerI didn't see Rockbox playing a song yet :P
15:34:36_ironi_you need images
15:34:49_ironi_i could provide images from ipod video if you want me to
15:34:54_ironi_if needed.
15:35:48alphakillermy ipod is video too :)
15:35:54alphakillerso it worked fine on yours ?
15:36:31webguest84_ironi_: That would probably be very much appreciated. Best is a scan of the player
15:36:42webguest84(stick it in a flatbed scanner)
15:37:06webguest84although I don't know if video is one of those missing
15:37:24linuxstbalphakiller: The new pacman emulator would be a nice example to get running on the cop at the same time as music. Currently, I'm carrying two ipods around with me....
15:38:30amiconnWow, interix supports 384KB command lines (binutils build told me)
15:38:44webguest84That's... a lot
15:38:46linuxstbThat's handy...
15:39:06webguest84in case you need to ls $(find /)
15:39:23alphakillerheheh, that would be nice ;)
15:40:18linuxstbalphakiller: Is your microkernel project using both processors?
15:41:15linuxstbalphakiller: Also, have you noticed the LCD driver improvements that have been made for the 5g since you were last around?
15:41:28alphakillerhmm, no ... I just used the iPod to get some testing on Arm processor. I wasn't really interested on the co processor ... but now I am ... and it has nothing to do with my own OS ... I just want to see the rockbox working indeed
15:41:50_ironi_i dont have a scanner
15:42:06_ironi_does th eipod have a coprocessor?
15:42:19_ironi_video or all ipods
15:42:26alphakillerall ipods
15:42:27crashdthe pp system on chip has 2 arm cores
15:42:34crashdso all of them
15:42:36crashdexcept the shuffle ;)
15:43:07amiconnTeh shuffle might be an interesting target...
15:43:18Mongeyi got some errors when trying to compile
15:43:25Mongey <<errors
15:43:34amiconn...we ought to make voice work properly
15:43:34linuxstbamiconn: Debugging would be an interesting challenge with no display.
15:43:37Bg3rdoes the shuffle has any other interface apart from the USB ?
15:43:57linuxstbamiconn: And also the fact that there is no gcc for it...
15:44:11amiconnWhat architecture is it?
15:44:11alphakillerI thing I was the first one to build on the cygwin, if My memory is right
15:44:25alphakillerI didn't have such error .. but that happenned with old sources
15:44:52 Join webguest24 [0] (
15:44:56 Join damaki_ [0] (
15:45:09_ironi_what would the pouint be of doing _anything_ witht the shuffle
15:45:13_ironi_do explain ;)
15:45:21*Mongey hints at hint in his pic
15:45:23crashdshits and giggles
15:45:26Mongeyalarm :D
15:45:27linuxstbMongey: Maybe you have multiple copies of cygwin1.dll on your PC.
15:45:31*webguest24 gently pokes safetydan
15:46:02webguest84_ironi_: Why not?
15:46:09Mongeythis popped up just before
15:46:25linuxstb_ironi_: It's cheap, and apparently has very good audio quality.
15:46:37linuxstbAnd Rockbox has a voice UI already
15:46:40 Part LinusN
15:46:57Bg3rthis remembers me the old joke with the message box "Format C:?" and buttons "Yes", "Why Not" "Of course"..
15:47:08webguest24Linuxstb, any chance o getting this multifont patch committed
15:47:38_ironi_linuxstb: well then why mod it?
15:47:40Paul_The_Nerdwebguest24: Doesn't the very entry say it actually breaks things too?
15:47:43_ironi_linuxstb: to make it play flac?
15:47:51Mongeyany ideas?
15:47:57webguest24I have no idea, just asking
15:48:06webguest84_ironi_: to make it less of a "shuffle"
15:48:07_ironi_linuxstb: ok voice ui, i see.
15:48:08linuxstb_ironi_: Mainly to give it a voice UI.
15:48:17webguest84That is, to give more control over what is played
15:48:20_ironi_and directory baes support and so on
15:48:26Paul_The_Nerdwebguest24: It says right in there that it breaks the recording screen drawing...
15:48:34crashdthere was some work done on the shuffle iirc
15:48:35_ironi_i would focus on the ipod nano/video :)
15:48:39crashdmaybe there's links on the ipl page
15:48:45amiconnWhat formats does the shuffle play? What cpu arch is it?
15:48:46tucozhow would you go forth porting rockbox to the shuffle. Debugging will be a hard task
15:48:56crashdhit and miss!
15:48:57 Quit SereR0KR ("XChat Aqua")
15:49:03crashdamiconn: mp3 and aac i guess
15:49:08linuxstbIt has a motorola DSP
15:49:13webguest24Paul, so bowl me over for not knowing, is there any harm in aslking, you huys are the developers after all
15:49:15Zagortucoz: isn't there a serial port in the dock interface?
15:49:26tucozI don't know. Maybe it is.
15:49:35Paul_The_Nerdwebguest24: I'm just saying, did you read the patch description there? And, I'm not a developer.
15:49:36 Join SereR0KR [0] (
15:49:40linuxstbZagor: On the other iPods, yes. I don't know about the shuffle.
15:50:27webguest24Paul, then why are you replying if you aren't a developer, sp what it has bugs, that's for prograamers to fix if they want to, thats what I'm here to find out
15:50:39tucozIs there a led somehere on the shuffle? In that case, the same approach as amiconn used when fixing the compression could be used ;-)
15:51:24Paul_The_Nerdwebguest24: I'm replying because I know enough to know that if a patch knowingly introduces bugs then generally speaking someone has to fix those bugs before it would get committed. It's common sense that you don't intentionally add bugs.
15:51:42*webguest24 shakes his head
15:51:45 Part webguest24
15:51:49linuxstbwebguest24: As Paul said, the patch itself isn't good enough to commit, so it needs someone to finish the job.
15:52:03amiconnurgs, DSP core...
15:52:14*Paul_The_Nerd sighs.
15:52:21webguest84thin skin
15:52:43tucozoh, I thought you left linuxstb
15:52:45Paul_The_NerdApparently I'm not allowed to have an opinion, since I'm not a dev.
15:52:56linuxstbtucoz: No :) I was saying farewell to #24
15:53:03*Paul_The_Nerd has maybe 2 or 3 lines of code somewhere in CVS now.
15:53:17 Join gracchus [0] (
15:53:34 Quit SereR0KR (Client Quit)
15:54:08 Join SereR0KR [0] (
15:54:36 Quit SereR0KR (Client Quit)
15:54:54Mongeyi have 2 cygwin1.dll 's , now which to keep C:\windows C;\rockbox
15:55:02 Join SereR0KR [0] (
15:56:32linuxstbMongey: I don't know for sure, but I would probably copy the one in C:\rockbox over the top of the one in C:\windows.
15:56:42Paul_The_NerdMongey: Which one's newer?
15:57:18amiconnMongey: Did you use the devkit before?
15:57:25Mongeythe rockbox\ one
15:58:01amiconnI would keep the one installed by cygwin, which is probably the one in windows/
15:58:18 Quit SereR0KR (Client Quit)
15:58:21amiconnNote: newer date doesn't necessarily mean newer dll version
15:58:50 Join SereR0KR [0] (
15:58:54Paul_The_NerdAmiconn: I do not have a cygwin1.dll in C:\windows\
15:58:56amiconnThe file properties should show a 'version' page for cygwin1.dll
15:59:03 Quit damaki__ (Read error: 110 (Connection timed out))
15:59:13Mongeyamiconn; i know
15:59:45Mongeythe rockbox one is 1005.19.0.0
15:59:50Mongeyi think#
16:00:20Mongeyeh woops
16:00:23Mongeyboth the same
16:00:59Mongeymeh ill delte the rockbox
16:01:07Paul_The_NerdI have no cygwin1.dll in C:\Windows, *or* in my base cygwin install folder
16:01:21Paul_The_NerdMines in C:\rockdev\bin\
16:01:51Paul_The_NerdSo, that's where cygwin's installer seems to put it by default.
16:02:04amiconnMine is in c:\cygwin\bin
16:02:06Mongeymine too
16:02:27*Mongey deleted the windose one
16:02:27Paul_The_NerdThat's the one you should keep, most likely
16:03:00Paul_The_Nerd"That" was C:\rockbox\bin, by the way.
16:03:26amiconnThe shuffle has just a battery led. Maybe software controllable, maybe not.
16:03:51 Quit DJ_Dooms_Day ("Trillian (")
16:04:15 Nick kclaf is now known as AwAyhOmO (
16:04:45*safetydan awakes, notices that webguest24 has left, goes back to sleep/work
16:05:47 Join quobl_ [0] (n=quobl@tor/session/x-5a69cf12a787c21e)
16:07:07 Quit sharpe (Read error: 110 (Connection timed out))
16:09:28Mongeyi got a load of things coming up after i said make file, command not found
16:10:04 Quit needleboy ()
16:12:40 Join XavierGr [0] (
16:15:29linuxstbamiconn: I think preglow has worked (is working?) with similar DSPs to the one in the Shuffle.
16:15:37 Quit Mongey ()
16:17:23***Saving seen data "./dancer.seen"
16:19:15 Quit damaki_ (Remote closed the connection)
16:19:54 Join DreamTactix291 [0] (
16:19:54 Quit DT291 (Read error: 104 (Connection reset by peer))
16:20:06 Nick AwAyhOmO is now known as kclaf (
16:22:09XavierGramiconn:are you here?
16:22:14 Quit quobl (Remote closed the connection)
16:26:12 Part tucoz ("Leaving")
16:27:27 Join Siku [0] (
16:32:50amiconnXavierGr: Somewhat.
16:36:36XavierGrI have the patch if you are interestedI have a small patch about the zoom thing. Do you want to take a look or should I leave this to Bger?
16:36:58XavierGropps sorry about the weird sentence
16:37:45 Quit Sinbios (Read error: 104 (Connection reset by peer))
16:37:50 Join TCK [0] (
16:38:03XavierGrit will not fix the problem you were saying about the voice file though. I didn't look at this yet
16:40:48 Join Sinbios [0] (
16:44:23XavierGranyway amiconn: here is the patch if you have some time to test tonight. I will try to make some more tests to be sure.
16:45:11 Join _FireFly_ [0] (
16:47:58amiconnBleh, I can't build cross-gcc on interix; I get syntax errors from 'sh'
16:48:10amiconnI need to find out how to replace sh with bash
16:48:18amiconnI already have bash installed...
16:48:27_FireFly_amicon symlink :)
16:48:46_FireFly_amiconn i mean
16:53:15linuxstbOr if symlinks aren't supported, "cp bash.exe sh.exe"
16:53:46linuxstbAt least on Linux, I've never seen /bin/sh be anything other than a symlink to /bin/bash
16:55:20amiconn. /bin/sh is a link to ksh on interix
16:55:33*amiconn changes
16:59:16amiconninterix standard installation features csh and ksh, but no bash
17:16:34 Quit safetydan ("Leaving")
17:16:47 Part Paul_The_Nerd
17:18:47 Join savagenator [0] (
17:21:01 Quit Zagor ("Client exiting")
17:23:18 Quit savagenator (Client Quit)
17:27:58 Join dpassen2 [0] (
17:31:35linuxstbI've been fighting with the plugin lib's configfile system, and just spotted the problem. The comments in configfile.h say that for TYPE_ENUM, the "max" field is the max value for the enumeration. But in the configfile_load() function, the for loop is for (i=min;i<max;i++) - so if a setting is set to the last value in an enumeration, it's not loaded.
17:32:06linuxstb(sorry, the for loop is from i=0;i<max
17:32:32 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
17:32:33amiconnI figured that as well when adding config support to oscilloscope
17:32:42tucozamiconn, does the ondio have a rtc?
17:33:01amiconn..and in addition, you can't use 'enum xxx' for the variable itself, you have to use in
17:33:07amiconntucoz: no
17:33:12tucozok, thanks
17:33:39*amiconn points to
17:33:45tucozhehe, I was lazy
17:34:21amiconnlinuxstb: ...unse 'int'
17:35:02linuxstbWhat do you mean?
17:36:14amiconnYou can't have enum xyz my_enum_setting and then use &my_enum_setting in the configfile struct
17:36:28amiconnIt must be int my_enum_setting
17:36:35linuxstbThat's not my problem though - I am using ints.
17:37:25 Quit TCK (Read error: 104 (Connection reset by peer))
17:37:27linuxstbThe problem was that used an int with values (for example) of 0-3, but setting min and max to 0 and 3 doesn't work. I need to set max to 4.
17:37:41linuxstbi.e. the comments in configfile.h and the code in configfile.c are conflicting.
17:38:02amiconn[17:31:54] <amiconn> yes
17:38:08amiconn[17:32:09] <amiconn> I figured that as well when adding config support to oscilloscope
17:38:09tucozWhich targets have the Car adaptor mode
17:38:51linuxstbamiconn: I know - but why did you then tell me to use enum?
17:38:53amiconntucoz: archos player & all recorders, but it only works correctly on the recorders, and only when flashed
17:38:59linuxstbI mean not use enum...
17:39:09 Quit _FireFly_ (Remote closed the connection)
17:39:11tucozamiconn, thanks.
17:39:34amiconnlinuxstb: Because that was another mistake I did...
17:39:36Mikachulinuxstb: "and in addition"
17:39:54 Quit dpassen2 ("CGI:IRC (EOF)")
17:39:54amiconnImho one should use an enum for enum, that's what it's meant for
17:39:55 Join _FireFly_ [0] (
17:40:01linuxstbAnyway, should I correct the comment in configfile.h so that it matches configfile.c, or the other way around? I think correcting the comment will be easier.
17:40:43 Quit coob (Read error: 110 (Connection timed out))
17:40:58amiconnI wanted to ask linus about it, it's his work..
17:41:06amiconnWhat's also missing is bool support
17:41:37linuxstbI was thinking it could also be nice to somehow merge it with the set_option() function.
17:42:09linuxstbIn pacbox, I now need to define the settings twice - once for the configfile, and once for set_option()
17:43:05markunamiconn, linuxstb: do you guys know what the slowest response time is before a UI feels laggy?
17:45:43linuxstbWhy are you asking?
17:46:45markunI was trying to put some more sleeps in the rockbox code
17:47:20alphakillerdoes any of you has a finnancial support to develop rockbox ? or everybody does it in their respective spare time ?
17:47:37markunI made the wps thread sleep if the while(1) loop takes less than x ticks
17:48:11tucozalphakiller, it's totally voluntary. But new hardware etc is sponsored by donations (see the pay-pal on the front page)
17:50:30 Quit _FireFly_ ("Leaving")
17:55:03amiconnmarkun: You don't need to put sleep() in the ui thread (and shouldn't)
17:55:24markunamiconn: why not?
17:55:26amiconnbutton_get_w_tmo() does the same as sleep() unless a button is pressed
17:55:55amiconnsleep() would make the interface laggy with no reason.
17:56:30markunbattery could be a reason..
17:57:48webguest84cvs -q annotate $FILENAME 2>/dev/null |grep "^[0-9\.]" |cut -b 15-22 |sort |uniq -c |sort -n #contribution scorecard! (assuming the output of cvs annotate is predictable)
17:59:02markunamiconn: rockbox used 100% of my CPU in the simulator with the WPS showing the peak meter, that's why I tried to slow it down a bit.
17:59:26amiconnmarkun: button_get_w_tmo() does the same as sleep() unless a button is pressed, i.e. it also sleeps the CPU
17:59:43amiconnThe peakmeter is a different thing, it needs to be drawn fast
18:00:48markunalso, one of the threads is not sleeping when playback is paused
18:04:42markunamiconn: thanks for pointing out button_get_w_tmo, don't know why it doesn't restrict the CPU usage in the SDL sim
18:05:52Mikachulinuxstb: there's a line with a tab in pacbox.c
18:08:36Mikachulinuxstb: 302
18:12:13linuxstbMikachu: Thanks. It will go in my next commit.
18:16:58 Quit alphakiller (Read error: 104 (Connection reset by peer))
18:17:24***Saving seen data "./dancer.seen"
18:17:58linuxstbmarkun: Have you managed to transfer any music to your gigabeat from FreeBSD yet?
18:36:27preglowanyone care to take a guess how probably it is an ipod will function with no battery and only external power? :>
18:38:38 Join Hyperactive [0] (n=OliverKa@
18:39:01Hyperactivecan someine help me plz?
18:40:27preglowfirst you've gotta ask a question
18:41:49Hyperactivei have iriver 3400
18:42:08Hyperactivei download rockbok
18:42:32Hyperactivebut i cant see inglist all i see is ._~
18:42:51Hyperactivelike this. hwo can i change the long?
18:44:39amiconninterix doesn't like ipod_fw.c
18:44:58amiconn...and the *only* crosscompiler I managed to build is arm-elf-gcc
18:45:40preglowHyperactive: i don't understand what you mean
18:45:42amiconnipod_fw.c:290: error: `opterr' undeclared (first use in this function)
18:46:04Hyperactiveok. i install rocnok on my iriver 340.
18:46:07amiconnAnyone knwoign what this 'opterr' (and also optarg, optind and optopt) is?
18:46:50Hyperactivebut i cant see the menu..the letters are like this: ... only pionts.
18:47:21Hyperactivewho can i change the language to inglish?
18:48:21webguest84hold down record when you're booting and it'll reset all settings
18:52:17preglowso we've got fonts bundled that haven't got ascii characters?
18:52:25Hyperactivethe hebrew funts wont work 4 me.
18:52:27preglowi think we really need to clean up our bundled fonts
18:52:59tucozpreglow, I agree. Get rid of the 7bit fonts at least
18:53:04tucozLike chicago 12
18:54:44webguest84Maybe there could be a check each time a font is selected whether it has the characters needed for the current language
18:54:54webguest84and the same for selecting new languages
18:56:13webguest84binlang could build a string containing all used characters
18:57:11 Quit webguest84 ("CGI:IRC (EOF)")
18:58:30 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
19:05:46markunlinuxstb: no, it didn't work. And I still don't know if it's a FreeBSD specific problem
19:06:36Hyperactivewhay does all my setings changes after aboot?
19:06:47Mikachudid you upgrade rockbox?
19:07:00Mikachuare you holding down the 'reset settings' button? :)
19:07:45Hyperactiveis this uptodate?
19:08:02Mikachuif you change from one version to another, sometimes the settings are reset the first time you start the new version
19:08:13Mikachuyou want to save your settings in the Manage Settings menu
19:08:23t0masMikachu: that's when the config block version changed
19:08:23Mikachuthen you can load that .cfg file after the settings are reset to restore them
19:08:49t0masdo you can see from the CVS commit messages when it happens :)
19:08:49Mikachuthat's the one
19:09:23Hyperactiveyou where right..
19:09:35Mikachuanyway it's a good idea to keep a .cfg handy
19:12:47 Quit muesli- ("ich will Kühe!!!")
19:13:15PapricaOK, tetrox works on: archos recorder/ondio, ipod color/4gen/3gen/nano, iriver h300,h100
19:13:26MikachuPaprica: i'm done with school stuff now
19:13:53Papricai will compile for you
19:14:01Mikachuplease don't
19:14:16Mikachui don't like precompiled stuff on principle :)
19:14:57Papricai'll send the source and the bitmap
19:17:49Mikachuif you wait 3 minutes, don't expect me to type /dcc get in 5 seconds :)
19:19:04Papricaoh, forgot the iaudio x5 on the working targets =]
19:19:12 Join safetydan [0] (n=dan@rockbox/developer/safetydan)
19:20:28Mikachuwow, that's really short :)
19:20:54Papricawhat short?
19:20:58Mikachuthe code
19:21:08Mikachuonly 687 lines
19:21:26Mikachubrickmania is 1844
19:21:26*amiconn now managed to build at least one target on interix
19:21:40amiconnIt's not really worth the hassle...
19:21:43Papricabrickmania is lonnnnnnn
19:21:53Mikachuand a bit uncommented i felt
19:22:18amiconnIt's not measurably faster than cygwin
19:22:35PapricaMikachu, dont talk, try it
19:23:02Mikachui would if linux' usb mass storage driver worked better :)
19:23:55Mikachuokay, seems to work with replugging the ipod
19:24:56Mikachuhang on
19:25:05Mikachu(or do other things in the meantime)
19:26:05Mikachuah, forgot to add tetrox.c to sources so it used my old one :)
19:26:12Mikachui should make clean more oftne
19:29:00 Quit SereR0KR ("XChat Aqua")
19:29:25 Quit Hyperactive ("nothing can break me like a woman can.")
19:31:30Mikachuhrm, it has that annoying feature that you can't rotate at the edge of the screen
19:32:31Papricait was like this
19:33:12 Join SereR0KR [0] (
19:33:55 Join Rob2222 [0] (
19:34:16 Join Paul_The_Nerd [0] (
19:35:06 Join Rondom [0] (
19:36:37PapricaMikachu, it works fine?
19:37:17preglowsafetydan: sproing?
19:40:55MikachuPaprica: yes, except for it restarting when i press select
19:41:04Mikachuthat could be construed as slightly annoying
19:41:36Papricaok i can replace it with the drop
19:41:41Mikachu#define TETRIS_DROP BUTTON_MENU
19:41:50Mikachui think reversing those two will be better
19:42:21Mikachuand i think the names TETRIS_UP and TETRIS_NAV for rotating left and right are a bit weird too :)
19:42:58Papricahaha you dont have another keys =]
19:43:29Papricamy wwrong
19:43:43Papricai think you talking about the keys
19:44:00Papricayou are*
19:44:05Mikachuanother suggestion is making TETRIS_RESTART into (BUTTON_MENU|BUTTON_REL)
19:44:20Mikachuthat way you can press menu+select to exit without looking at the splash 'restarting' for two seconds
19:44:36Papricawhat is BUTTON_REL?
19:44:44Mikachuit's sent when a button is released
19:44:52Papricaoh ok
19:44:53Mikachuso when you release menu, the game will restart
19:44:56Mikachuinstead of when pressing
19:45:20safetydanpreglow, sprong?
19:45:41safetydanWhat is the response to that? sprang? sproinga? spong?
19:45:47preglowgod knows
19:45:52preglowwhat with the cordic code?
19:45:55preglowthe pastebin url expired
19:46:09safetydangive us a sec
19:46:27preglowcan do
19:46:46MikachuPaprica: i have never in my life seen a tetris clone where the pieces come out in random orientation
19:47:19Papricai didnt write the code..
19:47:27Mikachuwell, you can change it :)
19:47:35amiconnMikachu: huh? What do you mean?
19:47:38 Join obo [0] (
19:47:53safetydanas I said, it's a bit rough, and I've no idea of performance
19:47:55Mikachuamiconn: when a piece appears, it is usually always oriented in the same way
19:47:57 Quit SereR0KR ("XChat Aqua")
19:48:05Mikachuwhich way depends on the clone, but in tetrox it's different every time
19:48:07Mikachuit's very confusing
19:48:08 Join SereR0KR [0] (
19:48:17preglowsafetydan: i'm just gonna test it anyway
19:48:21preglowif it works, it's good
19:48:31amiconnI only know variants where the orientation is random...
19:48:58linuxstbPaprica: I would be happy to test Tetrox on my 5g.
19:49:20safetydanoh, and the input angle range is limited to -pi/2 to +pi/2
19:49:35Papricalinuxstb, im working on the background bitmap now for the video
19:49:35Mikachuamiconn: should the picture in Next Piece at least agree with the orientation?
19:49:50PapricaMikachu, ok, you right
19:49:52linuxstbPaprica: Ah, OK.
19:50:03Papricai'll change it
19:51:56preglowsafetydan: we should just support from 0 to 2pi anyway
19:52:00Mikachua pause button would also be nice :)
19:52:08Mikachubut other than those things, i like it very much
19:52:28linuxstbWhat are the ipod key mappings? Is the clickwheel used?
19:52:32Mikachuall the colors are correct! i have only seen one clone where they are. and of course by correct i mean the colors used in that clone
19:52:35Mikachulinuxstb: yeah, for rotating
19:52:42Mikachuit could perhaps do with being a tad less sensitive
19:52:52Mikachuand i've already lowered my delta from 4 to 10
19:52:59Mikachuthat would actually be an increase
19:53:09linuxstbI've just tried RockBlox - the button mappings are a bit crazy (MENU exits), but it uses the wheel to move left/right which works well.
19:53:29Mikachui wouldn't like that i think
19:53:37linuxstbGive RockBlox a go.
19:54:01linuxstbI was thinking the wheel could move left/right and NEXT/PREV would be rotate right/left.
19:54:11Paul_The_NerdI was gonna suggest wheel for L/R movement too.
19:54:21preglowit's the most logical
19:54:58Mikachuth esensitivity is good, but i need to move two laps to move from one edge to the other...
19:55:29linuxstbMikachu: Are you using the standard button driver?
19:55:31Paul_The_NerdBy the way, I think it's time to enable battery status on at least the iPod Nano. I don't know if anyone's tested it on others, but the reporting seems at least reasonable.
19:55:41Mikachuwith the exception noted above
19:55:49 Join SereRokR [0] (
19:55:50preglowsafetydan: if nothing else, this thing should be really fast on arm :-)
19:56:18Mikachulinuxstb: but if it moved any faster i wouldn't be able to position it accurately
19:56:31Mikachulinuxstb: remember the nano wheel is a bit smaller than the 5g one (i think)
19:56:57linuxstbOK, but Rockblox and the standard button driver is perfect on the 5g.
19:58:29Mikachunow that i think about it three of the colors are "wrong" :)
19:58:30safetydanpreglow, good to hear
20:00:20 Join alphakiller [0] (n=bohas@
20:00:45alphakillercan anyone answer me an ipod question ? heheh
20:00:54Paul_The_Nerdalphakiller: Ask it, and we'll find out
20:01:01alphakillerhow to upload a picture onto an ipod video
20:01:18Mikachujust put it there
20:01:21linuxstbcp file.jpg /mnt/ipod and then open it using the lovely Rockbox jpeg viewer.
20:01:35alphakillerMikachu: where ?
20:01:41Paul_The_NerdWherever you want it.
20:01:42Mikachuon the ipod
20:01:51linuxstbalphakiller: Are you talking about Rockbox or the Apple firmware?
20:02:17alphakillerapple firmware
20:02:42Paul_The_NerdAaah, might've wanted to specify that, y'know, with what the channel is and all
20:02:47Paul_The_NerdWith the apple f/w, you have to use iTunes.
20:02:54Mikachuor gpixpod
20:03:04Paul_The_NerdOr some other thing that deals with their database
20:03:18*linuxstb shrugs
20:03:31Paul_The_NerdI have no idea why you'd rather view it in the apple firmware though
20:03:46Mikachuit's like rockbox but you can't zoom, but it still saves the fullsize picture
20:04:27preglowsweet god
20:04:28preglowi _HATE_ adsl
20:04:30Paul_The_NerdMikachu: It *does* have a slideshow mode, for extra quick battery draining.
20:04:38preglowsafetydan: this thing is leaps and bounds more accurate than my thing
20:04:40Mikachuoh yeah, forgot about that
20:05:06lostlogicdo we know why ipod audio cuts out at volume < -57dB?
20:05:10lostlogicon 5g?
20:05:21linuxstbBecause the range of the DAC is -57dB to +6dB
20:05:26safetydanpreglow, \o/
20:05:45Paul_The_NerdMan, and when it's silent I often listen to my music at about -53...
20:05:50safetydanit should be since it's the same algorithm the goold ol' 387 FPU used :)
20:06:03safetydangood even
20:06:05lostlogicI'm listening to -57 now and it's too loud... time to use preamp more
20:06:30linuxstbFeel free to play with the DAC....It may not be initialised perfectly.
20:07:06preglowsafetydan: what i'm wondering about, though, is if it's possible to increase the precision to the full 32 bits
20:07:31preglowsafetydan: right now it's only precise to the 30th bit, two of the top bits of the answer goes unused
20:07:45Mikachuis the scale really different from the nano? i usually have mine at -15
20:07:59linuxstbYes, the Nano has a different DAC.
20:08:15safetydanpreglow, if you up the iteration count it should increase the precision... but I'm not certain if there's an inherent level of inaccuracy
20:08:34Paul_The_NerdShouldn't -15 on Nano be the same as -15 on a 5G?
20:08:46preglowsafetydan: shouldn't be, the current problem is that the output format is the same precision as the input format, which needs to have some integer bits to represent angles bigger than 1
20:09:03preglowsafetydan: the result, however, never has any values higher than +/- 1
20:09:07preglowso it doesn't need any int bits
20:09:10amiconnlinuxstb: apps/plugins/pacbox/z80.c is rather demanding at buold time
20:09:43 Quit KN|stiff ("( :: NoNameScript 3.81 :: )")
20:09:50preglowsafetydan: but yeah, great work, this'll fix the problems we're having
20:09:51amiconnIt can't be compiled with less than ~500MB of free (virtual) memory
20:10:16linuxstbIt does have over a thousand functions in it...
20:10:16preglowahhh, gcc
20:10:28preglowthe crack addict compiler
20:10:49linuxstbI am planning to try different z80 cores if I can find any suitable candidates.
20:10:49amiconnShouldn't be an issue nowadays, but I ran into that problem when testing interix in a VM
20:11:23amiconnThis was for arm target (iPod nano)
20:11:24linuxstbBut I don't think that code is that inefficient - I think the video-related code is the bottleneck in pacbox.
20:11:32 Join petur [0] (i=petur@rockbox/developer/petur)
20:12:11safetydanpreglow, woo! good to hear. Should be able to turn that same cordic rotation stuff in to a general purpose math library
20:12:51safetydanWhat was the other function needed asinh?
20:12:54Mikachushould rb->splash(0, ...) from a plugin wait for a button press or just go away instantly? it seems to do the second
20:13:06linuxstbWhich reminds me, does anyone have any ideas why putting a very frequently used array in pacbox (the palette lookup table) into IRAM would actually slow pacbox down on the iPod?
20:13:24linuxstbMikachu: It returns instantly, but it's up to the calling function to clear the display (IIUC).
20:13:25preglowsafetydan: and to be quite honest, i don't think the last bits matter for the eq
20:13:32preglowsafetydan: i shall do some tests in octave and see
20:13:44Mikachumaybe i can just put a get_button thing after
20:13:51linuxstbMikachu: Yes, that should work.
20:14:15lostlogicis there a way to write replaygain tags to an existing mp3 file with lame (or any other linux software)?
20:14:17preglowlinuxstb: i take it you are absolutely certain that is the case?
20:14:32preglowi don't like the sound of that at all
20:14:37Mikachulostlogic: i haven't found any.. only mp3gain which modifies the mp3 frames
20:14:56 Quit alphakiller ()
20:14:58linuxstbpreglow: Yes - only very slightly slower, but still slower.
20:15:14linuxstbI'm wondering if the IRAM is cached.
20:15:21preglowwhy should it matter?
20:15:27preglowit should
20:15:29Paul_The_NerdMikachu: If you tell Mp3gain to scan, and then never tell it to apply, if I recall it writes tags but doesn't modify the frames.
20:15:29linuxstbBecause it leaves less space for other things.
20:15:39preglowi meant for your performance problem
20:15:40*lostlogic tries this with mp3gain
20:15:52Paul_The_NerdThough I've only used the windows version
20:16:01preglowbut if the cache is direct mapped, it might very well have an impact
20:16:12MikachuPaul_The_Nerd: there's no such option in the command line version, and i think that sounds suspicious
20:16:27Mikachualso, i found a semi-bug
20:16:28amiconnlinuxstb: Maybe it's a code cache aliasing problem?
20:16:29 Quit SereR0KR (Connection timed out)
20:16:36Mikachuplugin.h doesn't export button_hold on ipods
20:16:43amiconnMikachu: splash(0,...) is supposed to return instantly
20:16:47Paul_The_NerdMikachu: Fair enough. It was my very first experience with replaygaining at all, and being a windows user, I soon moved to foobar and never dealt with it again.
20:17:03amiconnsplash() will also never clear the display itself
20:17:07Mikachuamiconn: i had a vague memory of someone saying it should wait, but maybe it was something entirely different
20:17:25Mikachujust making the game over thing in tetrox not disappear and start a new game after 1 second
20:17:26preglowamiconn: code cache aliasing?
20:17:26***Saving seen data "./dancer.seen"
20:17:27 Join tvelocity [0] (
20:17:31linuxstbMikachu: It should do now - I added it a few days ago with pacbox.
20:17:45Mikachuhm, maybe something weirded out
20:17:48Paul_The_Nerdlinuxstb: That broke the Doom patch by the way. :)
20:17:51Mikachuthe line is close to my wheel_status thing
20:18:16lostlogicMikachu: it seems to work with the commandline version as well
20:18:36amiconnpreglow: Moving something into/out of iram might move functions, and make functions 'fight' for cache lines
20:18:37Mikachulostlogic: are you sure you don't mean the undo tags?
20:18:49amiconnI don't know how the iPod cache is organised though
20:18:56lostlogicMikachu: they are the same thing, and it put the right stuff in the tags
20:19:02Mikachuah, okay
20:19:14Mikachuit messed up all my unicode tags so i always run with /s s
20:19:40 Quit tvelocity (Client Quit)
20:20:07amiconnIt may also be that iram isn't cached and iram is also not single-cycle on pp
20:20:15Mikachulinuxstb: i just checked out a clean copy and it only has #if CONFIG_KEYPAD == IRIVER_H100_PAD as far as i can see
20:20:17 Join tvelocity [0] (
20:20:24 Quit didj (Connection timed out)
20:20:29Mikachulinuxstb: oh so sorry, it is just in another place
20:20:44linuxstbMaybe I should have merged them - but I don't like breaking the plugin API...
20:20:59preglowamiconn: ahh, right, that was what i meant too
20:21:01lostlogicdoesn't calculate album gain it looks like though.
20:21:13preglowamiconn: we don't really know what associativity the cache has
20:21:53preglowsafetydan: actually, just upping the iteration number did help...
20:22:02Mikachuhooray, now it pauses while the hold button is on
20:22:30amiconnpreglow: You can test whether iram is cached or not, and whether it's single-cycle
20:22:48preglowamiconn: whether iram or cache is single-cycle?
20:22:51amiconnIf it's single-cycle, testing whether it's cached will be harder of course
20:22:57preglowsafetydan: so, ok if i modify it a bit and commit it?
20:23:04safetydanpreglow, go right ahead
20:23:09Mikachulinuxstb: what #ifdef should i put if i want to use button_hold in a plugin?
20:23:11amiconnpreglow: Whether iram is single-cycle, postulating that cache is
20:23:32preglowi think both are single-cycle
20:23:35preglowaccording to ipl people
20:23:40linuxstbI've just taken all of the pacbox data out of IRAM (which is all the Z80-related data such as the RAM/ROM arrays), and on my 5g, I'm getting almost identical speed to the version with everything in iram.
20:23:51preglowipod has cache, yes
20:24:01preglowwe should try doing a rockbox build with no iram whatsoever
20:24:05preglowmight be very interesting
20:24:26amiconnMight also be interesting to test on PP5002...
20:24:35linuxstbIt would make using the COP a lot easier - all that IRAM to spare.
20:24:43preglowlinuxstb: exactly
20:24:51Mikachuanyone: what #ifdef should i put if i want to use button_hold in a plugin?
20:25:25lostlogicMikachu: ok, I take back all the shitty things I've said about mp3gain, it actually does do what is needed with a simple script wrapping it to feed it one album at a time.
20:25:28 Quit tvelocity ("Ex-Chat")
20:25:33lostlogicPaul_The_Nerd: thanks for the tip.
20:25:43linuxstbMikachu: Find the definition of that function - probably in button.c
20:25:56linuxstbAnd then use whatever is surrounding that function.
20:26:16linuxstbWe should probably fix plugin.h to use the same.
20:26:37Mikachuit's spread out with lots of different keypad definess
20:26:44linuxstb:) Just saw that...
20:26:46preglowsafetydan: you tried adjusting that K constant?
20:26:46amiconnPaul_The_Nerd: What happened to rockdoom?
20:26:51Mikachuso maybe not a big success
20:26:57amiconnWhat happened to rockdoom in general??
20:26:59Mikachuwhat do you do in pacbox?
20:27:06lostlogic5 bits changes when the hold switch is switched in 5g GPIO.
20:27:23linuxstbAh, button.h surrounds it by #ifdef HAS_BUTTON_HOLD
20:27:27Paul_The_Nerdamiconn: Doesn't build at the moment for most people, because of the changes to plugin.h/.c. I'm not sure if that's the only thing that's breaking it, I'm having a little trouble untangling my custom build. =/
20:27:32safetydanpreglow, no, I did calculate it though... but looking it again it might off by about 3
20:27:53linuxstbMikachu: I've actually taken it out of pacbox now. Previously it was just used for the ipods, because there were no other spare keys.
20:28:04Mikachuhow do you enter the menu now?
20:28:18Mikachuah, of course
20:29:02preglowsafetydan: there's some small discrepancy here that i don't like and i suspect it's rooted there
20:29:21preglowand why do you mention 2.16 fixed point? you're using s2.29 fixed point
20:29:21Mikachulinuxstb: in button.h, wouldn't it make sense to combine the two ifs for button_hold?
20:29:40Mikachulinuxstb: like #if defined(HAS_BUTTON_HOLD) || CONFIG_KEYPAD == IRIVER_IFP7XX_PAD
20:30:08 Join tvelocity [0] (
20:30:13linuxstbI would prefer not to touch the ifp7xx code - it's still in an early state.
20:30:40Mikachuokay, i have no idea about other targets :)
20:30:41linuxstbAnd plugins aren't feasible on the ifp7xx anyway.
20:30:47linuxstb(it only has 1MB RAM)
20:30:55linuxstbForget that....
20:31:12linuxstbBut plugins are disabled now anyway.
20:31:22amiconnPaul_The_Nerd:plugin api clashes shouldn't be hard to fix. I mean, what's the general status of rockdoom
20:31:37*amiconn wonders whether/when kkurbjun will return...
20:31:45linuxstbI think the version on the patch tracker was very close to be committable.
20:31:54Paul_The_Nerdamiconn: Oh. Few warnings, and it still doesn't fit in the plugin buffer. Other than that, it's pretty good, I'd say.
20:32:08amiconnlinuxstb: Not that much different from the archoses (with 2MB RAM)
20:32:10linuxstbIt needs a little work to bring the plugin size down on ARM, but I think it fits in 512KB on Coldfire.
20:32:17Paul_The_NerdYeah, it fits on coldfire
20:32:24linuxstbamiconn: I know - that's why I retracted my statement :)
20:32:27Mikachui think i've had mine on 28MB for a while
20:32:43linuxstbI'm not sure if it compiles in the sim either.
20:33:44linuxstbamiconn: Are you thinking that someone else should try and clean rockdoom up and commit it, rather than wait for kkurbjun to return?
20:34:04 Join Shadowarrior13 [0] (
20:34:04Paul_The_NerdI'll at least get it running again.
20:35:06safetydanpreglow, yeah the comment is wrong about the format
20:35:08amiconn-logbot_- kkurbjun ( signed off 13 days and 11 hours ago (Read error: 110 (Connection timed out))
20:35:25safetydantypo on my part
20:36:46preglowand is it (1/2)^n or 1/(2^n) ?
20:37:09preglowi'm not all that when it comes to operator precedence
20:37:16lostlogiclinuxstb: do buttons not even cause interrupts when the hold switch is on hold on ipod 5g?
20:37:21Mikachupreglow: that's the same thing
20:37:25Mikachupreglow: 1^n = n
20:37:42Mikachu1^n = 1
20:39:07preglowi just realised how dependent i am on ordinary maths notation :)
20:39:48XavierGramiconn: do you know if it will be easy to adjust the jpeg viewer to show the picture as much as it can without zooming. (default view of a picture can vary from a tiny to a big one (unzoomed))
20:39:50safetydanEven calculating it out to n = 64 the value of K doesn't change much
20:39:56preglowsafetydan: i just found out
20:39:56safetydanat all even
20:40:03preglowit converges well before n = 16
20:40:24safetydananyway, going to watch a movie
20:40:40preglowhave fun
20:40:41XavierGrwhich one? :)
20:40:42 Join BHSPitLappy2 [0] (
20:40:44preglowhope it's something decent
20:42:00amiconnXavierGr: That won't be easy at all
20:42:37amiconnThe jpeg viewer can only display zoom levels of 1/8, 1/4, 1/2 and full original image, utilising a special feature of jpeg images
20:43:14Paul_The_NerdYou could always check if 1/8 of the width or height is less than 1/2 of the screen, and it both are, start at 1/4, or something
20:43:14linuxstblostlogic: I think preglow told me the hold switch stops the interrupts.
20:43:15XavierGrso arbitary resising is out of the question.
20:43:27amiconnIt chooses the largest level that fits the screen initially, but if the image is larger than a certain size, even 1/8 won't fit
20:43:54XavierGryes I know that. And I've seen about the downscale factor.
20:44:15XavierGrI just hoped that a modification can alter the zoom by a smallest factor
20:44:29amiconnThis certain size is for H300 (simple calculation) 1760x1408
20:44:57XavierGrit is lcd_height*8 * lcd_width*8 IIRC
20:45:24amiconn896x512 on archos...
20:47:04XavierGrso amiconn, you say that rockbox zoom is based upon a jpeg format trick on specific downscale factors?
20:47:09preglowlinuxstb: that i did
20:47:25*Paul_The_Nerd has uploaded a new doom patch, that at least compiles.
20:47:27amiconnThe technique used for scaling can be adapted to allow more zoom levels *between* 1/8 and 1/1 - if you know a lot about IDCT
20:47:32linuxstbPaul_The_Nerd: Do you know which targets Doom compiles for? Is it just the colour targets?
20:47:39preglowamiconn: i say we just implement some interpolator
20:47:53Paul_The_Nerdlinuxstb: All colour targets, I believe.
20:47:57preglowamiconn: newer targets should be able to cope, and it's tons and tons easier than making custom idcts
20:47:57lostlogiclinuxstb: that's why 5 GPIO bits change (guess, but an educated one) 1 is the hold switch indicator and 4 are interrupt mask bits.
20:48:17amiconnpreglow: Interpolating down still needs to decode the higher resolution, meaning we won't save any ram
20:48:22Paul_The_Nerdlinuxstb: My recent patch doesn't build it for any targets but the colour ipods, and both iRiver Hs (because I didn't know what to actually put in that line other than IRIVER
20:48:27preglowamiconn: sure no, but you can choose the zoom factor yourself
20:48:46linuxstbPaul_The_Nerd: I was just about to do what you've just done... So I'll go grab your patch.
20:49:12preglowamiconn: and i mean having the draw routine do the rescaling, just like the it also currently converts from yuv
20:49:25amiconnpreglow: XavierGr's question was whether we can scale down lower than 1/8, for viewing large jpegs while listening
20:49:35preglowthat'll be harder
20:49:36Paul_The_Nerdlinuxstb: It just fixes it to compile and run. It also resizes the plugin buffers, so I don't have to explain it every 5 minutes in the forums. :)
20:49:58preglowit is possible, however, but requires hacking the decoder routine severely
20:51:00XavierGramiconn: I don't say that only for very large pics
20:51:06XavierGrImagine a small picture
20:51:28*amiconn wants two custom IDCTs for archos
20:51:41amiconnXavierGr: Smaller pics aren't scaled down that much
20:51:57amiconnYou'll never end up with an image smaller than 1/2 of the LCD
20:51:59preglowamiconn: why two? all of them would need to be if you mean correcting for aspect ratio
20:52:22amiconnpreglow: Yes, but I guess it's not possible to decode fractional pixels...
20:52:23preglowof course, correcting 1x1 and 2x2 will be... special we could only correct the 2 largest levels
20:52:56amiconn...needing a 10x8 and a 5x4 IDCT
20:52:59preglowi think i know how to do it
20:53:22preglowhmm, 10x8
20:53:24preglowcan you even do that?
20:53:27preglowit would to upsample
20:53:41preglowhave to
20:54:21amiconnWell, it is possible. [IDC]Dragon mentioned that he got one such IDCT routine from an author who explains on the net how to make fast custom IDCTs
20:54:59amiconnThese beasts are supposed to handle about every scaling factor. If only I would understand enough of this...
20:55:20 Quit BHSPitMonkey (Read error: 110 (Connection timed out))
20:55:38 Join BHSPitMonkey [0] (
20:56:05preglowi guess you have to resample for 5x4 too
20:56:14preglowbut of course, in an idct is a very good place to resample
20:56:28preglowsince you can do it ideally in the frequency domain
20:56:38 Quit BHSPitLappy (Connection timed out)
20:58:19 Join damaki [0] (
21:00:01XavierGrI thought that this stuff was a lot more easy. That is completely out of my knowledge
21:00:37linuxstbAnyone understand why gcc would say "warning: operation on eventtail may be undefined" with the folllowing code?
21:00:38linuxstb for ( ; eventtail != eventhead
21:00:38linuxstb ; eventtail = (++eventtail)&(MAXEVENTS-1) )
21:00:47linuxstb(the warning refers to the second line)
21:01:30linuxstbIt's from Doom...
21:02:33*Paul_The_Nerd was just about to go look at the warnings.
21:02:56amiconnlinuxstb: I would think it has problems with the prefixed ++
21:03:15amiconnI'd simply replace that by eventtail = (eventtail + 1)&(MAXEVENTS-1)
21:03:20LinusNyes, there are two assignments to eventtail on the same sequence point
21:04:00linuxstbOK, thanks. I'll do that.
21:04:45 Join imphasing [0] (
21:04:51linuxstbIt's in the network code, so it's probably unused anyway...
21:05:11XavierGrlol imagine 2 h300s linked up to play doom
21:05:25Paul_The_NerdXavierGr: Too late, safetydan and I already imagined 4. :-P
21:06:04*XavierGr drools
21:06:18Paul_The_NerdIt's not really possible without someone rather insane though
21:06:24Paul_The_NerdOr a USB networking driver.
21:06:37XavierGrlinked up with USB OTG bluetooth adapters.
21:06:45*linuxstb drools over the gaming potential of DAPs with wifi or bluetooth
21:06:46XavierGror wi-fi
21:06:49 Join webguest04 [0] (
21:07:08 Part webguest04
21:07:09XavierGryeah maybe in like what.... 4-5 years?
21:07:26linuxstbPortalPlayer have announced such a chip already.
21:07:54linuxstbSo it wouldn't surprise me to see iPods with wifi or bluetooth this year.
21:08:16XavierGrI was speaking for an H300 utilizing its USBI OTG port
21:08:32Paul_The_NerdMan, bluetooth ipods. Automatically sync whenever in range...
21:08:32 Quit ender` (" A psychiatrist is a fellow who asks you a lot of expensive questions your wife asks for nothing.")
21:09:20linuxstbAutomatically sync with any other bluetooth-enabled Rockbox devices in range...
21:09:52lostlogicugh, why doesn't the hold switch on ipod work?
21:10:00linuxstbBut now to squeeze Doom in the ipod's plugin buffer....
21:10:08lostlogicor at least why doesn't it display hold on status/wps
21:10:32linuxstbIt does for me. The wps tag isn't implemented though afaik.
21:10:44Paul_The_NerdYeah, I remember you had a patch that made Doom fit, but it caused some funny graphical artifacts (I think it broke the z-buffer for sprites)
21:11:08linuxstbDo you still have it?
21:11:25lostlogicoh, the wps is just coded wrong *fixes*
21:11:31linuxstbIt's OK - I've found three patches on my website.
21:11:38Paul_The_NerdIt was doom_memory, if I recall
21:11:43Paul_The_NerdOr rather, if this is it
21:12:07linuxstbYes - that's the latest one, so I'm going to try to get that working.
21:12:27Paul_The_NerdI try to keep any patches you send my way until they show up in CVS, or become inconsequential.
21:14:41linuxstbAh, I see the problem... timer_register() isn't defined in the simulator.
21:18:56Kohlrabiah freind of mine seems to have problems with H3x0 rockbox:
21:19:03KohlrabiThe WPS-fonts don't align correctly
21:19:15KohlrabiWhat could be the cause?
21:19:31Paul_The_NerdKohlrabi: That WPS looks like one of the ones that uses various patches that aren't in CVS
21:19:58KohlrabiAh, so another WPS could work?
21:20:09Kohlrabi(he told me all of them had errors)
21:20:20Paul_The_NerdKholrabi: Any of the album art ones will *definitely* have errors
21:20:30Kohlrabiah OK
21:21:05Paul_The_NerdIn fact, the only ones guaranteed to work are those he creates himself, and those that are included with Rockbox. Other than that, many people seem to enjoy creating WPSes that only function with special patches, many of which may not even get included because they aren't how the project wants to handle certain aspects of WPS rendering.
21:22:02KohlrabiAs far as I know he only D/Led rockbox, but I'll check that... :)
21:23:12Paul_The_NerdKohlrabi: Well, that WPS isn't one of the included ones. :)
21:23:25KohlrabiAh, OK
21:23:48Paul_The_NerdAlbum art isn't a feature yet, as there are a few changes left needing to be made to the patch before it can be included.
21:23:56Paul_The_NerdSo any WPS with an album-art section is definitely not going to work
21:24:11amiconnlinuxstb: Do we need the timer in the sim?
21:24:30amiconnIf so, we'll probably need yet another thread providing the user timer
21:24:47amiconn(host thread of course)
21:29:43linuxstbIt's used in Doom to give a 35HZ tick. I think we can just approximate it with current_tick/3
21:29:54 Join ashridah [0] (
21:31:58KohlrabiProblems solved :)
21:31:59 Join Matze [0] (
21:37:26amiconnthis will jitter a bit, but give 35Hz on average
21:39:00linuxstbGood idea.
21:39:42amiconnBtw, how does doom work on iPod then?
21:39:56linuxstbI use the hardware usec tick
21:39:57*amiconn didn't notice user timer code going in for iPod
21:40:16linuxstbI was just wondering when that will wrap... It's a 32-bit unsigned value.
21:40:37linuxstbSo it's only about 4000 seconds.
21:40:43Papricasomeone is against commit tetrox?
21:41:00MikachuPaprica: could you run it through indent first too?
21:41:38amiconnlinuxstb: About 71 minutes...
21:42:08MikachuPaprica: i added a basic pause when hold is enabled..
21:42:14linuxstbNot sure if the battery will last that long....
21:42:20Mikachubut of course i can submit changes after commit too :)
21:42:26PapricaMikachu, i do it
21:42:48*linuxstb whispers "user timer" in preglow's ear
21:43:01preglowahh, right, there's that
21:43:06peturLinusN ?
21:43:11preglowi'll have a look at it now
21:43:28preglowjust finish this new trig code
21:43:52linuxstbI also do "(USEC_TIMER*35)/1000000" which will overflow nicely ...
21:44:00 Join Aditya_ [0] (
21:44:34 Quit Aditya (Read error: 104 (Connection reset by peer))
21:44:37preglowwhat's preferable? long sincos(long angle, long* cos) or void sincos(long angle, long* sin, long* cos) ?
21:45:04amiconnlinuxstb: Just reduce the fraction...
21:45:12*amiconn is quite used to reducing fractions
21:46:10 Quit tvelocity ("Ex-Chat")
21:46:12amiconnAn example is video.rock: The cpu clock frequencies of recorder and Ondio relate 11059200 : 12000000 == 576 : 625
21:46:21Mikachui cheated and used maple there
21:47:01amiconnAnother example is that the iriver (base) clock and archos recorder clock relate 11289600 : 11059200 == 49 : 48
21:47:29 Join sharpe [0] (
21:47:44Mikachuamiconn: can you factor arbitrarily large numbers instantly?
21:47:56amiconnPrepared for video.rock on H1x0...
21:47:58Papricammm how could i remove a file from cvs?
21:48:13preglowMikachu: he's got a cybernetic implant that does it for him
21:48:15Mikachucvs delete file
21:48:16Mikachui think
21:48:22Mikachucould be remove
21:48:37linuxstb"rm file" followed by "cvs remove file" followed by "cvs commit -m 'message' file"
21:49:29*amiconn wonders what happened to Wett
21:49:47Papricaok im asking again, someone is against commit tetrox?
21:50:22Papricaand remove rockblox
21:51:04linuxstbI don't think you should remove rockblox yet.
21:51:15linuxstbIt's useful to compare tetrox with.
21:51:55 Quit qwm (Read error: 110 (Connection timed out))
21:52:10linuxstbIMHO, commit tetrox now, and when everyone agrees it's better than rockblox, then delete rockblox.
21:52:20Paul_The_NerdDoes tetrox run on the Archos' now?
21:52:48*linuxstb would still prefer a rotated screen
21:53:19linuxstbPaul_The_Nerd: Should doom look for wads in /games/doom/ ?
21:53:35Paul_The_Nerdlinuxstb: That's the current behaviour, at least.
21:53:54linuxstbMmm. I've got the sim compiling, but it's not finding the wads
21:54:38Paul_The_Nerdlinuxstb: The filenames are correct? It doesn't scan for arbitrary wads, but rather looks for specific ones.
21:54:52linuxstbIt's the same as I've used on my ipod - doom1.wad
21:56:19preglowamiconn: timer code currently assumes the timer.c base freq is CPU_FREQ, yes?
21:56:32preglowi'll just introduce TIMER_FREQ, then
21:56:43linuxstbbbl. dinner time.
21:56:48 Nick Rondom is now known as Polizei (
21:57:56Papricalinuxstb, did you see the screenshot of tetrox on ipod video?
21:58:02 Join fiftyfour123 [0] (
21:58:51Papricatetrox on archos
21:59:07Paul_The_NerdThat is _small_ :)
21:59:17Papricaits like the rockblox
21:59:32sharpeonly different...
21:59:35 Join Rondom_ [0] (n=Rondom@
21:59:40Paprica3x3 blocks
21:59:54sharpeinstead of what... 2x2?
21:59:55 Nick Rondom_ is now known as Rondom (n=Rondom@
22:00:31 Part gracchus
22:00:51Papricamm let me check, dont remember
22:01:00 Nick Rondom is now known as Polizeiverst (n=Rondom@
22:02:47Papricasharpe, tetrox is 3x3 on archos and rockblox is 3x4
22:03:36sharpethanks... now i can sleep today :)
22:03:42 Nick Polizeiverst is now known as Polizeiprivat (n=Rondom@
22:03:47 Nick Polizeiprivat is now known as Rondom (n=Rondom@
22:04:04sharpeseems to be cycling names...
22:04:44 Join solexx_ [0] (
22:06:06Papricalinuxstb, so what do you think? commit it without remove rockblox?
22:06:08amiconnPaprica: Guess why rockblox is horizontal?
22:06:15*petur wonders what [BROS] has to do with usb host
22:06:31Mikachui find it really hard to use the nano rotated
22:06:36Papricayep =]
22:06:41sharpeum... because they both have an 'os'... is why...
22:07:17sharpeand i know i'm not helping today... :(
22:07:25Mikachusuccess, now my hold-button pause also enables the backlight timeout
22:07:41amiconnpreglow: Sure you want long*, not int32_t* ?
22:09:33Papricaamiconn, rockblox have 10x17 tetris screen while the orginal tetris have 10x20
22:12:07preglowamiconn: talking about sincos?
22:12:39preglowamiconn: in that case it doesn't matter much, it assumes it's got 32 bits, but doesn't care if it's got more
22:13:00amiconnIn other cases it does matter
22:13:05preglowi'm testing the algo on a 64 bit box and works fine
22:13:09amiconn(like the playback-on-speed thing)
22:13:26amiconnI suspect the dsp code
22:13:38preglowsomething new is wrong?
22:13:59 Quit Paul_The_Nerd ("Leaving.")
22:14:01amiconnNothing new
22:14:18preglowthen i can't remember what you're refering to
22:14:27 Join actionshrimp [0] (
22:14:33 Quit Polizei (No route to host)
22:14:52amiconnIirc the dsp input is 32bit ( 6.26 ??)
22:15:18amiconn...but it seems to read the data with a long*
22:15:29amiconn..and that's clearly wrong on a 64 bit machine
22:15:29preglowthat it does indeed
22:15:49amiconnI did not yet check thoroughly
22:15:53preglowthe dsp code has always assumed it's either got 16 bit data or 32 bit data
22:15:58amiconnThis is code I didn't touch often...
22:16:34amiconnpreglow: It can continue to assume that, but it must use explicit int types to do so
22:16:45preglowi haven't looked too closely at it myself
22:17:00preglowbut i've always tended to forget the sim when coding, so i've probably always used long* everywhere myself
22:17:30***Saving seen data "./dancer.seen"
22:19:58 Quit fiftyfour123 ("Chatzilla 0.9.71 [Firefox]")
22:20:20 Join fiftyfour123 [0] (
22:20:23 Quit solexx (Read error: 110 (Connection timed out))
22:23:36 Quit ashridah ("Leaving")
22:24:40 Part LinusN
22:25:49 Join stripwax [0] (
22:27:55 Quit fiftyfour123 ("Chatzilla 0.9.71 [Firefox]")
22:28:16MikachuPaprica: 85880 points :)
22:28:28MikachuPaprica: 168 lines
22:28:32Papricalol =]
22:28:36MikachuPaprica: but my thumb hurts
22:29:29Papricado you find better button mapping?
22:29:46linuxstbstripwax: I was thinking about pacbox on the h100 - what do you think about simply replacing rgb_to_gray() with a 16-case switch statement (based on the 24-bit rgb value) which just assigns pre-determined greyscales for each of the 16 pacman colours.
22:31:45amiconnlookup table...
22:32:07linuxstbFrom a 24-bit value?
22:32:10*amiconn wonders whether the grayscale lib would be fast enough
22:32:26amiconnlinuxstb: No, for the 16 pacman colours
22:33:50*amiconn also wonders why colur is computed in full 24 bit instead of just using a palette
22:34:29linuxstbIt's not quite as easy as that... The palette[] lookup table for a reason I haven't investigated yet contains the 16 entries spread out over 256. But it's not performance-critical, the switch statement I suggested is just called during initialisation.
22:35:56stripwaxlinuxstb - maybe... I was thinking of a "hi-contrast" mode which replaces everything that isn't BLACK with WHITE for ease of visibility (wdyt?). So, in your scheme, that would be two lookup tables. If there's only 16 unique colours then the 256-entry table could contain indicies into the lookup table :-)
22:36:21stripwaxHowever that would make it hard to implement a dithered renderer for H1x0. Not that that is ever necessarily going to happen
22:36:44linuxstbamiconn: The original emulator initialises a 256-element palette array with 24-bit rgb888 colour values. I've changed the format of the palette array so it contains "rockbox native" values - 2 bit greyscale or the appropriately packed RGB565
22:37:16 Join Daishi [0] (
22:37:19linuxstbThe issue is how the palette array values are initialised for greyscale.
22:37:37stripwaxAs for optimisation - I was thinking we could 1) only render the pixels that are going to make it as far as the LCD (i.e. every fourth pixel of the sprite and every fourth pixel of the background chars) and 2) optimise which rects we blit, rather than doing a full lcd update each time. Both should improve framerate considerably
22:37:40linuxstb(to map the 16 used colours into something that looks clear on the h1x0)
22:37:52amiconnlinuxstb: Why not just compute the palette at build time?
22:38:04stripwaxamiconn - hm, how?
22:38:11linuxstbBecause it's contained in the ROMS
22:38:19linuxstb(I think...)
22:39:06 Part tucoz ("Leaving")
22:39:34*amiconn thinks it's odd to run pacman on an emulator instead of natively
22:40:01linuxstbIt's because pacman clones are never the same as the original.
22:40:34linuxstbThere are patterns around the maze you can perform on the arcade version, but I wouldn't expect them to be the same on a clone.
22:40:42peturthey're probably faster ;)
22:41:01stripwaxamiconn - actually I was beginning to write a clone (it wouldn't be too hard). but I gave up as soon as I saw pacbox :-)
22:41:06linuxstbYes, but at least on the 5g, it's now realtime.
22:41:15linuxstb(almost - about 95% realtime)
22:41:28amiconnA native version would be much less demanding
22:41:28stripwaxDoes the 5g have cpu boost? and if so is pacbox running at 100% cpu...?
22:41:31Mikachushouldn't the 5g be faster than the nano with it's four cpus? :)
22:41:46amiconnIt would very probably run realtime even on archos
22:42:05linuxstbamiconn: Very true. But I want the original.
22:42:13amiconnlinuxstb: Does playing music in parallel work without skipping?
22:42:19linuxstbNo chance.
22:42:21*petur excuses for starting another speed discussion
22:42:30stripwaxanyone can write a pacman clone, but few are as good as the original
22:42:55Mikachui found a page with mods to the original roms when i was er, googling for something related
22:43:43 Quit phaedrus961 (Read error: 110 (Connection timed out))
22:45:15stripwaxAnyone like the idea of a text adventure engine for rockbox (TADS and/or INFOCOM)?
22:45:27Mikachuwouldn't it be a tad hard to enter text?
22:45:27linuxstbDoesn't that involve a lot of typing?
22:45:29sharpesounds interesting :)
22:45:40Mikachuthere is one for gba, and i don't like it :)
22:45:41stripwaxselecting from a dictionary of words rather than actually typing letter by letter ...
22:45:51sharpehey, what about a nethack clone...
22:47:01linuxstbI've discovered the Doom sim doesn't work because it uses the system access() function which obviously isn't aware of the fake archos/ directory.
22:47:12cs_weaselstripwax: I am actually looking at the source code to frotz at this very moment :)
22:47:32stripwaxcs_weasel - spooky
22:47:52sharpei compiled my first simulator :)
22:47:57cs_weaseli think a simple port would be easy, just using the funky kbd input thing
22:48:11stripwaxcs_weasel- yeah.. but ... :-)
22:48:11cs_weaselmight look into something more sophisticated if I can get that going
22:48:26stripwaxcs_weasel that would be v. cool.
22:49:16amiconnlinuxstb: Why does doom need access()?
22:49:34safetydanWell... that wasn't really worth two hours...
22:49:42amiconnNote that I didn't know about this fn until now, so I judge from quick googling.
22:49:45 Join webguest55 [0] (
22:49:52*safetydan just finished watching the Roman Polanski version of Oliver Twist
22:50:30amiconnlinuxstb: Imho it should be possible to macrofy away access()
22:50:32webguest55didn't someone already create a z-interpreter?
22:51:00linuxstbamiconn: I didn't recognise it either. It's just used to check if a file exists - so I've changed the name to fileexists() and kept the internal implementation in rockdoom.
22:51:31 Join ender` [0] (i=ychat@
22:51:45linuxstbBut now I get a segfault in sim_readdir() Investigating....
22:51:55cs_weaselwebguest55: did they? i really hadn't thought to look around too closely
22:52:34webguest55cs_weasel: I can't seem to find it in the tracker yet, but I'm quite sure there was one
22:52:36 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
22:52:40stripwaxwebguest55 - I mentioned it about a month or so ago.. could you be thinking of that?
22:52:53 Quit amiconn (Nick collision from services.)
22:52:53 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
22:52:59webguest55stripwax: no, this was many months ago
22:53:20cs_weaselwell don't count what I'm saying as much more than a hollow half-promise either, but it'll give me something to tinker with in my spare time.
22:53:21stripwaxah, nemmind then. if someone has already done that, then it's yet another thing I don't need to bother with :-)
22:53:33*stripwax wishes he actually had spare time
22:54:04cs_weaselyeah, i've got an hour here and an hour there :)
22:54:05sharpei have spare time!
22:54:42stripwaxwhen I'm not at work, I'm either eating or sleeping at the moment
22:54:52webguest55"Rick" was porting pinfocom
22:54:56webguest55not sure what came of it
22:55:20webguest55A year ago, roughly
22:55:22stripwaxoh! well, good on him!
22:55:53cs_weaselfrotz has a nice dumb build that doesn't need curses, doesn't rely on much other than itself, and already has code to adjust for tiny terminals so
22:55:57stripwaxor did he just mock up four screenshots in mspaint? ;-)
22:55:59cs_weaseli'm thinking of working from there
22:56:14Mikachuwhat's minotaur?
22:56:30sharpemythical creature
22:56:38Mikachui meant the game in that dir
22:56:49sharpe;) couldn't tell you...
22:56:51stripwaxcompile it and find out! or, check out the first two screenshots
22:56:53webguest55stripwax: would be a silly thing to hoax
22:57:15cs_weaselheh hunt the wumpus would be fun
22:57:25Mikachui'm a bit slow in the head
22:57:26stripwaxwebguest55 - well, yeah. but I don't see anything of substance there as evidence..
22:57:38stripwaxor a pixel maze .. ;-)
22:58:37stripwaxcs_weasel - yeah, I think I must have checked out frotz as a reference too, I remember looking at about six z interpreters and one that had a 'dumb terminal' point stuck out as being a good starting point
22:58:39webguest55I guess he disappeared
22:59:12Mikachuwould you be surprised if i said it doesn't compile ootb? :)
22:59:23sharpepixel maze...
22:59:29sharpethat's what i'm supposed to be doing
22:59:42stripwaxyeah ;-)
22:59:50webguest55Mikachu: probably needs adopting to a) new graphicsapi, b) new pluginapi
22:59:53sharpethank you, i had almost forgotten
23:00:05stripwaxnamed 'Mack' iirc..
23:01:17sharpeokay, i'm going to get started on that pixel maze, with the scaling ability which will make it, not a definite pixel maze. :)
23:01:28webguest55Wonder if it's the same Rick that's here now
23:03:00lostlogiclinuxstb: the bit we are using to detect USB and send to disk mode is indeed the charging bit.
23:03:10lostlogicso now how do we _really_ detect USB mode? :(:)
23:05:22linuxstbThe code that's there for the other 4g ipods works fine on my 5g - but one person reported that it always gave a false positive.
23:05:49linuxstbSo I changed it. Maybe we should just put it back and see if anyone shouts again.
23:06:19lostlogiclinuxstb: maybe
23:06:53linuxstbIf you check the cvs logs, you should find the commit message.
23:07:34linuxstbThis was the diff which changed it:
23:07:37lostlogiclinuxstb: but yeah, I'm pretty confident that bit 0 of gpiob definitely looks to be the charging status bit
23:08:11linuxstbYes - now that I've started using my 5g more, I noticed that the times it failed to detect USB were the times when it was fully charged...
23:08:59lostlogicyep, and I just deliberately fully charged my battery, checking the gpio occasionally, upon full charge, the bit went high, and I verified full charge by forcing disk mode and waiting a few secs.
23:10:12lostlogicI can't really futz with it until my plane ride tomorrow, cuz I have critical bugs at my real job ATM.
23:13:40petur(coldfire asm Q) does the bitnumber passed with BTST start at 0 or 1?
23:14:24ohrnanyone around that could plz take a look at my profiling bugfix and commit it?
23:14:34ohrnI can give a deeper explanation if necessary
23:17:01stripwaxlinuxstb - hey, looks like there's a pacbox bug when pacman goes off the left-hand tunnel, his sprite gets all mashed up.
23:17:01 Join jocke1s [0] (
23:19:41 Part jocke1s
23:19:48 Join scf [0] (
23:20:01sharpestripwax, i've decided on a name
23:20:40sharpemore like a temporary name 'mixel' :)
23:21:35stripwaxmixel's cool
23:22:43lostlogicohrn: looks good, will commit.
23:22:53 Join matsl [0] (
23:23:22 Join kkurbjun [0] (
23:23:25 Quit Matze ("Miranda IM! Smaller, Faster, Easier.")
23:24:09webguest55It's the Doom guy!
23:24:22 Join oumouk [0] (
23:24:38 Quit actionshrimp ("( :: NoNameScript 4.01 :: )")
23:24:45kkurbjunyep, I have a new release now
23:24:47lostlogicohrn: ermh, maybe not −− why don't codecs have this problem?
23:24:47*amiconn is tempted to commit his H300 lcd changes
23:24:48 Join azha [0] (
23:25:10azha #bashfr chan ou on parle pas d'info, chan entre ami, venez on s'amuse bien
23:25:10azha #bashfr chan ou on parle pas d'info, chan entre ami, venez on s'amuse bien
23:25:10DBUGEnqueued KICK azha
23:25:10azha #bashfr chan ou on parle pas d'info, chan entre ami, venez on s'amuse bien
23:25:10***Alert Mode level 1
23:25:10azha #bashfr chan ou on parle pas d'info, chan entre ami, venez on s'amuse bien
23:25:11***Alert Mode level 2
23:25:11***Alert Mode level 3
23:25:11azha #bashfr chan ou on parle pas d'info, chan entre ami, venez on s'amuse bien
23:25:11***Alert Mode level 4
23:25:11***Alert Mode level 5
23:25:11azha #bashfr chan ou on parle pas d'info, chan entre ami, venez on s'amuse bien
23:25:12***Alert Mode level 6
23:25:12***Alert Mode level 7
23:25:12azha #bashfr chan ou on parle pas d'info, chan entre ami, venez on s'amuse bien
23:25:14***Alert Mode level 8
23:25:14azha #bashfr chan ou on parle pas d'info, chan entre ami, venez on s'amuse bien
23:25:16 Quit azha (Excess Flood)
23:25:25 Part YouCeyE ("Leaving")
23:25:26webguest55Yes, because now we'll want to go there...
23:25:45 Join webguest37 [0] (
23:25:51 Join azha [0] (
23:25:53sharpejust a guess... but, i'm guessing french bash...
23:26:23azha#bashfr,venez parler entre amis de tous et de rien
23:26:25azha#bashfr,venez parler entre amis de tous et de rien
23:26:27***Alert Mode level 9
23:26:27DBUGEnqueued KICK azha
23:26:27azha#bashfr,venez parler entre amis de tous et de rien
23:26:29linuxstbkkurbjun: Just as I've spent an hour debugging and cleaning up your last version.....
23:26:29***Alert Mode level 10
23:26:29azha#bashfr,venez parler entre amis de tous et de rien
23:26:31***Alert Mode level 11
23:26:31azha#bashfr,venez parler entre amis de tous et de rien
23:26:33 Quit azha (Excess Flood)
23:26:33webguest55stop that, kindly
23:26:40preglowlike i've said before: we need more ops
23:26:55kkurbjunlinuxstb: oy, which version were you working with?
23:26:55CtcpIgnored 2 channel CTCP requests in 1 minute and 17 seconds at the last flood
23:26:55*Paprica play's pacbox in 19.5 fps on his h300
23:27:05linuxstbThe one linked to from the patch tracker.
23:27:08stripwaxPaprica - cool
23:27:20 Join herz42 [0] (
23:27:26kkurbjunahh, not the test version
23:27:36kkurbjunthis new one has all that stuff from the test version
23:27:39linuxstbstripwax: I'm about to commit it - it's a simple patch that just puts the z80 registers and other small things in IRAM.
23:27:40Paprica - sweeeeet
23:27:54kkurbjunI fixed it so it doesn't have the stack problems
23:28:09 Quit webguest37 (Client Quit)
23:28:30kkurbjunand it "should" fit on the ipod although there is quite a bit more code
23:28:35stripwaxlinuxstb - nice. i'll fix the sprite mangling and optimise drawChar a bit too, and upload a patch... any chance I could get cvs access (please? ;-))
23:28:58stripwaxif not no worries
23:28:58linuxstbI'm not the person with that power.
23:29:04stripwaxI know ..
23:29:30sharpewoohoo. i'm on my way.
23:29:36safetydanpreglow, how goes the trig?
23:29:41kkurbjunI have all the tables loading from disk with a malloc along with the visplanes being allocated dynamically it cuts down on the plugin size
23:29:47preglowsafetydan: btw, it seems to give better results if you round the atan lookup values beforing saving to int
23:29:52 Quit oumouk (Excess Flood)
23:30:04preglowsafetydan: which also has the curious property of making all the entries 2^x
23:30:39stripwaxHere's why the palette table is 256 entries: each 8x8 char can contain at most 4 colours, and *which* four colours are chosen are determined by lookup into one of 64 4-colour entries in the palette table. even though there are only 16 distinct colors. boy, Pacman video hardware must be wildly strange.
23:31:11 Join phaedrus961 [0] (
23:31:18stripwax(lines 414,415 of arcade.c)
23:32:40preglowsafetydan: and yeah, the maximum error in the routine currently seems to be 5
23:33:07preglowsafetydan: which is probably explained by our not using the two last available bits
23:33:45kkurbjunlinuxstb: what changes/bugs did you fix in the code that you were working with?
23:34:02kkurbjunI might be able to incorporate some of them
23:34:16sharpehmm... how should i have a copy of the maze stored in memory?
23:34:51stripwaxsharpe - its a simple maze so I'd go for an array of bytes, each byte containing 8 bits, each bit being 'wall' or 'floor'
23:35:01linuxstbkkurbjun: I'm mainly trying to get it to work in the sim at the moment.
23:35:03stripwaxit's not like speed is important here ;-)
23:35:18sharpewhat i was thinking
23:35:46kkurbjunlinuxstb: that would be great, I had it compiling, but I didn't have it running as it wasn't finding any of the files I had in the player directory
23:35:57sharpeoh... noes! i need to find a maze generation algorithm ...
23:36:13stripwaxsharpe - there's so many .. google in no time
23:36:32***Alert Mode OFF
23:36:41linuxstbkkurbjun: Yes, I fixed that - it was using the system version of the access() function, not the version in rockdoom.c - so it wasn't looking in the archos/ directory.
23:36:44sharpei know :) trying for that sarcasm again...
23:37:09linuxstbkkurbjun: I've just been distracted with a pacbox optimisation which I'm about to commit to CVS, and then I'll go back to it.
23:37:58kkurbjunlinuxstb: great, I need to try pacbox, it looks very interesting, I just havn't had much time lately
23:38:23stripwaxlinuxstb - is there enough icode space to put drawsprite there too?
23:40:00linuxstbYes, there should be. But it's now running at a full 20fps on the H140.
23:40:09 Join DJ_Dooms_Day [0] (
23:40:22linuxstb(with the patch I just committed)
23:41:09safetydanpreglow, cool, I look forward to seeing a commit in the morning :)
23:41:12safetydannight all
23:41:13 Quit safetydan ("Leaving")
23:41:52kkurbjunlinuxstb: if you get a chance could you see if this new code I submitted compiles/runs on the ipod.
23:41:53 Join tinodeleste [0] (
23:42:59 Join [TCK] [0] (
23:49:29stripwaxis there a quicker way than "make install" to 'refresh' the archos dir for a sim build?
23:49:59 Quit linuxstb (Read error: 104 (Connection reset by peer))
23:50:09 Join linuxstb [0] (
23:50:46linuxstbkkurbjun: Here is a patch with my Simulator-related changes:
23:51:03linuxstbIt's now working fine in the sim :)
23:51:31linuxstbstripwax: If you are just testing a plugin, then just copy the .rock file into archos/.rockbox/rocks/
23:52:32stripwaxcool.. and the .rock is located where? I guess I've been a bit getting by a bit backasswards all this time..
23:54:14linuxstbapps/plugins/ in your build directory.
23:54:18kkurbjunlinuxstb is that code for the timing in the sim accurate?
23:54:47kkurbjunoh, great, I can use that instead of a timer on the H300 too then
23:55:36linuxstbYou can test it and see - it averages to 35 ticks per second, but it isn't 35 ticks every second...
23:55:56*webguest55 notes a multitude of c++ comments in that diff
23:56:09linuxstbBut what's the problem with the timer? You've already done the work and implemented it.
23:56:30kkurbjunlinuxstb, I'll give it a try and see how it works, it would be nice not to have to use a timer thoguh
23:56:44kkurbjunlinuxstb: it does have that overhead though
23:57:02kkurbjunit's not too noticable at that slow of a timer..

Previous day | Next day