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

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

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

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

#rockbox log for 2006-08-30

00:00:49 Join Rob2222 [0] (
00:00:59 Join Febs [0] (
00:02:52 Join jborn_ [0] (
00:08:09 Quit gromit` ("Coyote finally caught me")
00:10:05 Part Paul_The_Nerd
00:10:39 Quit tucoz ("Leaving")
00:11:45 Quit webguest67 ("CGI:IRC 0.5.7 (2005/06/19)")
00:13:13 Join xNibbler [0] (
00:14:38 Quit netmasta10bt (Read error: 54 (Connection reset by peer))
00:15:44 Quit rretzbach (Connection reset by peer)
00:24:38 Quit Nibbier (Read error: 113 (No route to host))
00:24:56 Quit bluebrother ("Leaving")
00:26:28 Join webguest87 [0] (i=42c07542@gateway/web/cgi-irc/
00:27:53 Nick webguest87 is now known as SoapOutaTown (i=42c07542@gateway/web/cgi-irc/
00:29:02 Join midkay [0] (n=midkay@rockbox/developer/midkay)
00:32:26 Quit TeaSea (Remote closed the connection)
00:32:34 Join TeaSea [0] (
00:32:53SoapOutaTownStupid corporate firewall blocks the foruums even when I log in remotely. :(
00:34:51 Quit SoapOutaTown ("CGI:IRC")
00:47:49preglowi take it it's a bit optimistic to think that vanilla c/c++ has some kind of endianness define? :/
00:51:25 Quit lodesi_ ("leaving")
00:53:48 Join barrywardell [0] (
00:56:33 Quit Kohlrabi ("Quit")
00:59:26 Quit ender` (" Programming is like sex becuse: 1. One mistake and you have to support it for the rest of your life.")
01:05:06 Join safetydan [0] (i=cbca159f@rockbox/developer/safetydan)
01:06:10 Quit TeaSea (Remote closed the connection)
01:07:36 Join TeaSea [0] (
01:13:31*preglow piledrives autoconf
01:17:04barrywardellBagder: sound on H10
01:17:14barrywardellMrH was right again!
01:20:54 Join linuxstb_ [0] (i=5343d4aa@gateway/web/cgi-irc/
01:21:09linuxstb_barrywardell: What was the problem?
01:21:59barrywardellthe i2c address is 0x1b for the h10, not 0x1a as in the ipod
01:22:13preglowbarrywardell: !!
01:22:23preglowdamn, this is progressing fast
01:22:35barrywardellthanks to MrH
01:22:51preglowdoes he ever stop by here?
01:23:17ArathisI hope he does't stop having good ideas
01:24:33preglowbarrywardell: so, how far away from a gentleman mail are we?
01:24:59barrywardellit's not perfect yet. the volume is very low at 0db and gets distorted at higher volumes
01:25:16barrywardellpreglow: gentleman mail?
01:25:58preglowbarrywardell: "gentlemen, we have sound"
01:26:12preglowtradition of ours
01:26:25preglowlike that
01:27:03barrywardellwe're not far off
01:27:24preglowdoes the h10 use the ata interface?
01:27:28barrywardelli only got it working about 5 minutes ago
01:27:53barrywardellpreglow: yes
01:27:54preglowi can't remember if it uses flash or microdrives
01:28:09preglowbarrywardell: hell, then you shouldn't be too far from a working rockbox, then
01:28:10barrywardellit uses 1.8" and 1" drives iirc
01:28:37barrywardelland we'll have a working rtc tomorrow thanks to lorijho
01:29:40preglowreally a very speedy port
01:29:49preglowbut i guess the fact that we already have portalplayer targets helps
01:30:20barrywardellyes, the ipod port made life a lot easier
01:30:36 Join uski [0] (
01:30:45barrywardellvery similar hardware.
01:30:51uskihi, anyone familiar with ipod minis ?
01:31:03preglowuski: somewhat
01:31:29uskia friend gave me one without the harddrive, and i would like to try to replace the microdrive with a plain CF card compatible with ATA mode; i have read the instructions on how to install the firmware (...)
01:31:39uski(...) but it says i need first to extract apple's bootloader
01:31:53uskias i don't have the original microdrive, i don't have the original apple bootloader :)
01:32:03preglowwhat generation mini is this?
01:32:05uskiso my question is: how can i install rockbox on my ipod mini ?
01:32:15uskiaccording to the SoC version PP5020
01:32:29preglowis the drive size inscribed on the back of it?
01:32:29uskiand not 5022; i have read that 5022 is for the second generation
01:32:36uskii don't have the case
01:32:56preglowit should be inscribed on the back of the unit itself
01:33:14uskii only have the pcbs
01:33:24uskilemme check if i see something
01:33:27preglowbut ok
01:33:34preglowthen i can't help you anyway
01:33:39preglowi've only got 2g stuff
01:33:48linuxstb_You can install Rockbox without the Apple firmware if you want to.
01:33:52uskibut don't worry, it boots, it will work if i'm able to install a firmware and if it's able to use a real compactflash
01:34:06uskilinuxstb_, hi; ok, i haven't seen that in the PDF, lemme read it again
01:34:14linuxstb_No, it's not documented...
01:34:21 Quit lightyear (Remote closed the connection)
01:35:26linuxstb_You can just use any file instead of "apple_os.bin" in the (almost) final step of the installation instructions.
01:35:39linuxstb_Do you know how to partition an ipod?
01:36:08uskii'm totally new to ipods, but i'm not new in "computer stuff" nor in rockbox (i've used rockbox for many years on several archoses)
01:36:22uskii'm using linux but i can access a windows box if necessary
01:36:42linuxstb_No, Linux is easier
01:36:44uskicreating a partition is not a problem for me, is there something special that should be done ?
01:36:47uskiok great
01:37:02uskiit seems that there is a system partition and a data partition
01:37:10uskior sth like that
01:37:58linuxstb_Yes. The first should be as small as you can make it (if you don't want the Apple firmware), and have partition type 0 (zero). The second is the FAT32 partition, and is of type "b"
01:39:11uskiit seems that doing dd if=rockbox.bin of=the_first_partition will do the job ?
01:39:12linuxstb_(1MB is more than enough for the firmware partition with just Rockbox - give it about 20MB if you want to install the Apple firmware in the future).
01:39:38linuxstb_Yes, but you need to create them first.
01:40:20uskiusing ipow_fw -g mini -o rockboot bootloader-mini1g.bin ?
01:40:26uskiomitting -i apple_firmware.bin ?
01:40:39uskior using a 0-length file as -i apple_firmware.bin ?
01:42:43linuxstb_0-length probably won't work. Try a 512byte file - create it using dd if=/dev/zero of=apple_os.bin count=1
01:44:00 Join OolonColuphid [0] (i=3fa98405@gateway/web/cgi-irc/
01:45:16uskilinuxstb_: ok, i created a file, i'll now partition the CF card using a CF reader and copy the raw image file generated by ipod_fw to the "type 0" partition using dd...
01:46:25 Quit Rondom ("Ex-Chat")
01:46:53 Quit Genre9mp3 ("I don't suffer from Rockbox psychosis. I enjoy every minute of it.")
01:47:04***Saving seen data "./dancer.seen"
01:48:34preglowbedzor time
01:48:42preglowgood night
01:50:44 Join lucas42 [0] (
01:53:47uskilinuxstb_: i successfully (i guess) partitionned and copied the firmware image to the card; when booting the iPod, should I see something special instead of the Apple logo ? I get the apple logo then a folder icon with a ! mark, which is bad i guess :)
01:53:52 Join lorijho [0] (n=lorijho@
01:54:02elinenbebarrywardell: nice work on the h10 port... congrats!
01:54:46barrywardellthanks elinenbe
01:55:16barrywardelli don't deserve all the credit though
01:55:18linuxstb_uski: That means something went wrong... Either the partitions are wrong, or the rockboot.bin is wrong.
01:55:49uski(or the card doesn't support IDE mode... i'm trying to check that with a CF->IDE adapter)
01:56:22uskiit'sz a very small card, 16mb, here is my partition table FYI:
01:56:25uski Device Boot Start End Blocks Id System
01:56:25uski/dev/sda1 1 15 944 0 Empty
01:56:25uski/dev/sda2 16 244 14656 b W95 FAT32
01:57:34 Quit spiorf (Remote closed the connection)
01:57:36linuxstb_That looks OK. You can try settings the Boot flags on both partitions - that's the only difference I can see.
01:58:03uskiok i try that, just in case
02:00:11uskisame problem, i need to find my CF->IDE adapter to check for IDE compatibility
02:01:18Arathislinuxstb_: just a dumb question, but it says the first partition is empty, but if I understand you correct the bootloader should be on this partition, right?
02:02:08linuxstb_Yes. That's how it needs to be.
02:06:02 Quit Febs ()
02:06:16uskiArathis, it says it's empty because the type "0" means empty, but the bootloader searches for a type 0 partition, so it looks for a so-called "empty" partition which is not actually
02:07:04Arathisso the bootloader is there and that's not the problem.
02:08:26elinenbebarrywardell: when's the commit coming?
02:08:56barrywardellIt's a very small change
02:09:01barrywardelllook for it in a minute
02:12:37 Join TCK [0] (
02:12:46uskiArathis, yes, because it's not because the partition is marked as Empty that it is actually empty
02:13:05 Part lucas42
02:13:39 Quit OolonColuphid ("CGI:IRC")
02:14:29uskitime to plug in the CF->IDE adapter: reboot time, brb
02:14:31 Quit uski ("Leaving")
02:15:54elinenbebarrywardell: is everything coming along with it? Does the H10 support wav, mp3, aac, etc...
02:16:16barrywardellit supports mp3 anyway
02:16:45barrywardellit probably supports all the other formats the ipod supports but i haven't got any test files here to try out
02:17:31lorijhobarrywardell: I tested ogg,wav,flac files just now. The all play.
02:17:43barrywardellgreat :)
02:18:10elinenbebarrywardell: sounds great! so what's left to do?
02:18:26lorijhoelinenbe: a lot :)
02:18:26elinenbebarrywardell: what's next for you? Helping Bagder with the Sansa? :-)
02:18:28barrywardellelinenbe: quite a lot
02:18:38barrywardelllcd still has some problems
02:18:44barrywardellscroll pad doesn't work
02:19:00barrywardelland lots more
02:19:06lorijhoelinenbe: we're getting there slowly but surely
02:20:24lorijhobarrywardell: you'll find the rtc patch in your inbox in a few minutes
02:21:25barrywardellgreat :)
02:24:38 Join JdG|uni [0] (i=82c20d6a@gateway/web/cgi-irc/
02:26:37 Join uski [0] (
02:27:03uskilinuxstb_: this is getting strange, my card works in ATA mode with my adapter; i suspect the minis need either a big card, or a microdrive, but not a plain CF card
02:27:26uskii checked for voltage problems (5v crd supplied with a 3.3v supply ?) but it seems that all CF cards car work both in 5v and in 3.3v
02:27:43uskithanks a lot for your help, i don't want to annoy you more with that :)
02:28:41uski(or the image is bad... but i'm not sure of the way to check it)
02:29:05uskimaybe apple's bootloader doesn't like the 73kB image (too small?)
02:31:01uski$ ./ipod_fw -g mini -o rockboot.bin -i bootloader-mini1g.bin
02:31:01uskifread failed: Success
02:31:04uskii love this error :D
02:33:18 Quit linuxstb_ ("CGI:IRC (Ping timeout)")
02:33:38linuxstbuski: That's the wrong command...
02:33:57uskii tried previously with ./ipod_fw -g mini -o rockboot.bin -i apple_os.bin bootloader-mini1g.bin
02:34:04uskibut i get the folder/! error
02:34:19uskii think that the apple_os.bin needs to hold something else that zeroes
02:34:31uski -i: image extracted with -e, load and entry address preserved
02:34:40uski==> it seems that the iPod is trying to execute this
02:34:47uskiso it may check for some header or so
02:34:47linuxstbOther people have done it OK with a 512 byte 0 file.
02:34:52uskihmm ok
02:35:32linuxstbThe ipod will execute the Rockbox bootloader. The Rockbox bootloader will use the apple_os.bin if you hold MENU as you boot.
02:36:24linuxstbLet me create a rockboot.bin a slightly different way.
02:37:01 Join Snyper [0] (
02:37:19barrywardellelinenbe: the change is in cvs now
02:37:22 Quit TCK (Read error: 113 (No route to host))
02:38:17linuxstbJust try writing that to /dev/sda1
02:38:33uskisure, one moment
02:39:57uski$ sudo dd if=rockboot-mini.bin of=/dev/sda1
02:39:57uski142+1 records in
02:39:57uski142+1 records out
02:39:57DBUGEnqueued KICK uski
02:39:57uski73192 bytes (73 kB) copied, 0.000556 seconds, 132 MB/s
02:40:01uskii suppose that's correct
02:40:13uski(followed by a "sync" before card removal)
02:40:17 Join TCK [0] (
02:40:45uskiit must be my card, i'll try to find another one (bigger) tomorrow
02:41:16linuxstbThat would be my guess. I know someone has put a CF card in a mini before with Rockbox, so it should work.
02:41:42uskiok, thanks a lot :)
02:42:07JdG|uniamiconn: if you happen to read logs... dropping settings_menu.c from the recorderv2 build dropped the size down 12Kb (gcc 4.0.3)
02:43:25 Join gasdghasdklghasd [0] (
02:43:46 Quit Sinbios (Nick collision from services.)
02:43:48 Nick gasdghasdklghasd is now known as Sinbios (
02:48:58 Join Nico_P [0] (
02:51:53uskilinuxstb, if you happen to remember the approximate size of the CF card that the person you know used in a mini, feel free to tell me ;) - it would help me determining if i need to use a recent card or not
02:56:09linuxstbuski: I've just found this page....:
02:56:46linuxstbIt appears to need a patch to work in Rockbox, but it works fine in the Apple OS.
02:57:11uskinice !
02:57:32uskii'll patch the source, it's not a problem for me, i just need to find the time to install the crosscompilation toolchain
02:57:57linuxstbThere's a script in the tools directory of the Rockbox source which makes that easy. IIRC.
02:58:07uskiok, ty
02:58:26uskia friend will give me a lot of old CF cards, i'll do some tests to see which one works and which ones do not
02:58:44linuxstbFeel free to add your experiences to that wiki page.
03:00:15JdG|unicongrats barrywardell
03:00:28uskilinuxstb, i'll do that :) thanks a lot again for your patience
03:00:40barrywardellthanks JdG|uni
03:09:09JdG|uniif the settings are not stored in the in the config block anymore, is there still need for the RTC settings?
03:09:54 Join saratoga [0] (i=9803f578@gateway/web/cgi-irc/
03:10:10safetydanJdG|uni, I believe the RTC settings are still needed for things that need to be read before ata is inited
03:10:12safetydancould be wrong though
03:10:30safetydanactually... ignore that
03:11:00saratogaBarry: I was wondering about the H10 port since its progressed so fast
03:11:11saratogais it just that the H10 used a lot of the Ipod code?
03:11:39JdG|unisafetydan: i thought that, the RTC settings are loaded very early, but i doubt they are actually needed untill after ata_init anyway
03:11:48 Quit TeaSea (Remote closed the connection)
03:12:16JdG|uniare they stored in there for speed while its running? or only for speed at astartup?
03:12:42barrywardellsaratoga: that's exactly it. most of the ipod code just worked
03:12:48BigMacJdG|uni:i got all of house 2 and am watching it now
03:13:18saratogawhat did you guys actually have to redo? the LCD code and sound driver?
03:13:42JdG|uniBigMac: burn them onto dvd and send em to me..
03:15:09safetydanJdG|uni, the comment in the settings.c file is pretty self-explanatory
03:16:33safetydanthose we either need early, save frequently, or without spinup
03:17:41barrywardelllcd code was similar to ipod but needed some changes
03:18:03barrywardellsound was very similar to ipod. just required one change
03:18:16barrywardelladc was all new
03:18:30barrywardellrtc is largely new
03:18:33saratogawhere the changes based on spec sheets, disassembly, or something else?
03:18:39saratogai'm curious what the process was
03:18:40BigMacyou send me a blank one or pay for some new ones and its a deal
03:18:45BigMacin hd to
03:19:22barrywardellthe lcd is the same as in the X5 and the interface is the same as the ipod so combining the two drivers worked
03:20:03barrywardellthe adc was reverse engineered by MrH
03:20:30barrywardellas was the address for the sound codec
03:21:04barrywardellthe rtc is being done from a spec sheet by lorijho
03:21:57lorijhothe RTC (Epson) spec sheet is excellent!
03:22:39 Quit TCK (Read error: 104 (Connection reset by peer))
03:23:14Nico_Pwill it be hard to get the scrollpad working ?
03:23:47JdG|uniis the global_settings memory watched so we know when its changed to save it? or is it done periodically automatically?
03:24:37lorijhoNico_P: I guess the scrollpad needs further reverse engineer investigation
03:26:16 Join TCK [0] (
03:28:04 Quit barrywardell ()
03:29:23 Quit TCK (Read error: 104 (Connection reset by peer))
03:30:46 Quit Arathis ("[rl_bot quit]")
03:30:54JdG|unianswering own question... it looks like its explicitly called to save the hdd/rtc blocks
03:33:36 Quit JdG|uni ("CGI:IRC")
03:35:16 Quit lorijho ("leaving")
03:39:25 Part pixelma
03:44:09 Join TCK [0] (
03:45:47 Quit saratoga ("CGI:IRC (EOF)")
03:47:06***Saving seen data "./dancer.seen"
03:49:09 Join lucas43 [0] (
03:49:27 Join Drkepilogue [0] (i=478df32d@gateway/web/cgi-irc/
03:54:41 Nick lucas43 is now known as lucas42 (
03:57:00 Join CrossF [0] (
03:58:35 Quit CrossF (Client Quit)
04:08:03 Join ryanpg [0] (
04:18:06 Join webguest15 [0] (i=42c07542@gateway/web/cgi-irc/
04:18:19 Join JFKennedy [0] (i=deefbf35@gateway/web/cgi-irc/
04:18:41JFKennedyi have a question
04:19:28 Nick webguest15 is now known as SoapOutaTown (i=42c07542@gateway/web/cgi-irc/
04:19:32SoapOutaTownjust ask
04:20:13JFKennedyi just tried to playing the Movie files in my nano?
04:20:36JFKennedybut it couldn't foward or rewind
04:20:46SoapOutaTowna .m2v?
04:21:16JFKennedyhow can i do that?
04:21:36JFKennedyit drives me crazyyyy
04:21:42SoapOutaTownmpeg playback is about as raw and prelim as it gets at this point. You can help code that part. Or wait.
04:22:58safetydanJFKennedy, I don't believe seeking in movie files is implemented
04:23:01JFKennedyso have to wait till it'll develop?
04:23:05safetydan at least doesn't mention anything
04:24:13SoapOutaTownit also doesn't mention how scaling isn't there, nor post-processing...
04:24:50JFKennedyit's really gr8 stuff but...
04:26:37SoapOutaTownno buts.
04:27:28JFKennedyno sounds no skip
04:27:42JFKennedyanyway goooood job!
04:28:55ryanpgwould it be any use to try to run objdump on an iaudio m5 firmware?
04:29:54 Quit Nico_P ()
04:30:37 Quit JFKennedy ("CGI:IRC (EOF)")
04:31:21 Quit Drkepilogue ("CGI:IRC (EOF)")
04:32:06ryanpghrm... well tried it anyway... got objdump: m5_fw.bin: File format not recognized
04:44:14 Join EbErT [0] (
04:45:41 Join EbErT_ [0] (
04:45:41 Quit EbErT (Read error: 104 (Connection reset by peer))
04:45:54 Part EbErT_
04:46:53SoapOutaTownlordy it is quiet, is the webclient worthless on a 24.4 modem, or did all the devs give up on the signal-to-noise ratio here?
04:48:03safetydanSoapOutaTown, it's the middle of the night for most of the devs
04:51:46SoapOutaTownOh, I know it's going on 4-5 there, but I'm still used to more chatter this time of day/night. Oh well, I'll shutup and stop screwing the S/N ratio.
04:53:05 Part BigMac
04:53:15 Join BigMac [0] (
04:55:24 Quit ryanpg ("Leaving")
05:23:50 Join Rim [0] (i=458c45c3@gateway/web/cgi-irc/
05:24:44 Nick JoeBorn is now known as JB_away (
05:32:50 Quit Rim ("CGI:IRC (EOF)")
05:39:41 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
05:40:32 Quit linuxstb (Read error: 110 (Connection timed out))
05:40:59 Quit Criamos ("( :: NoNameScript 4.02 :: )")
05:41:22 Quit jborn_ (Read error: 60 (Operation timed out))
05:45:08 Join dan_a [0] (n=dan-mirc@
05:47:09***Saving seen data "./dancer.seen"
06:01:09BigMachey is yotto here?
06:01:21BigMacdunno if he comes on irc
06:01:29BigMacor what his irc sn is
06:13:28 Quit idnar (Nick collision from services.)
06:13:31 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
06:21:48 Join BHSPitMonkey [0] (
06:24:15 Join webguest02 [0] (i=467eee95@gateway/web/cgi-irc/
06:26:25 Quit webguest02 (Client Quit)
06:26:25BigMacwhy is irc always so dead right now
06:27:31dan_aFor most of the devs and users it's around 5am
06:27:40 Join JoeBorn [0] (
06:29:39BigMacoh i see
06:30:28dan_aIt's only insomniacs and Australians in here!
06:30:45BigMaci guess im an insomniac
06:33:17billytwowillyyah, and Canadians eh.
06:33:22billytwowillyit's only 10:30 pm here in alberta
06:33:41BigMacalright im going to start another wos as well
06:33:46 Join Davide-NYC [0] (
06:34:35 Quit rotator ("zzzzzzzzz")
06:34:48dan_aLooking at the map of Rockbox users, I'm surprised there's not more Americans in here at the moment
06:38:29BigMacim here
06:39:08billytwowillyI'm Canadian eh.
06:39:18billytwowillyno doubt aboot it.
06:39:48CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
06:39:48*billytwowilly has no idea where ^ the above stereotype came from noone talks like that here.
06:40:45dan_aBut I've seen Canadian people talking like that in loads of films - it must be true!
06:40:59Davide-NYCis there a command which will only list the applied patches in a cygwin envirnment?
06:41:02safetydanbillytwowilly, the Canadian guy next to me at work definitely says "aboot", though it's only a slight accent :)
06:41:20billytwowillywher'es he from?
06:41:31billytwowillythe maritimes?
06:41:34Davide-NYCsafetydan: did you ever watch that vid?
06:42:22billytwowillyoh yah, I installed rockbox and it seems mostly decent.
06:42:25arpdavide, where in nyc are ya?
06:43:25billytwowillythe chinatown in edmonton is about one block by 7 blocks.
06:43:31*Davide-NYC can't believe he actually typed that
06:43:48arpbilly, thats bigger than nyc
06:43:49billytwowillythere, now you shouldn't feel so bad.
06:43:51arpwell sorta
06:44:03Davide-NYCyo, is there a command which will only list the applied patches in a cygwin envirnment?
06:44:20billytwowillyreally? the china town in toronto is like 40 blocks by 40 blocks.
06:45:10dan_aDavide-NYC: If they're applied to a CVS checkout, you can do "cvs diff -u" - that will show you the code that has changed
06:45:15billytwowillyI assumed NYC would have an even bigger one.
06:46:47Davide-NYCdan_a: just in stdout right? Not going to mess with my changes or produce a diff file, correct?
06:47:08dan_aDavide-NYC: Correct
06:48:33arpbilly, I think its like 3 blocks by 4 blocks
06:48:35arpits not that big
06:48:45arpbut, chinese are all over nyc anyway
06:49:03arpa lot of what was little italy became chinatown.
06:49:18arpbut again, italians are all over the place
06:50:08Davide-NYCThis will remove the blah.patch "patch −−binary -p0 -R < blah.patch" yes?
06:50:24Davide-NYCor is it > ?
06:50:35arp< means from
06:50:35JdGordonwhy are u doing −−binary?
06:50:37arp> means to
06:50:56arpyah, thats probably unneeded
06:51:00Davide-NYCI dunno, it says to do so in the IdiotsGuideToCOmpilingRockbox
06:51:10JdGordoni dont think it is....
06:51:14JdGordonbut if it works..
06:51:15Davide-NYCarp: Chinatown here is HUuuuuge.
06:51:37arpdavide, not really (I'm in nyc as well)
06:58:51 Quit JoeBorn (Read error: 110 (Connection timed out))
07:02:14Davide-NYCarp: alright! what part of NYC?
07:03:20 Quit safetydan ("CGI:IRC (EOF)")
07:03:20 Quit SoapOutaTown ("CGI:IRC (EOF)")
07:04:02Davide-NYCBED TIME
07:04:09 Quit Davide-NYC ("Chatzilla 0.9.75 [Firefox]")
07:05:08 Join sharpe [0] (
07:05:29 Quit sharpe` (Read error: 54 (Connection reset by peer))
07:10:29JdGordondoes the lcd flip display setting not work in the sim?
07:14:46 Join JoeBorn [0] (
07:15:09amiconnJdGordon: no.
07:15:41JdGordonoh goody :)
07:15:45JdGordonand good morning
07:16:34amiconn07:16 here.
07:17:23JdGordoni buzzed you a few hours ago, i dunno if you read, but if you remove settings_menu.c from the build the recorderv2 .elf drops by 12Kb so maybe it might be good moving that to a rock if you really want to slim down again
07:18:38amiconnWhat are the savings with your table-based settings stuff?
07:19:13JdGordoni gave up on that.. too much work to get it working enough to check the difference
07:20:36amiconnI thought you already had it working...
07:20:56JdGordonit is working, but the whole menu tree needs to be converted, which is broing work
07:21:37amiconnYe, sometimes there is boring work to do...
07:22:07JdGordonyes, but it involved changing a few thousand lines :p those emnu files are huge :p
07:22:07amiconnReworking .lang is no fun at all, but it also has to be done (again)
07:22:15amiconnI already did that once...
07:22:55JdGordoni havnt deleetd the files, so i can resume it when im feeling really bored :p
07:23:04JdGordonim playing with removing the config blocks today
07:23:41amiconnRemoving the config block doesn't save space...
07:23:54JdGordonno, but its something to do :p
07:24:13JdGordonhow much over is it?
07:24:21amiconnIf nobody else does it, I will have to further postpone viewports and stuff, doing size reduction work instead :/
07:24:38amiconnIt _has_ to be done
07:25:28JdGordonright, but by how much are we over? because it will get to a stage where rombox isnt feasable anymore.. unless you start doing a sepearate uber-skinny version especially with it
07:25:58JdGordonmoving the general settings menu to a .rock would save about 12kb easily without much effort which is why i checked it
07:26:34*JdGordon personally could care less if the generel settings menu was in a rock, but some dont like that idea
07:26:35amiconnIt means a spinup everytime you edit the settings...
07:27:24amiconnIt would make more sense to make the recording a plugin, but that's impossible atm
07:27:45JdGordonnot the whole menu... just generel settings... which isnt used very often after its initially setup
07:28:37amiconnOn player and Ondio they're used quite often
07:28:45amiconnThese targets have no quickscreen
07:29:54JdGordonoh woops.. i got confused.. ye, the playback menu gets moved up one level and then the rest get rock-ified
07:30:10JdGordonfile-view, display, system, language, voice
07:30:18JdGordonand maybe bookmarking..?
07:30:31amiconnfile view is very important too...
07:30:48amiconnlanguage isn't a menu, it's a browser
07:31:06JdGordon... you cant have everything...
07:32:13amiconnBut afaics a lot of slack has accumulated again
07:32:29amiconnWhen that is cut away, rockbox will become small enough again
07:33:03amiconnDead code, both completely unused functions and unreachable case:s because of impossible actions,
07:33:14JdGordoni had another idea, split off heaps of the core to rocks, and add a 2nd buffer for these core rocks, and instead of unloading them swap them out to disk if the user wants to swap? (i dont know if this would work at all)
07:33:43amiconnunused .lng strings (being one of the main reasons why langv2 was introduced, but nobody else than me seems to care) :-(
07:34:11JdGordoneventually tho there will be a stage where there is no more slimming to be done nd rombox wont be feasable anymore...
07:34:11amiconnAnd on top of that, your table based settings
07:34:37amiconnYes, but that stage can be pushed further
07:35:25amiconnKeeping code size down is a good idea in general, even though most swcodecs have less of a problem in that area
07:36:00JdGordoni agree... but i feel it wil get to the stage where good stuff gets left out because it slightly adds to code size
07:36:22amiconnI don't think so
07:37:38amiconnIf the memory shortage makes developers code their features efficiently, this can only be a good thing
07:37:57amiconnThere's enough bloatware in the world imho
07:38:21JdGordonefficient code is one thing, but sometimes code cant get any more slim..
07:38:39amiconnAnd if a feature is really so complex & big that it doesn't fit on archos - then it has to be left out
07:39:06JdGordonim talking about small stuff.. which add to the rb experience, but are rejected because of "bloat"
07:40:54amiconnWe're 4KB above limit for fm/v2
07:41:49JdGordontrimming the wps and radio would proably help a fair bit
07:42:06JdGordonand the recorder screen (which imho looks horrible atm :p )
07:42:09amiconnThe radio screen needs rework anyway
07:42:32amiconnThe operation is rather inconsistent with the rest of rockbox
07:44:15JdGordonok, well im gonna play with removing the config blocks (coz they annoy me) for a bit, then ill get back to the menus.. I just that does hack the settigns size down enough..
07:46:21JdGordonscratch that... ill numb my brain for a while then play with the configs blocks :p
07:47:12***Saving seen data "./dancer.seen"
07:48:54amiconnThe 2 largest contributors to the growing binary size in the last 2 months were (1) tagcache and (2) actions
07:49:15amiconn(2) puzzles me a lot - it should have had the opposite effect...
07:54:31amiconnaction.c+keymap_recorder.c only add ~2.3KB of code to the fmrecorder build
07:55:03amiconnI remember the size increase being more than that - with the keymap being smaller than today
08:02:31 Join theli_ua [0] (
08:04:07 Join yesyesyo [0] (n=yesyesyo@
08:05:24yesyesyohello audience. I've noticed that recharging an ipod mini (with rockbox active) takes a good while longer than with the default firmware. Is there a way for me to (temporarily) boot up into Apple firmware (not just in disk-mode) to do recharges- if it's faster. If there's a way for a speedier recharge while using rockbox, i'd try that too
08:14:03 Part yesyesyo
08:28:06BigMachold menu down while booting
08:28:17BigMacalright bed time
08:28:18JdGordontoo slow
08:28:26BigMaci was busy making my wps
08:28:51BigMacplanning for release tomorrow
08:29:00BigMacits simple but w/e
08:29:46 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
08:30:34 Nick BigMac is now known as BigMac|catching (
08:30:58 Nick BigMac|catching is now known as BigMac|comatose (
08:32:42midkayhow interesting...
08:33:42JdGordontis tisnt tit?
08:34:32*Bagder reads the latest we have sound mail
08:35:28*LinusN too
08:36:18Bagdervery cool
08:38:57daurnimatorhi all
08:40:29 Join fejfighter [0] (
08:41:11Bagdermight be time to offer h10 downloads
08:46:11JdGordonamiconn: i found a nice place to drop heaps of space... change everything that needs STR() to just pass the lang id and then the function itself can get the correct string and voice...
08:46:12 Quit LinusN (Read error: 104 (Connection reset by peer))
08:48:30JdGordonlong is 64bits yeah? is it really nescacery for the lang id variable to be long instead of int? we dont have more than 16million lang strings do we?
08:52:33Bagderlong is 32bit
08:52:33Bagderon our targets
08:52:33JdGordonso the same as int? or is int 16bits?
08:52:33DBUGEnqueued KICK Bagder
08:52:33Bagderint is 32 too
08:52:33JdGordonok then
08:52:33amiconnlong is 32bits on 16bit and 32bit architectures, 64bit on 64bit archs
08:52:33Bagdernot always
08:55:05Bagderwin64 is an example
08:55:05amiconnint is 16bit on 16bit archs, 32bit on 32bit archs, and 32bit or 64bit on 64bit archs
08:55:05 Join gromit` [0] (
08:55:05JdGordonhmm.. im sure i was tought that long is usually 2xint
08:55:05Bagderyou were taught wrong then ;-)
08:55:05JdGordonobviously :p
08:55:05amiconnshort and char are always 16bit and 8bit, respectively
08:55:05JdGordontime to change unis
08:55:05amiconnBagder: long isn't 64bit on Win64?
08:55:05***Alert Mode level 1
08:55:05***Alert Mode level 2
08:55:05Bagderonly pointers are 64bit
08:55:05DBUGEnqueued KICK amiconn
08:55:05***Alert Mode level 3
08:55:05***Alert Mode level 4
08:55:05Bagdergoogle for LP64 P64 and win64 and you'll learn ;-)
08:55:05amiconnDo you refer to x86 or Itanium?
08:55:05***Alert Mode level 5
08:55:05Bagderits win64, not the cpu
08:55:11Bagdersince linux on these cpus are LP64
08:56:19BagderI assume they use long long for 64 bits
08:58:32 Join LinusN [0] (i=linus@gateway/web/cgi-irc/
08:58:36 Join Hideo_Kuze [0] (i=irc@gateway/tor/x-6561782ca40cb939)
08:59:25JdGordonok, im trying to save space so i want to put a long and a char* together in a union to keep lang support.. does this make sense? if P2ID( of the char*) == -1 then use it as a char, ortherwise use it as a long with ID2P()?
09:00:29Hideo_Kuzehi, can someone test this for me, while playing a song in the wps, go to the filetree mode and press stop, and the music stops
09:00:51Hideo_Kuzeit isn't supposed to do that right?
09:00:59JdGordonit is supposed to
09:01:13Hideo_Kuzewhile in browsing mode?
09:01:35LinusNi don't like it when it does that, but it is in fact intentional
09:01:43Hideo_Kuzeweird, it didn't do that until recently
09:02:00Hideo_KuzeLinusN: I'm not used to it either
09:03:10LinusNi think it's a fringe feature, since the only thing you can do in the browser is to stop the music
09:03:22CtcpIgnored 3 channel CTCP requests in 2 minutes and 32 seconds at the last flood
09:03:22*Bagder agrees
09:03:31LinusNi can't see why that is important, or even wanted
09:03:51midkayonly thing you can do is stop music? what?
09:03:51midkayyou can't switch to a new song or browse? or access the menu or quickscreen?
09:04:05LinusNmidkay: you know what i mean
09:04:24JdGordonenum is always converted to int yeah? even if the values are small enough for a char?
09:04:30LinusNthe only thing you can do to control the playback is to stop the music
09:04:32midkayi truly don't, sorry - that doesn't make any sense at all. "the only thing you can do in the browser is to stop the music". what about all the stuff i listed?
09:04:34 Join einhirn [0] (
09:04:34 Quit einhirn (Client Quit)
09:04:43midkayah.. aren't the other keys already in use?
09:04:53LinusNexactly, so why should Stop work
09:05:06***Alert Mode OFF
09:05:18midkayhaving the functionality is probably nicer than not having it and leaving the key unmapped?
09:05:24*amiconn likes that stop-in-browser feature
09:05:35LinusNmidkay: i don't think so
09:05:52midkaydo you accidentally press stop a lot?
09:05:54LinusNmidkay: what if Left was unused, should we enable rewind?
09:06:03midkaythen it's as well a problem in the WPS..
09:06:13midkayLinusN: if left and right were BOTH not used, then yes, why not rw/ff?
09:06:21midkaysince they go together it's silly to have only one..
09:06:22Hideo_KuzeI thought the stop key used to have the same functionality as the left key, that is to back out of a directory when in browse mode
09:06:32midkaystop is independent functionality, it doesn't complement anything. stop is stop.
09:06:32LinusNmidkay: because that's the purpose of the wps
09:06:45LinusNmidkay: that depends on the target
09:06:55midkayLinusN: right, but if you're a habitual stop-presser then it's as well a problem there. if you aren't then it's no problem right?
09:07:10midkayLinusN: exactly, it doesn't happen in the browser on iPods afaik..
09:07:26midkaybecause there's no dedicated stop button, you just hold play to stop...
09:07:28LinusNmidkay: so if i take it to the extreme, you believe that all unused buttons should be mapped to their WPS mapping?
09:07:54LinusNthat will surely confuse the user
09:08:40LinusNi believe that consistency is important, and that each screen should do its own thing
09:09:02LinusNif you want to control the playback, use the wps
09:09:14Hideo_Kuzeoh yeah, the user would certainly get more confuse since the wps and browse mode aren't kept consistent
09:09:18 Join midkay_ [0] (
09:09:44midkay_since stop happens to be unmapped it's logical that it stops in the browser.
09:10:22midkay_and you can't just lump things into "if the key is unmapped it should be mapped". there's some thought involved, some things make more sense than others. i think stop makes sense to stop the playback in the browser.
09:10:32midkay_you can resume it immediately after as well, with play, no?
09:11:03midkay_er, this didn't get through before i dropped out: <midkay> LinusN: not at all.. again, for example, Left complements Right as far as rewind and fast-forward.
09:11:09 Quit midkay (Nick collision from services.)
09:11:09 Nick midkay_ is now known as midkay (
09:12:07LinusNi actually don't think stop in the browser makes sense
09:12:11LinusNbut that's me
09:12:26Hideo_KuzeI agree
09:12:29midkaythen it's entirely a matter of opinion. :) i've never used rockbox on the irivers to be fair. it sounds logical.
09:12:32*Bagder walks over to LinusN's corner
09:12:48Hideo_Kuzebut it's not like I'm a developer and have a say in things anyway =)
09:12:55LinusNmidkay: i agree that it probably depends on which target you use
09:12:58JdGordonwhen actions first went in off went up one directory... and everyone complained..
09:13:08JdGordonbut im in the corner with you guys :p
09:13:09BagderHideo_Kuze: well, when you agree with us we count your voice! ;-P
09:13:25LinusNJdGordon: this is a general rockbox rule - you can't make everyone happy
09:13:47JdGordonforget rockbox.. its just a given for anything this size
09:13:48Hideo_Kuzesomeone has to always suffer huh?
09:13:48Bagderthe general rule is that whatever you change, there will be a minor riot of upset users
09:13:57Bagderuntil they learn to appreciate it
09:14:23blxthey can always fork otherwise.
09:14:25LinusNi remember when i removed the arrow cursor
09:14:37Bagdera golden moment
09:14:45LinusNi should do it again
09:14:46JdGordonarrow cursor? must have been before my time
09:14:49Hideo_KuzeI hate the arrow cursor
09:15:01JdGordonoh, you mean the selection icon?
09:15:10LinusNthat may in fact save some bytes in the code
09:15:19JdGordonwho uses that? the highliten line is much better :p
09:15:30LinusNabout 3 people, iirc
09:16:35LinusN3 very loud people
09:16:46midkayso you put it back? i never noticed it disappear :)
09:16:54midkayhow silly.. so much easier to read the inverse bar, saves space...
09:19:14 Quit Hideo_Kuze (Remote closed the connection)
09:19:38 Join Hideo_Kuze [0] (i=irc@gateway/tor/x-d0d5f7ed85afbbdf)
09:24:08JdGordonwhich is the important file to compare bin size? rockbox.elf or .bin ?
09:25:26 Quit xNibbler (Read error: 113 (No route to host))
09:26:47LinusNbut the .bin is a pretty good measure
09:27:06JdGordonbin will have to do.. i cvs up'd without saving the map.. is tiny compared to the other 2?
09:28:17 Join xNibbler [0] (
09:30:54JdGordonamiconn: i just did some very odd guesswork and i dont tihnk fixing the menus will help... with no menu code the bin (recv2) is 11kb and elf is 15Kb smaller than cvs... but by my probably very off estimates of averaging 30x4bytes for each menu (x130 items in the current menu) that is 15.6Kb...
09:31:42LinusNJdGordon: the size of the .map file is not important, you are supposed to read it
09:36:57amiconnComparing rockbox.bin size before/after is the most consistent method across all targets
09:36:57JdGordonconst variables add ~1:1 to the bin size tho yeah?
09:37:42amiconnThe firmware files are named different per target, and for some targets there's self-extraction applied
09:38:03amiconnrockbox.bin is the easy-to-remember way :)
09:39:08JdGordonso whatcha rekon about the possible size diff?
09:39:11amiconnJdGordon: I've just read about your menu rework idea (trying to catch up with my mail backlog :/)
09:39:31amiconnI think even without big savings it sounds like a good thing to do
09:40:13amiconnIt will consolidate the menu handling
09:40:32LinusNi agree
09:40:45amiconnThere are some differences in voice handling in menus vs. settings I wonder about
09:40:46JdGordonim just scared that ill do all this work and it will get rejected because it does blow out the size by a bit..
09:41:41amiconnI'm more concerned about devs stopping their rework half-way through, leaving the system without a real benefit
09:41:51amiconnThere are now at least two examples for this
09:41:58LinusNJdGordon: if we reject it, it will not be because of the size, only because we don't like you ;-)
09:42:11JdGordonof course..
09:42:40amiconn(1) the gui stuff is still not ported to all screens (some not at all, some in a really hackish way)
09:42:55JdGordon3 guesses what 2) is.. :p
09:43:02amiconn(2) actions also need more work (okay, actions aren't in for very long yet)
09:43:07*JdGordon thinks he hears someone calling
09:43:27LinusNJdGordon: that's the hitman we sent
09:43:29amiconnlangv2 is the 3rd
09:43:41amiconn(but the actual rework didn't start yet on that one)
09:44:08*amiconn is trying to catch up with mails :(
09:44:12 Join webguest53 [0] (i=c27f0812@gateway/web/cgi-irc/
09:46:47webguest53Hi. What are the primary RB goals atm (besides porting it to new players)? And what are the criteria for picking up and committing patches? There are very simple ones (e.g. 5738) and there are rather complicated (e.g. R.E.P.) And I saw a discussion about a new screen "management" system a couple of days ago. So what comes next?
09:47:06Bagderwe don't have "primary goals"
09:47:14***Saving seen data "./dancer.seen"
09:47:16JdGordonapart from "take over the world"
09:47:18Bagdermost of us are in it for the fun
09:47:46 Quit dan_a (Read error: 110 (Connection timed out))
09:49:00 Join daurn|laptop [0] (n=quae@unaffiliated/daurnimator)
09:49:34 Join dwihno [0] (n=dw@
09:53:02webguest53So how do you decide what patch to commit? Is it personal preference of a dev ("I like this feature...")?
09:53:38JdGordonno. they look at the name of the develper and then as long as its not me they accept it...
09:53:45LinusNwebguest53: more or less, but it's also up to the patch creator to make the patch work on all targets
09:54:21LinusNthe patches are often incomplete, and can't be committed without much work
09:55:49 Join safetydan [0] (n=dan@
09:57:45webguest53LinusN: well, the 5738 is complete. I use nimbus-12 and was happy to find that patch. As I was distracted by the round W (you know, once you've seen it once you're aware of its ugliness :-)
09:59:37webguest53LinusN: I also want to create a bigger nimbus (e.g. nimbus-16 or 18 or 20). I like the shape but 12/14 is too small for me. BTW: 12 and 14 are almost the same, only line spacing is different
10:00:05markunwebguest53: good fonts are always welcome
10:00:43markunAnd it would also be nice if you could add some cyrillic and hebrew to your font
10:03:57*JdGordon is buying dark tranquility tickets tomorow \m/ :D \m/
10:06:02blxis there any rockbox simulator for gnu/linux avaiable?
10:06:24 Quit dwihno_ (Read error: 110 (Connection timed out))
10:06:48markunyou can compile it from the rockbox source if you have SDL installed
10:06:56webguest53markun: there's also a version of nimbus-12 and 14 with cyrillic and hebrew (and some other) glyphs in the patch tracker. I've used them, worked well.
10:07:06blxthank you markun
10:07:42markunwebguest53: I think we should update the rockbox font then
10:08:34 Join spiorf [0] (
10:14:53 Join oxygen77 [0] (i=c1c2842b@gateway/web/cgi-irc/
10:16:03oxygen77amiconn: would you be here for a question on MAS ?
10:16:29daurn|laptopno - but he would be here for other questions
10:16:42 Join ender` [0] (i=null@
10:18:03markundaurn|laptop: there was a guy working on a port for another DM320 player. Did he talk to you about it?
10:18:04 Join dan_a [0] (
10:18:28markunI have to check the logs
10:18:29BagderI talked to him a bit
10:18:37daurn|laptopwas it for mrobe
10:18:42Bagderguided him to the first steps of configure fixing etc
10:18:49Bagdermrobe yes
10:18:54daurn|laptopok, was shirour
10:19:00daurn|laptopi haven't talked to him in AGES
10:19:03 Join barrywardell [0] (
10:19:05 Part webguest53
10:19:12Bagderbarrywardell: congrats on the sound!
10:19:33barrywardellthanks Bagder. and thanks to MrH yet again!
10:19:40 Join bbroke [0] (
10:20:34dan_abarrywardell: Well done!
10:20:35markundaurn|laptop: he was asking me for help with the same things I helped you with. So I just thought it would be good for you guys to work together.
10:21:00dan_aHow come it isn't in the News section on the front page?
10:21:13Bagderthose darned lazy web guys
10:21:26daurn|laptopmarkun: sure :P - if you can find him
10:22:57markunbarrywardell: I'm unifying the wolfson code. Could you take a look some time and make sure I keeps working on the H10?
10:23:17barrywardellmarkun: yeah sure.
10:23:37barrywardellis that the patch you showed me before? i can't remember the url
10:23:45Bagdershirour's email is on that first page
10:24:27dan_aWould there be any objections if I was to split out the PortalPlayer section of crt0.S into a separate file (crt0-pp.S)? It makes the remaining code more readable, and linuxstb suggested on Friday that it would be a useful thing to do.
10:24:45markunbarrywardell: yes:
10:24:45Bagderimho, the ctr0.S one should be split into separate ones for separate CPUs
10:24:46LinusNdan_a: i think it should move to the target tree
10:24:50barrywardelldan_a: I vote in favour of that
10:25:06Bagderor at least for separate archs
10:25:08markunBagder: and all the targets should be changed to target tree
10:25:13dan_aLinusN: Where in the target tree?
10:25:33markunThere is a patch for the ipod 4g to target tree
10:25:41LinusNdan_a: probably target/arm/ipod
10:25:55barrywardellmarkun: I'll try that later today and let you know
10:26:27markunbarrywardell: ok, don't forget to change wmcodec_write again
10:26:27LinusNor maybe even target/arm
10:26:39daurn|laptopum... i need help with my port
10:26:39Bagdertarget/arm I'd say
10:26:44daurn|laptopi haven't touched it in a week
10:26:48dan_aLinusN: It isn't ipod specific - H10 and Sansa use it too - so target/arm sounds best
10:26:51daurn|laptopand i've forgottern wtf i was doins
10:27:07barrywardellyes, target/arm - it applies to H10 and Tatung and others too
10:27:10daurn|laptopok, current build error(s) are for buttons
10:27:25Bagderdaurn|laptop: are you really the only gmini402 owner who wants to see Rockbox on it?
10:27:43markuntarget/arm/portalplayer ?
10:27:47daurn|laptopBagder: the only one thats not a total noob
10:28:43barrywardellthere's already ata-pp5020.c in target/arm which is specific to PP5020
10:29:17barrywardellso keep the same idea for crt0.S too?
10:29:30dan_abarrywardell: Sounds good
10:30:15daurn|laptopBagder: search for rockbox
10:30:30dan_aFor now I'll leave the other code in crt0.S, including the generic ARM code which I think is for Gigabeat
10:30:31 Quit akaias (Operation timed out)
10:30:47Bagderno results found
10:31:06 Join akaias [0] (
10:31:22daurn|laptopi found lots
10:31:33Bagdercrappy forum them
10:31:44BagderI bet it behaves differently if you're logged in
10:31:50daurn|laptop31 threads
10:33:05daurn|laptopjoin if you want
10:33:29BagderI found a rockbox link on the front page, but that just loads and loads and never gets anywhere ;-)
10:34:07Bagdermight be a very slow site perhaps
10:34:22Bagderdoesn't like me at all
10:34:46daurn|laptopi've had no other complaints
10:34:49Bagderdead it seems
10:35:29Bagderfrom two entirely different networks I can't get the front page anymore
10:35:44daurn|laptopsame for me
10:35:48daurn|laptopjust last minute
10:36:02Bagderit connects so its the web server that is stupid
10:37:23 Quit barrywardell ()
10:37:38oxygen77anybody else than amiconn could help me on MAS config ?
10:38:23LinusNjust ask the question instead of asking for pepole
10:38:59LinusNpeople, evel
10:39:05LinusNeven, even :-)
10:41:27oxygen77ok, I have some doc on MAS read it ... There is some missing info in them, for the codec config regs 0x2 to 0x5 have no description
10:42:24oxygen77I have now a code that runs on MAS that apparently needs some config, most of it is done by setting value in memory but it also uses reg (0x1 (which is notdocumented)
10:43:12oxygen77wondering if you anybody would have info on this rge 0x1 in codec conifg of MAS
10:46:22LinusNhmm, it's been a long time since i worked with the MAS, but i don't remember any codec config regs
10:47:18LinusNor are you talking about CONV_CONF and friends?
10:48:28oxygen77hum, talking of the reg used to configure the DAC, ADC, volume, bass, ...
10:49:47LinusNin the data sheet, 0001-0005 are undefined
10:50:04oxygen77yup, that's my prob
10:51:14oxygen77the code I use need some config in reg 0x1 in order to set the sample-rate
10:51:40oxygen77this code is for PCM playback from first SDI port which connected to TMS320 DSP on av3
10:54:36linuxstb_Bagder: Do you think MrH would be interested in helping to work out the Elio's LCD?
10:54:41 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
10:55:09Bagderdo you have the mi4 lying around?
10:56:02LinusNoxygen77: hmmm, i'm not sure how we could find this out
10:56:20linuxstbYes - I'll upload it somewhere for you. I don't have any bootloader at the moment, but I could try dumping the flash contents to disk.
11:03:15linuxstbBagder: Something that's also confusing me is the memory remapping that happens at the very start of the firmware. The PP5020 has a set of 8 pairs of memory-mapping registers. The Apple firmware for the ipods just does one remapping - remapping the whole of SDRAM from 0x10000000 to 0x00000000. The Elio firmware appears to do 2 mappings, but I don't understand what. There also appears to be a lot of references all over the firmware to
11:03:15linuxstbmemory locations in both the 0x00000000 and 0x10000000 ranges, so I'm guessing it splits the memory somehow. I would be interested to know if MrH has seen this in the other PP firmwares he's looked at.
11:04:01BagderI don't recognize that from our discussions at least
11:04:07amiconnoxygen77: I have no idea what the undocumented codec regs 2 thru 5 do
11:04:16 Quit firenx ("Leaving")
11:04:26amiconnWe don't need them for the pcm codec to work
11:05:17 Join MarcoPolo [0] (
11:05:41 Join aliask [0] (
11:09:27oxygen77ok thx amiconn, it is used in another pcm codec from archos, which I need as it reads data from SDI not SDIB as the codec you use
11:19:17 Quit fejfighter (Read error: 104 (Connection reset by peer))
11:32:33 Join _FireFly_ [0] (
11:33:59dan_alinuxstb: Are you OK with the PortalPlayer section of crt0.S being moved to target/arm/crt0-pp.S (as well as the other changes I suggested on Friday)?
11:34:21dan_aIf so, I'll do a bit more testing and then commit that later today
11:36:42 Quit dwihno (Read error: 110 (Connection timed out))
11:38:15 Join ghode|afk [0] (
11:38:16preglowdan_a: i don't see a problem with that, crt0.S is too convoluted anyway
11:38:44 Quit daurn|laptop (Read error: 110 (Connection timed out))
11:40:13 Join TeaSea [0] (
11:41:00linuxstbdan_a: Definitely.
11:42:26preglowjust take care to not move all the arm stuff too
11:43:19preglowthough crt0.S would most definitely benefit from being split severely up
11:43:24dan_apreglow: I have done - I've removed the things that are #ifdef CPU_PP from crt0
11:43:36dan_abut CPU_ARM is still there
11:44:41*preglow curses his cold, autoconf and c++ in quick succession
11:45:15 Join PaulJam [0] (
11:47:16***Saving seen data "./dancer.seen"
11:57:23 Join webguest57 [0] (i=c27f0812@gateway/web/cgi-irc/
11:57:34 Quit Snyper (Read error: 110 (Connection timed out))
11:58:39webguest57Hey, the newly committed font nimbus-14 has an error: the W is bad. I think the last line in the chardef should be F380, not FF80!
12:04:56markunwebguest57: oops
12:05:04 Join daurn|laptop [0] (n=quae@
12:05:11markunOr was it not somthing I did?
12:06:30markunwebguest57: I see F380
12:07:08markunAh, nimbus-14..
12:08:10linuxstbAnyone know if I am right in thinking that the image data in a .bmp file is in reverse order - i.e. the bottom row of pixels first?
12:08:45webguest57markun: or nimbus-12 is wrong. They should be the same!
12:08:55 Join mirak [0] (
12:09:28markunwebguest57: well, nimbus-12 was updated by LinusN today to change it to F380
12:09:39webguest57markun: with FF... it looks even better IMHO
12:09:45markunI'll check when I get back home
12:10:15markunthere are also some problems with the space between the cyrillic glyphs
12:10:22webguest57markun: but that change corrected C to F. Now we should correct also 3 to F :-)
12:10:59markunwebguest57: you could check the whole font if you want
12:11:06markunmaybe you find some more problems
12:12:14markun"Pixels are stored from the bottom to the top of the images, and then from the left to the right
12:12:31 Join barrywardell [0] (i=892b7a6e@gateway/web/cgi-irc/
12:13:14linuxstbmarkun: Thanks. That's very weird then - I've been looking at a dump of the flash in my Elio, and there are 24-bit bitmaps there, stored in the same byte order as .bmp files.
12:13:43markunnice that you got a dump!
12:14:17linuxstbi.e. I took the 54 header bytes from a 24-bit 220x176 .bmp file, appended 220x176x3 bytes from the flash image, and I get the main bitmap the original bootloader shows.
12:14:47linuxstbmarkun: Just a simple write(fd,(char*)0,1024*1024) in my test bootloader. (the standard PP5020 ATA driver works).
12:15:06 Join daurn|afk [0] (n=quae@
12:16:31linuxstbIn fact, the .bmp header is in the flash image as well :)
12:16:44 Quit PaulJam (Connection reset by peer)
12:20:53 Quit MarcoPolo (Read error: 104 (Connection reset by peer))
12:21:47 Quit daurn|laptop (Nick collision from services.)
12:21:50 Nick daurn|afk is now known as daurn|laptop (n=quae@
12:22:00 Quit webguest57 ("CGI:IRC (EOF)")
12:22:20daurn|laptopmy powers gone out 4 times in the last 10 minutes
12:22:50 Quit spiorf (Remote closed the connection)
12:23:00amiconnlinuxstb: BMP files are definitely bottom-up; there are numerous places in rockbox with a comment stating this
12:23:08amiconn(screendump, bmp loader...)
12:23:20preglowi think even the windows bitmap layout is like that
12:23:23linuxstbamiconn: Yes, I saw those shortly after I asked the question....
12:23:25preglowas in internally
12:24:17linuxstbBut the mystery is solved - the bootloader contains the .bmp headers, as well as the image data. So it makes sense for it to be reversed (although not for them to use .bmp files...)
12:24:38amiconnThey _can_ be top-down afaik, but it's rather unusual
12:29:04 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
12:30:05 Quit amiconn (Nick collision from services.)
12:30:06 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
12:36:40 Quit daurnimator (Read error: 110 (Connection timed out))
12:38:22safetydanJdGordon, has House started yet?
12:39:20JdGordoni tinhk its a repeat
12:40:05safetydanit's all new to me
12:40:18safetydanlousy Australian Idol going late
12:40:49 Join MarcoPolo [0] (
12:41:54Bagder <= spot the difference!
12:43:06safetydanThe page takes an extra second to load? :)
12:43:30Bagderyeah, due to two new images...
12:44:11dan_aI'm not near a DAP at the moment - could anyone confirm that everything still works if you apply - it looks right, and compiles cleanly, but I want to be sure.
12:44:22safetydanSo H10 gets the official build blessing. Woo!
12:44:44safetydanso many choices for buying a new dap
12:45:34 Quit barrywardell ("CGI:IRC (Ping timeout)")
12:45:45 Quit linuxstb (Read error: 110 (Connection timed out))
12:46:03 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
12:47:07 Join lorijho [0] (n=lorijho@
12:47:23 Join PaulJam [0] (
12:48:56linuxstbIs there an installation howto etc for the H10 available if anyone asks?
12:50:18Bagderbut I guess the h10 team will provide some prebuilt bootloaders when they feel the time is ripe
12:52:01 Quit oxygen77 ("CGI:IRC (Ping timeout)")
12:53:18lorijhoBagder: we will
12:54:40linuxstbI guess we also need to figure out a way to load the original firmware. I can try to do that for the Elio as well - I'm guessing the same method will work on the H10.
12:55:30linuxstbDumping the register contents at the very start of our bootloader should hopefully tell us what we need...
12:57:13lorijholinuxstb: it probably would
12:58:32lorijhoHow well do the pp5020 based players deal with ogg and flac decoding? Volume is very low on H10 so I can't really tell yet.
12:59:01linuxstbIt works fine. FLAC is very efficient, Ogg is reasonable.
12:59:31lorijholinuxstb: any suggestions for -q when encoding ogg ?
12:59:50linuxstbNo, I don't use it. I know others here do though...
13:00:40lorijhothe oggs I encoded in the past have all -q 8.5, I wonder if it isn't too much
13:01:42 Join Mmmm [0] (
13:02:41MmmmjhMikeS: are you about?
13:02:51mirakblock is a pointer to an array of int16_t. the C version is doing " int i = block[0] " to get the first word, then "int j = block [1] to get the second word." now in asm I put the adress of block into a6 register. I do "move.w (a6),d0" to get block[0], then "move.w (1,a6),d0" to get block[1] . Is this right or will I have some problems with endian ness on coldfire ?
13:03:32mirak "move.w (1,a6),d1"
13:03:39preglowlinuxstb: i've finally gotten a 24 bit flac file i like to listen to
13:03:47preglowlinuxstb: so i might end up doing some work in that apartment again
13:04:16 Join Paul_The_Nerd [0] (
13:05:10linuxstbpreglow: OK, cool. What's the boost ratio at the moment?
13:09:29 Join CrossF [0] (
13:09:50CrossFsomeone know's when there will be rockbox for Sansa e200?
13:10:05mirak the assembly function idct_row is at line 327 . I tried in that one to replicate the C version. I have read it several time, I am out of ideas about what can be wrong
13:10:22BagderCrossF: nobody knows
13:11:28 Quit theli_ua ("Leaving")
13:12:49 Quit CrossF (Client Quit)
13:18:55 Quit PaulJam (Read error: 60 (Operation timed out))
13:19:29mirakwhat's the better place to have my code reviewed ?
13:21:07preglowlinuxstb: haven't tried on target yet
13:21:22preglowi'll give it a go right now
13:22:34 Join PaulJam [0] (
13:24:37preglowlinuxstb: never boosts on coldfire
13:25:11 Join theli_ua [0] (
13:25:20linuxstbWhat samplerate is it?
13:25:34preglowso not that much of a surprise, granted
13:28:07pregloweven ipod takes it in stride
13:28:08preglowso hooray
13:29:33 Quit Sinbios ("If the definition of a klutz is someone who doesn't have eyes on their ass, then yes, I suppose I am a klutz.")
13:36:32linuxstbI don't think FLAC compresses 24-bit very well though. You'll probably be better off converting to wavpack.
13:37:03preglowit doesn't
13:37:06preglowso yeah
13:37:08linuxstbI think it's an encoder issue, rather than a problem with the format. I think Josh is planning to improve it before the next release.
13:37:10preglowbut i can't be bothered :>
13:38:28 Join GliGli [0] (
13:41:20daurn|laptopwhy does time go SO slowly
13:41:37 Join webguest79 [0] (i=c0647cda@gateway/web/cgi-irc/
13:41:42daurn|laptopi feel like i've been waiting for thins for weeks
13:41:52daurn|laptopbut its been only a day(or 2)
13:43:05 Join oxygen77 [0] (i=c1c2842b@gateway/web/cgi-irc/
13:44:57 Quit Mmmm (Read error: 110 (Connection timed out))
13:47:17***Saving seen data "./dancer.seen"
13:47:43 Join Criamos [0] (
13:48:33 Join tucoz [0] (n=martin@rockbox/developer/tucoz)
13:49:08tucozare there any feature differences between the h10-5gb and the h10-20gb?
13:49:30tucozlike recording, radio etc
13:51:19tucozi.e. do we need two different manuals for those players?
13:56:02 Join lorijho_ [0] (n=lorijho@
13:56:23 Join Arathis [0] (
13:57:01*Bagder is the human proxy
13:57:17Bagder <= anyone like for configure menu ?
13:58:30 Quit Hideo_Kuze ("tada")
13:58:45BagderI've updated all (I hope) scripts now so we can change the numerics in the menu without killing all autoscripts
13:58:57preglowjust make sure it fits in 80 columns
13:59:01Bagderoh yes
13:59:12BagderI'm priest in the 80 column church
13:59:35Bagderthis is 63 columns at the longest line
14:00:12 Quit safetydan ("Ex-Chat")
14:00:21lorijho_Bagder: nice
14:00:43Bagderit'll be a lot easier to find in that one
14:01:49LinusNwhat irc nick does mark arigo have?
14:01:53PaulJamthis is maybe a silly question, but do i have to run configure again, if this gets committetd (because h300 was 10 previously)?
14:02:06BagderPaul_The_Nerd: nope
14:03:16 Join [San] [0] (
14:03:28Bagderuh, I mean that for PaulJam
14:03:42*Bagder can't even use tab completion properly
14:06:00 Quit [San] (Client Quit)
14:06:39 Quit lucas42 (Read error: 104 (Connection reset by peer))
14:07:47linuxstbLinusN: lowlight IIRC
14:07:52 Quit lorijho (Read error: 110 (Connection timed out))
14:08:01LinusNlinuxstb: ah, now i remember
14:08:15tucozLinusN, 20.06.08 # <lowlight> If anyone want's to test:
14:08:28tucozthe path reveals his true identity ;)
14:08:50 Nick lorijho_ is now known as lorijho (n=lorijho@
14:09:15 Join Genre9mp3 [0] (
14:09:52 Part oxygen77
14:10:19 Part Paul_The_Nerd
14:16:31elinenbeso, time to commit some of the nice WPS patches :-)
14:19:37 Join RoC_MM [0] (
14:19:59 Join Lear [0] (
14:22:53LearAnyone who has looked at this patch:
14:23:43preglowi want it commited, but i have no idea about the playback system
14:23:51preglowi've verified there is a problem, anywya
14:28:55LearI was thinking about committing it (will give it a try at least). Had a freeze probably caused by the bug it fixes.
14:29:29LearI'm don't know that much about the playback system, but the patch looks okay to me...
14:32:52maddog011hey, what's the progress on the H10 20GB player?
14:35:22dan_amaddog011: There is sound on it since last night. As I understand it the quality is not brilliant, but it's a big step forward
14:36:24 Join infamis [0] (i=4b0377fc@gateway/web/cgi-irc/
14:36:47maddog011dan_a: omg, you got sound out, congratulations !
14:36:51 Join Xen0n [0] (i=WinNT@
14:37:01maddog011dan_a, did u fix the touchpad
14:37:18maddog011dan_a, so it can detect the upper movement
14:37:24dan_amaddog011: Not me - barrywardell and lorijho are the guys working on it
14:37:32infamiskeep getting a compilation error with an x5 build (devel w/logf enabled) in mp3_enc:
14:37:44infamismp3_enc.c: In function `filter_subband':
14:37:48maddog011awsome, barry is really doing a good job
14:37:59infamismp3_enc.c:2431: error: can't find a register in class `ADDR_REGS' while reloading 'asm'
14:38:19infamisI've made veryclean && re-ran configure
14:38:24infamisstill no go...updated cvs
14:39:28Arathismaddog011: tochpad isn't fixed till now
14:39:34dan_ainfamis: I got a similar error on iPod a while ago when I wasn't using the recommended version of GCC
14:39:50infamisdamn...must be a cygwin thing (which I forgot to mention)
14:40:46infamisI noticed the bleeding edge builds fine so didn't know what could cause it with a fresh cvs repo
14:41:33 Quit apo` (Remote closed the connection)
14:41:51dan_aFrom the CrossCompiler wiki page in the iAudio section: "GCC 4.0.x seems to work fine when building Rockbox for Coldfire platforms. For the time being, it is not the recommended or preferred version to use. It should generate valid binaries, but if you run into problems when using 4.0.x, try the 3.4.6 version instead."
14:41:53 Quit RoC_MM ("Leaving")
14:41:53LinusNpreglow, Lear: let's assume that the patch author knows enough
14:41:54 Join bluebrother [0] (
14:42:02infamisgcc version 3.4.4 (cygming special) (gdc 0.12, using dmd 0.125)
14:42:33infamisI guess up to 3.4.6?
14:43:38dan_ainfamis: It's probably worth a try
14:43:39lorijhomaddog011: we do not yet detect the touching position on the scrollpad (or whatever that thing is called)
14:44:21 Quit Xen0n ("-=SysReset 2.53=-")
14:44:49infamisyea when I installed cygwin I didn't keep track of versions or anything, everything's been building (been building manually since about may) fine until now
14:45:48infamisthanks dan_a
14:45:51 Quit infamis ("late for school...")
14:47:40daurn|laptopi need some help please
14:47:52daurn|laptophow do you do a button driver?
14:48:38markundaurn|laptop: easy
14:49:02markundid you check the button-*.c file of any of the target tree players?
14:49:18daurn|laptopi copied yours across
14:49:21markunThen what's the problem?
14:49:25daurn|laptopand customized it for my button set
14:49:37markunthat should be all
14:50:03markunyou also need to have button-target.h
14:50:09daurn|laptopi did the .h
14:50:18daurn|laptopand i didn't understand the structure of the .c
14:50:56markunThe driver in firmware/drivers/button.c is generic and uses your functions
14:51:46daurn|laptopbtw, where is the button base address defined?
14:51:54markunwhat's that?
14:52:25daurn|laptopthe address that all button are offset up
14:53:06markunsometimes buttons are connected to the ADC, some to the IO pins
14:53:10markundepends on the target
14:53:53 Quit Farpnut (Read error: 110 (Connection timed out))
14:57:38 Quit aliask ("Chatzilla 0.9.72 [Firefox]")
15:05:01 Join spiorf [0] (
15:06:25 Join stamppot [0] (i=d4883814@gateway/web/cgi-irc/
15:08:15 Quit TCK (Read error: 104 (Connection reset by peer))
15:17:55 Quit eGen ("mýdlo končí ! ... ručník končí !")
15:20:17markundaurn|laptop: how are the buttons connected on your gmini402?
15:21:01daurn|laptopdpad (4 way) + 2 on left + 3 on botton + 2 on right
15:21:12daurn|laptopso..... 11
15:21:17 Join eGen_ [0] (i=generat0@
15:22:26markunand how do you read the buttons?
15:22:48 Join eddddddy [0] (i=deefbf35@gateway/web/cgi-irc/
15:23:10eddddddyim eddy
15:23:33daurn|laptopsee if some hardware addresses are true or not
15:24:54eddddddyi just want to know my pod version
15:25:22 Quit XavierGr ()
15:25:24eddddddyIn tells me 1.2 ///
15:25:33 Join pixelma [0] (i=pixelma@
15:25:46eddddddyis it newest version?
15:29:05 Quit eddddddy ("CGI:IRC (EOF)")
15:31:15 Join apo [0] (
15:33:37pixelmatucoz: I made a new rockboxlogo.svg and already updated the ipod drawings to use it... seems to work now
15:34:02tucoznice :) did you notice that there is a new target as well?
15:34:34pixelmayes - I have to do the H10 too... any nice scans of it yet?
15:34:39 Join rretzbach [0] (
15:35:50tucozhehe. not that i know of
15:37:12 Quit spiorf (Remote closed the connection)
15:40:21 Quit Xerion (" ")
15:42:55 Join thegeek [0] (
15:42:59 Join lorijho_ [0] (n=lorijho@
15:43:38 Part LinusN
15:46:06 Join [San] [0] (
15:46:57 Join webguest42 [0] (i=c182ea0f@gateway/web/cgi-irc/
15:47:21***Saving seen data "./dancer.seen"
15:47:34 Quit webguest42 (Client Quit)
15:47:41 Join webguest53 [0] (i=c182ea0f@gateway/web/cgi-irc/
15:48:21 Quit webguest53 (Client Quit)
15:56:34 Quit lorijho (Read error: 110 (Connection timed out))
15:56:49 Join Helen_42 [0] (
16:00:09 Quit Helen_42 ("Chatzilla 0.9.75 [Firefox]")
16:00:15 Join tuwe [0] (n=the_fuch@
16:02:25tuwei wonder if anyone can explain the following.
16:02:37tuwei have todays cvs source
16:02:49tuweand i applied the following patches:
16:03:49tuwethe latter didn't work perfectly, so i modified the sources by hand, as given in the corresponding *.rej files.
16:04:00 Join thegeek_ [0] (
16:04:18tuwethen i compiled the code to obtain a simulator. it worked perfectly.
16:04:31 Quit thegeek (Read error: 104 (Connection reset by peer))
16:04:51tuwei didn't experience any of the bugs mentioned on the album art page on rockbox flyspray
16:05:14tuwei'm on an ipod mini 2g
16:05:21tuweso i started creating a wps
16:05:49tuwethe problem is: album art doesn't show up, although i can set the cover.bmp as background and then it works.
16:06:07tuwefurthermore, i cannot view any bmp in the bmp viewer.
16:06:16tuweis this a bug in the bmp viewer?
16:06:27 Quit MarcoPolo ("Bye !")
16:07:24tuweor is it just because bmp viewer doesn't support monochrome bitmaps?
16:07:44tucozhang on and maybe someone can help you. i can't though.
16:07:54tucozi didn't even know there was a bmp viewer
16:08:34tuwei'm not sure if there is one. it think i read it on some pages.
16:08:54daurn|laptopcyas ALL
16:08:59tuwemaybe it's something with the bmp resize function
16:09:01tucozthere is a bmp-loader function in rockbox
16:09:41tucozand that will load native bmp's (i.e. bmp in your players resolution and bpp)
16:10:01preglowLear: so, we should just commit it?
16:10:16tuwemaybe that's what i was talking about? but the regular bmp loader seems to work, the wps on are all working.
16:10:21 Quit thegeek_ (Read error: 113 (No route to host))
16:10:48Genre9mp3tuwe: Or maybe the album art 5.10 (which btw has a "TEST" in caps in its name) breaks the bmp viewer...
16:11:14Genre9mp3tuwe: Why not use the album_art_v5.01.patch instead?
16:11:40tuweyou mean with the latest cvs and stuff?
16:11:43 Join thegeek [0] (
16:12:00tuwei wonder why i didn't think of that before :)
16:12:13 Quit Landus ("Leaving")
16:12:16tuwei think i tried 5.01 with an older source version
16:12:17Genre9mp3tuwe: If 5.01 is not outdated why not?
16:12:25 Nick BigMac|comatose is now known as BigMac (
16:12:30tuweok i'll give it a try.
16:14:49Genre9mp3tuwe: quote from the author of 5.10 patch: "I did some more tests and found some horrible bugs! So give me some time to fix them... −−> ATM: DON'T USE THE PATCH v5.1.0!!!"
16:17:19tuweyes i read that too, but since the older patches didn't seem to work (today is my second day of cvs and patching), i thought i'd try, and i even put it on my ipod and didn't run into any bugs yet.
16:17:57 Quit chendo_ (Success)
16:18:26 Join chendo_ [0] (
16:18:56 Quit PaulJam (".")
16:20:10 Join Febs [0] (
16:20:43 Join Landus [0] (
16:23:40 Join daurnimator [0] (n=daurnima@unaffiliated/daurnimator)
16:24:24 Quit thegeek (Read error: 104 (Connection reset by peer))
16:25:31 Join thegeek [0] (
16:26:48 Join ootput [0] (n=ootput@tor/regular/five0)
16:29:35tuweGenre9mp3: ok i have recompiled the simulator with album art patch 5.01, but i still get the same results when viewing a bmp. :-/ seems like i have to wait for someone who knows what he's doing.
16:31:38Genre9mp3tuwe: I use an older version of album art patch to tell you the truth (without the resize stuff etc.) I just gave you an idea...
16:31:57Genre9mp3Maybe someone else has something better in mind
16:32:13theli_uatuwe, i'm almost sure album art patch is only for color targets
16:33:09tuweGenre9mp3: ok i'll try that too, thanks for the help
16:34:12 Join netmasta10bt [0] (
16:34:32tuwetheli_ua: it somehow works, because the ipod mini simulator recognizes the tag, and something the expected size of the cover shows up, but it looks all screwed up.
16:37:39 Quit thegeek (Read error: 104 (Connection reset by peer))
16:37:43 Join thegeek [0] (
16:39:08 Join dwihno [0] (n=dw@
16:46:52 Join Crackerizer [0] (n=root@
16:47:27Crackerizerhello, dan_a.
16:48:16dan_aHello Crackerizer
16:49:48Crackerizerdan_a: do you have menu-delay problem on your 3g with rockbox?
16:50:58dan_aNone that I've noticed. When I go into the plugins menu there is a small delay while the disk spins up, but that is it.
16:51:41dan_aWhat kind of delay are you getting?
16:52:01Learpreglow: I think so, but I haven't really tested it much yet. :) Basic playback seems fine, but I'd like to play a bit with seeking first...
16:52:50Crackerizerdan_a: i dont know. I just feel that it's not smooth as the original firmware.
16:56:30 Quit thegeek (No route to host)
16:58:47Crackerizerdan_a: is 3g your main rig? or it's just for rockbox development?
17:00:32dan_aIt's my main rig, but I spend more time trying to make Rockbox play without skipping on it than I do listening to music
17:00:41 Join thegeek [0] (
17:00:48 Quit stamppot ("CGI:IRC")
17:03:49dan_aThere are a couple of things I'm working on at the moment, once they are out of the way I will see if I can see a problem with the menus, and if there is anything I can do about it.
17:03:54Crackerizer:), same here. after getting to know what rockbox is, it sits on my table and waiting for a new built of rockbox.
17:04:10 Quit theli_ua ("Leaving")
17:06:48CrackerizerI'm learning how rockbox works. hopefully, i have more time on the long holiday.
17:10:18BigMaccan %al an xy coordinate?
17:10:55 Join dishnecto [0] (
17:11:13dishnectohey can someone help me
17:11:19dishnectoi put a them on my ipod
17:11:26 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
17:11:32dishnectobut the writing is not in place
17:11:39dishnectoits all over the place
17:11:41Learpreglow: Hm... Got a codec failure when seeking backwards in a flac. MP3 seems fine though.
17:12:28dishnectolike the them is good but the fonts or not in the box's u know what i mean
17:14:38BigMacdishnecto:download the fonts package and place it inDrive:rockbox\root
17:15:04BigMacthen it should load the right font
17:15:54dishnectowhat i mean is when i am playing a song the time and everything is not in the boxes its all over the place
17:16:56 Join MarcoPolo [0] (
17:20:54dishnectoi dont have the folder roots
17:21:39bluebrotherdishnecto, that is the point. The display is line oriented, thus when using a wrong font (happens usually when the requested font isn't installed) everything shifts around
17:21:51bluebrotherdownload the fonts package from the daily builds page.
17:22:02bluebrotherand put its content to /.rockbox/fonts
17:22:53bluebrother"root" is the root of your player, i.e. the folder you're in when you didn't enter a folder. The topmost folder of your drive.
17:25:16 Join senab [0] (
17:28:25 Join Kingstone [0] (i=Snape@
17:30:36BigMachey senab
17:30:40BigMacgot a sec
17:30:54 Quit JdGordon ("Konversation terminated!")
17:31:39 Join barrywardell [0] (i=892b7a6e@gateway/web/cgi-irc/
17:32:46dishnectooh thx works good now
17:32:52barrywardellBagder: what do you think about hosting a build of the H10 bootloader mi4?
17:33:35barrywardellIt's not something that will likely change until we get it to load the OF or dan_a makes more COP changes
17:34:13dishnectohey guyz and how do i appley patches
17:35:33 Quit senab (" HydraIRC -> <- IRC has never been so good")
17:36:30tuwedishnecto: take a look at
17:37:58 Join Joey_D [0] (
17:38:12Joey_Danyone home?
17:38:16dan_abarrywardell: I'm pretty sure I shouldn't need to touch the bootloader again.
17:38:41dan_aJoey_D: Some of us are!
17:38:55Joey_Di have a quick ipod rockbox question
17:39:30 Join pondlife [0] (
17:39:47bluebrotherJoey_D, just ask −− if someone who knows is around you most probably get an answer ;-)
17:39:56Joey_Dthe first and second generation minis are mostly identical
17:40:42Joey_Dwhat would happen if 2G rockboxs was put on a 1G mini?
17:41:54 Join Kohlrabi [0] (
17:41:54 Quit apo (Read error: 104 (Connection reset by peer))
17:46:58preglowbarrywardell: so, anything big left before you're willing to announce the h10 port more or less fully functional?
17:47:03dan_aJoey_D: I would imagine that it wouldn't work - why?
17:47:24***Saving seen data "./dancer.seen"
17:47:37Joey_Di have a 1g mini and was wondering
17:48:03barrywardellpreglow: the lcd is still quite glitchy
17:48:18barrywardelland the scrollpad doesn't work properly
17:48:55barrywardelland it stutters with the default wps
17:49:25barrywardelland we don't have proper power control
17:49:32Learpreglow: I've committed patch 5879 now.
17:49:36preglowLear: so i see
17:49:40*barrywardell wishes the list was shorter!
17:49:42preglowi think i'll have a look at the musepack seek patch right now
17:49:50preglowbarrywardell: now where would the fun in that be!
17:50:54 Join adiamas [0] (n=adiamas@
17:51:16tuweJoey_D: i'm not sure but you could possibly try it out by compiling a) a 1g simulator and b) a 2g simulator and swap the archos directory
17:51:29tuweJoey_D: see for how it works.
17:51:44Joey_Dthank you
17:51:48adiamasanyone know off hand what tags the tag cache looks at?
17:51:50adiamasv1, v2?
18:01:01Learadiamas: ought to be what you have set in settings.
18:01:55 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
18:02:21 Quit BHSPitMonkey (Read error: 104 (Connection reset by peer))
18:02:51barrywardellmarkun: just tried your wmcodec patch
18:03:01barrywardell I had to make a couple of changes but it works fine otherwise
18:04:12 Join apo [0] (
18:04:50adiamaswill there be some message when the tag cache is ready?
18:05:53 Join oxygen77 [0] (
18:06:53bluebrotheradiamas, no. But you can see the status in the debug menu if you need to.
18:06:55 Join linuxstb_ [0] (i=d57b9aa9@gateway/web/cgi-irc/
18:07:07*Bagder makes Lear aware of a line of yellow columns
18:07:21adiamasthanks bluebrother
18:07:26adiamasBagder.. long time no see
18:07:36Bagderhey adiamas, good to see you alive! ;-)
18:08:09linuxstb_Joey_D: I _think_ the main difference between the 1g and 2g mini versions of Rockbox is the button driver - the button hardware is different (internally) between the two.
18:08:19Bagderah, the bleeding edge builds work for the h10 packages too now
18:08:45adiamasheheh its been a while...
18:08:57 Part oxygen77 ("Leaving")
18:09:26*Lear has already noticed that, and is working on it. :)
18:09:51linuxstb_Bagder: Would it be difficult to create .mi4 files using scramble? Am I right in thinking we don't actually encrypt anything for the current targets?
18:10:14Bagderlinuxstb: no, we just need to make it do crc32 and a dummy dsa
18:10:28adiamascould someone explain what a-b shuffle is?
18:10:40Bagderlinuxstb: should be easy enough
18:11:00*Bagder runs off, back in 4-5 hours
18:11:05linuxstb_Could I do it without looking at mi4code.c ? i.e. just from the info on your page?
18:11:15markunbarrywardell: could you commit it with your changes?
18:11:34barrywardelllinuxstb: we just need a mi4 header as described on the mi4 page
18:11:36markunOr is it not ready?
18:11:37linuxstb_Bagder: OK, see you later...
18:11:48merbananlinuxstb_: no the info is not that detailed
18:12:11 Nick tuwe is now known as the_fuchs|weg (n=the_fuch@
18:12:14merbananbarrywardell: some checks are not described on the webpage
18:12:19 Nick the_fuchs|weg is now known as tuwe|away (n=the_fuch@
18:12:22adiamashmmm my tag cache debug says initialize, but shows progress as -1%
18:12:27barrywardelllinuxstb: you could probably do it without looking at mi4code.c
18:12:56barrywardellmerbanan: which webpage?
18:13:05barrywardellmarkun: looks good to me anyway
18:14:24merbananbarrywardell: is there another one ?
18:15:05barrywardellthat's the updated wmcodec patch.
18:15:20barrywardellmarkun: has it been tested on all other wmcodec targets?
18:15:34barrywardellmerbanan: which checks are not described?
18:16:05linuxstb_What's the data that gets written to around 0xe0 to 0xe9 of the actual data itself (not the header)?
18:16:31markunmarkun: I'll fix it for the other targets. Can you give me your changes?
18:16:42markunah, sorry :)
18:18:18 Join BHSPitLappy [0] (
18:18:33 Join BHSPitMonkey [0] (
18:18:33merbananbarrywardell: at the end of the firmware a "magic" value has to be present
18:18:55 Join goldenwords [0] (
18:19:07 Part goldenwords
18:19:21linuxstb_merbanan: The 0xaa55aa55 ? That's mentioned on the page.
18:19:43Crackerizerhow to add a new .lang file and use it with rockbox?
18:20:44 Join spiorf [0] (
18:21:20 Quit _FireFly_ ("Leaving")
18:21:22barrywardellyes,you just append that 0xaa55aa to the firmware binary then pad with 0 to a multiple of 0x400
18:21:28bluebrotherCrackerizer, see
18:22:02markunbarrywardell: wmcodec-pp.c should only be included on portalplayer targets (the Gigabeat also uses drivers/wm8975.c)
18:22:32merbananlinuxstb_: missed it, then it should be complete
18:23:21barrywardellmarkun: ah yes. there was no SOURCES patch in the patch you sent me, so I just made one that I thought would work
18:23:32barrywardelldidn't think of non-pp targets
18:23:55merbananlinuxstb_: you only need to calculate the checksum for now
18:24:15BigMacanyone got a second
18:24:16markunbarrywardell: after the move to target tree it will be easyer I think
18:24:25BigMactrying to finish my wps before school
18:24:27barrywardelllinuxstb_: you can look at an mi4 file created by
18:24:36Crackerizerbluebrother: thank you.
18:24:37barrywardellit's not encrypted
18:24:49markunlinuxstb_: did you look at the patch to move the ipod (4g only so far) to target tree?
18:25:28 Quit Genre9mp3 ("I don't suffer from Rockbox psychosis. I enjoy every minute of it.")
18:25:54merbananlinuxstb_: I mean for the h10 you only need the checksum
18:26:18BigMachow could i make an arrow be the progress bar?
18:26:49BigMacor does it have to fill up? can it be one item that will go along the whole line?
18:28:21BigMacfor instance could i use an item like this ^ and it would just go down the bar
18:28:30BigMacwith no progress bar?
18:28:42BigMacbecause all it does is keep repeating that
18:28:54barrywardellfor all mi4 targets so far you only need the checksum
18:29:17barrywardellyou can use a dummy dsa key for 010301 versions
18:30:36 Quit webguest79 ("CGI:IRC 0.5.7 (2005/06/19)")
18:30:36barrywardellmake the plaintext length extend to the end of the file
18:32:10barrywardellso the plaintext length is the filelength - 0x200
18:33:31linuxstb_barrywardell, merbanan: OK, I'll try and give it a go this evening. And yes, examining the output of and comparing to the input file was my plan.
18:33:33 Join CriamosAndy [0] (
18:34:02linuxstb_Is the H10 a "0102" firmware?
18:34:11merbananlinuxstb_: yes
18:34:16barrywardellthe 5GB model is
18:34:21barrywardell20GB model is 0103
18:34:40*barrywardell notices the new H10 manual
18:34:44barrywardellnice :)
18:36:01Crackerizerbluebrother: another question, how to add the translated lang file to rockbox? I built rockbox but there is no language i added in the language menu.
18:37:49tucozbarrywardell, do you need two manuals for the H10's? That is, do they differ in terms of features?
18:38:11tucozrecording, radio, resolution etc
18:40:39Crackerizersolved, sorry for stupid question.
18:42:46barrywardelltucoz: the only difference I know of is the screen resolution and the hdd size
18:42:56barrywardelland the physical size of the device
18:43:23Arathisand that the H10 5/6GB Pure doesn't have radio..
18:43:48barrywardellyes, that too. Although no H10s have radio in Rockbox yet...
18:44:03 Quit Caliban (Read error: 110 (Connection timed out))
18:46:32barrywardelltucoz: thanks for doing the manual btw.
18:46:37 Quit pondlife ("byebye")
18:50:00 Quit Criamos (Read error: 110 (Connection timed out))
18:52:13bluebrotherCrackerizer, np.
18:54:03 Join niskel [0] (
18:56:10 Quit dan_a (Read error: 110 (Connection timed out))
18:59:03tucozbarrywardell, no problem. Ok. so screen resolution forces two manuals then.
18:59:22tucozand radio :)
19:04:23BigMacwoh woh woh
19:04:24 Join webguest70 [0] (i=c3ec4b1b@gateway/web/cgi-irc/
19:04:38BigMacwtf is up with julius not having an atist badge
19:04:42 Quit daurn|laptop (Read error: 60 (Operation timed out))
19:04:44BigMacsomeone should fix that
19:05:02preglowOH NOES
19:05:08preglowthe end of the world is upon us
19:06:13preglowi guess he should msg llorean, then
19:07:53preglowdoes anyone else have the bloody problem where linux suddenly stops to register usb mass storage devices?
19:08:36preglowit doesn't exactly help on the coding
19:10:09 Quit Crackerizer ("using sirc version 2.211+KSIRC/1.3.12")
19:11:44 Quit dishnecto (Read error: 110 (Connection timed out))
19:13:27 Join Rob2222_ [0] (
19:14:53 Join lightyear [0] (
19:19:31 Quit barrywardell ("CGI:IRC")
19:24:42preglowmusepack never boosts for ~160 kbps now
19:24:44preglowby a very nice margin
19:24:58preglow~180 even
19:25:14BigMachow do i lower the volumes in the ipod video archos simulator?
19:25:29preglowor 209...
19:27:29*preglow summons snowgoon
19:28:40preglowBigMac: ipod video archos? what sim is that?
19:29:55 Quit niskel (Remote closed the connection)
19:30:17 Join Nico_P [0] (
19:31:33 Quit Rob2222 (Read error: 110 (Connection timed out))
19:33:15 Quit BHSPitLappy (Read error: 104 (Connection reset by peer))
19:39:24amiconnmirak: around?
19:41:13markunpreglow: how it musepac for lower bitrates?
19:41:23BigMacit is the ipod video sim
19:41:28preglowBigMac: use the wheel keys musepack..
19:41:32BigMacbut it says archos in side it
19:41:36preglowmarkun: well, they say it's not optimal for lower bitrates
19:41:38amiconnVolume doesn't work in the sim
19:41:40preglowmarkun: but it's not too bad, really
19:41:51BigMacthats going to suck then
19:41:53preglowmarkun: but if lower bitrates is what you want, something better than musepack exists
19:41:56amiconnThe only option is the volume adjustment in the OS
19:42:14markunpreglow: I use Ogg Vorbis q3
19:42:21BigMacno i have to move the location of my zero volume
19:42:22amiconnpreglow: archos sims don't play music
19:42:29BigMacits different then the others
19:42:31preglowamiconn: no?
19:42:36preglowamiconn: ah, right, hwcodec
19:42:40amiconnWe would need to simulate the MAS
19:42:57amiconnIn fact that should be possible, now that libmad is part of the code
19:43:23preglownot worth the effort, imho
19:43:27amiconnThere is an ages-old patch that added playback using libmad
19:43:35preglowis there ,now
19:45:14 Quit Arathis (Read error: 110 (Connection timed out))
19:46:11 Join Arathis [0] (
19:46:14BigMac%pc/%pt %s%m|134|224|%ac%?It<%It|>
19:46:27BigMacwhat is wrong with that line
19:46:50BigMacpc/pt shows time passed/total time
19:47:05BigMac%sm should set the scroll margins
19:47:14BigMacmother fooker
19:47:23BigMaci figuered out my own problem
19:47:27***Saving seen data "./dancer.seen"
19:47:34BigMacthere is no scroll margins patch in the sim
19:48:10BigMacbut is that the right code for scrolling margins?
19:48:47amiconnHmm, why don't apps running under wine get a button in the gnome taskbar?
19:49:18billytwowillybecause this isn't the gnome support/wine support channel.... and god is punishing us.
19:50:04billytwowillythe long anser is the install script adds some stuff to some place that tells gnome to point to the program and wine isn't doing that.
19:50:25amiconnI don't mean the icons to start the app
19:50:42 Join gopp [0] (
19:50:58linuxstb_I think it's an option in .winerc (or whatever it's called) - to allow the X window manager to manage Wine windows.
19:51:02billytwowillyif they are running they should get a spot on the taskbar.. they do in kde. I'm not that familiar with gnome.
19:51:37goppone question how do I know if thier a ipod of rockbox for my ipod mini
19:52:17*billytwowilly shouts "!"
19:52:31*billytwowilly points to the manuals linked off the front page.
19:52:41linuxstb_gopp: That question doesn't make any sense... But there is a version of Rockbox for the ipod mini, if that's your question.
19:53:36goppOpps my english sucks
19:54:14goppMy question is, I have currently rockbox on my ipodmini, and it works great, but I was wondering how do I know if thier a update firmware of rock for my mini.
19:55:26linuxstb_Updated versions of Rockbox are released all the time (many each day). The recent changes are listed on - so upgrade whenever you want to.
19:55:33*linuxstb_ goes home
19:55:36 Quit linuxstb_ ("CGI:IRC")
19:56:01billytwowillyrockbox doesn't support video on the models that can play video right?
19:57:11FebsRockbox doesn't really officially support video at all yet.
19:58:30billytwowillythat's what I thought. thanks.
19:58:40goppall I wanted to realy know
19:58:52goppif any new verison for the ipod mini, support for less battery drain
19:58:56goppthat is battery last longer
19:58:59billytwowillyDo any of the supported models have an fm radio? if so, does rockbox support it?
19:59:15goppsome model do, and yes it supports it
19:59:33FebsYou can get information about the development of video support here:
19:59:59goppbillytwowilly Archos
20:00:00goppRecorder, FM Recorder, Player, and Ondio SP/FM
20:00:30FebsInformation of supported features on various platforms:
20:01:52markunFebs: do you know if the ipod does realtime mp3 encoding now?
20:02:12billytwowillysays no on that page.
20:02:35markunbillytwowilly: it's a bit out of date :)
20:03:06billytwowillythat's too bad.
20:03:21*billytwowilly is waiting anxiously for a rockbox port to the sansa e200 series players
20:03:44billytwowillyif it'll do radio and has bookmarks it's awesome.
20:05:48 Quit Nico_P ()
20:10:00 Quit adiamas ("Chatzilla 0.9.75 [Firefox]")
20:10:12 Join stripwax [0] (
20:10:54blxmy iAudio X5L just arrived :):):)
20:11:21Siimorslucky you :(
20:11:30SiimorsI wish I had one of those
20:11:37Siimorsbut I couldn't get it here
20:12:15billytwowillyWhere are you?
20:12:32billytwowillysend me money and I'll mail you one as long as it isn't illegal.
20:13:09Bagder995 people on the rockbox map now
20:13:27Siimorsbillytwowilly: too late
20:13:34SiimorsI already bought an ipod :(
20:13:53billytwowillyno worries. it'd just be a slight PITA for me anyway:)
20:14:25billytwowillylink to the map? I'll add myself
20:14:32Bagdersee topic
20:14:37SiimorsI like the hard drive though, 60gb is enough for me
20:14:53billytwowillywell, that's what I get for idling in the channel for a bajillion years;)
20:14:55blxL is 30G
20:15:18Siimorsfits my entire Keith jarrett discography @ lossless and then some more (like 25GB more)
20:15:51*billytwowilly would go crazy with just one guys music on his player
20:16:25Siimors j
20:16:34Siimorshes great
20:16:39Siimorsand he has like 100 albums
20:16:43billytwowillyI don't know who he is and it doesn't matter.
20:16:46Siimorsso I never get bored
20:16:51billytwowillyI need variety.
20:16:53Siimorsalso I learn while I listen
20:17:19Siimorsand it require a lot of listening
20:17:27billytwowillywoot! there's a developer in my home town!
20:17:38Siimorsto really appreciate a man like him and also learn something from him
20:18:26billytwowillybilstovmhz is from my home town!
20:19:00Siimorsdo u like keith jarrett?
20:19:17billytwowillyI don't know who that is.
20:19:20billytwowillyit doesn't matter.
20:21:41Siimorsis there at least a slight chance that someday I will be able to record with my ipod through phone jack?
20:21:47 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
20:22:23 Quit blx ("quit")
20:22:30preglowrecord through the phone jack???
20:23:15belzehow could that technically be possible? ;)
20:23:23 Quit linuxstb (Read error: 110 (Connection timed out))
20:25:47goppthey use ipod mic
20:26:13 Part stripwax
20:27:12 Quit lorijho_ ("leaving")
20:28:40 Join barrywardell [0] (
20:29:35zeif it has an audio input, then it'd be possible to make (or buy?) an adapter to go from where the handset of a phone would plug in... i dunno about a jack itself, maybe...
20:47:35 Nick linuxstb_ is now known as linuxstb (n=linuxstb@rockbox/developer/linuxstb)
20:48:17linuxstbSiimors: Which ipod do you have?
20:51:27Siimorslinuxstb: the most recent one
20:51:29Siimorswith video
20:52:36linuxstbThen no, I don't think the video supports recording via the headphone socket.
20:53:44linuxstbIt's definitely supported on the Photo/Color - I've done it. I've also managed to record via the line-in connector on my video - which is hidden within the dock connector.
20:54:13preglowlinuxstb: what the flaming hell's up with being able to do that?
20:54:27Siimorslinuxstb: so if I buy the dock, I can record then?
20:55:11linuxstbpreglow: Ask Mr Apple. I think it's selected via a gpio pin, I forget the details. The right-channel can be made an input.
20:55:32Siimorsand what is the best quality one can achieve?
20:55:39linuxstbSiimors: No, you don't need a dock. I'm just talking about the pins you can see in the connector at the bottom of your ipod - you need to make a cable.
20:55:48Siimorscan it handle 16bit/44khz uncompressed?
20:55:56 Join Rondom [0] (
20:56:01linuxstbSiimors: My tests were just 16-bit/44.1KHz stereo, but I'm pretty sure it can go up to 24-bit/96KHz.
20:56:57linuxstbpreglow: So you can even record just using the standard ipod white earphones as a poor quality mic.
20:57:37SiimorsI was thinking about hooking up it to my stage piano
20:57:46SiimorsI would like a easy way to record my ideas
20:58:30preglowlinuxstb: a very poor mic...
21:03:58 Join Massa [0] (n=Massa@
21:04:19ghode|afkSlasheri: is your latest tagcache comit related to this problem:
21:04:48bluebrothercan someone explain me a strange behaviour with the struct user_settings?
21:05:34bluebrotherI added a new char array to the end of it. When writing the value it works, but when trying to access it via a pointer it is off by 8 on the h120 sim
21:06:04bluebrotherthe strange thing is that it works as expected when I put the array somewhere else in the struct, not at the end.
21:07:01 Join Caliban [0] (
21:11:01MassaCan someone explain me the differences between the several drawmodes (DRMODE_FG, DRMODE_SOLID, DRMODE_BG)? I can't see any difference when I display a bitmap with them...
21:12:46zeMassa: i don't really know, but it sounds to me like something that'd involve how the bitmape relates to other things being drawn at the same time
21:14:54MassaOf course - but how exactly? And what's the difference between DRMODE_FG and DRMODE_SOLID?
21:15:33zewell again, just a wild guess, but that might have to do with transparency? FG allowing it, SOLID not? still, i don't know anything about it :p
21:16:00linuxstbMassa: I think the draw modes only affect mono bitmaps. The GraphicsAPI wiki page may help.
21:16:23MassaIf you use a "%x" tag, the bitmap will be displayed with DRMODE_FG; if you use "%xl" and later "%xd" DRMODE_SOLID will be used...???
21:17:48MassaI just wonder which is the best (or the correct one) to display albumart bitmaps... ;)
21:17:50mirakhum I did a PROFILE build, shouldn't there be a file at the root with the stats ?
21:18:31Learmirak: and you call the start/stop profiling functions?
21:19:24mirakLear: well, the codecs.c have this start/stop already isn't it ?
21:20:07mirakLear: rb->profile_thread(); rb->profstop(); ?
21:20:19Learmirak: yes, those are the ones I meant.
21:20:44mirakLear: they are already there for rockdoom for exemple or codecs.c, so I exepected something would happen
21:21:10mirakLear: I added them to mpegplayer.c and it by the way even without them the device stalls
21:21:22mirakin PROFILE build
21:21:33Learfor codecs, just having them in codecs.c doesn't do anything. You need to add it to the actual codec.
21:22:03Slasherighode|afk: yes, excactly
21:22:04 Join lodesi [0] (
21:22:27 Join CC [0] (
21:22:35 Nick CC is now known as CrisCr0ss (
21:23:02MassaBTW, why does the screen structure contain a "set_drawmode" but not "get_drawmode" callback?
21:24:18CrisCr0ssHow do u make menu fonts bigger?
21:24:29linuxstbMassa: I guess because no-one has needed it...
21:24:30CrisCr0ssjubda snakk'
21:24:35CrisCr0sskind of small**
21:24:58mirakLear: and about the crash of mpegplayer, any ideas what could be the cause ?
21:25:41Learnope, never tested it myself.
21:26:21 Quit _Veseliq_ (Read error: 104 (Connection reset by peer))
21:26:24Learmassa: afak, drawmode is for pixel-level and text functions, not bitmap drawing.
21:26:40 Join Strogg [0] (n=user@
21:26:54 Join solarflare [0] (
21:27:04Massalinuxstb: Well, actually I wanted to do a get_drawmode and store the value somewhere - then set_drawmode to what I need and set it back after displaying to the value it had...
21:27:30MassaBut if that function does not exist, I do what other functions also do - just set it back to DRMODE_SOLID ;)
21:27:36bluebrotherCrisCr0ss, use a theme with a bigger font.
21:27:58bluebrothere.g. UniCatcher
21:28:15MassaLear: aren't bitmaps made of pixels? :P
21:28:19StroggIs there a more reliable way of finding out which ipod version you have other than looking at the apple pictures? I've tried installing rockbox on my ipod (which I believe is an ipod photo.. same as an ipod color) and it's failed on me. I've tried the 4g image as well with no luck.
21:28:21miraklinuxstb: any idea why profile build crash mplagplayer ?
21:28:54Learmassa: except for mono bitmaps. I checked the 16-bit driver only, but...
21:31:00 Join lorijho [0] (n=lorijho@
21:32:10ghode|afkSlasheri: great, that bug was very annoying!
21:32:23CrisCr0ssknow one knows how to increase font size?
21:32:57 Quit Kingstone ("_/\__/\____/\______/\_________________ Jim, I think we lost him.")
21:33:29FebsCrisCr0ss, use the "Browse fonts" option to choose a bigger font.
21:34:03FebsMake sure that you have downloaded the font pack from the daily builds page.
21:35:03 Join freqmod [0] (
21:36:34linuxstbmirak: Sorry, I've no idea. I don't think mpegplayer does anything unusual.
21:37:23mirakok, or maybe I am not patient enough since profiling seems to slow down things, but it would be a bit to much
21:37:30miraklinuxstb: maybe it's a target problem
21:38:39 Quit mirak (Remote closed the connection)
21:40:41 Join Nico_P [0] (
21:40:50 Quit lodesi (Read error: 110 (Connection timed out))
21:40:53Nico_Pbarrywardell: you there ?
21:41:04Stroggthe video image didn't work either..
21:41:21Nico_Pi just uploaded rockbox on my brother's h10
21:41:32tucozis the rockbox site slow for anyone else?
21:41:56Nico_Pbut when we switch it on after putting the h10.mi4 bootloader on it, it shows "system files missing"
21:42:00 Join lodesi [0] (
21:42:20pixelmatucoz: for me too
21:42:28barrywardellwhich h10 model do you have?
21:42:32Nico_P5 GB
21:42:53tucozpixelma, ok.
21:43:02linuxstbI've just noticed that Bagder's change to has broken the firmware creation for my Elio - so I'm guessing it's also broken for the h10_5gb (that was the option I was using)
21:43:13barrywardelldid you follow the instructions in the wiki?
21:43:24Nico_Pis it normal the bootloader i compiled and coded is 54 KB when the iriver one is 3.8 MB ?
21:43:33Nico_Pi'm pretty sure i did everything right
21:43:54CrisCr0ssI hve a problem
21:44:04CrisCr0ssI removed RockBox but now want to install it
21:44:10linuxstbNico_P: Did you use the script, or did you run mi4code.c manually?
21:44:11CrisCr0sshowever i cannot access ipod in Diskmode again
21:44:20Nico_Pi did follow the instructions closely and i also have built rockbox regularly since it is available on the h300 so i have some experience
21:44:29Nico_Pi used the script
21:44:30CrisCr0ssto extract the ipod .rockbox file again
21:44:43FebsDoes anyone know what type of tags ALAC uses?
21:44:51linuxstbNico_P: I think the script is broken - could you test using the previous version?
21:45:01barrywardellNico_P: yes, my one is the same
21:45:20linuxstbFebs: They're ".mp4" files - so the native tagging format in that container. I'm not sure of the name.
21:45:37barrywardelllinuxstb: what's broken with the script?
21:45:38Nico_Pshould i use ver 1.7 of ?
21:45:52linuxstbbarrywardell: It worked before Bagder's commit, it doesn't any more...
21:45:54StroggAnyone around available for ipod help? when I got through the wiki instructions to install the bootloader and then boot it, I get the apple error screen
21:46:02FebsThere's another thread on Head-Fi about Tag Cache not properly reading ALAC tags, or displaying different tags from what iTunes displays. Link in a moment.
21:46:19linuxstbbarrywardell: It seems my Elio needs the .mi4 to be encrypted.
21:46:31tucoz1011 rockbox users on rashers page :)
21:46:48barrywardelllinuxstb: that's a pity
21:46:52linuxstbbarrywardell: Which I think is what the script did for the h10_5gb option before today's change.
21:47:29***Saving seen data "./dancer.seen"
21:47:29linuxstbNico_P: Yes, that was the version working for me.
21:47:46Nico_Plinuxstb: ok, thanks
21:47:49Nico_Ptrying now
21:48:00*barrywardell is looking at bagder's changes
21:48:30 Join CC [0] (
21:48:36MassaHi Nico_P!
21:48:41linuxstbFebs: The m4a metadata parser is in need of fixing. There's a patch on the patch tracker for it (I think by Lear), but I've not got round to looking at it.
21:48:55barrywardelllinuxstb: the h10_5gb didn't encrypt the firmware before Bagder's commit
21:49:07barrywardellit did before my last commit
21:49:17linuxstbAh, so you broke it? :)
21:49:21tucozmaybe the rockbox policy wrt anonymous contributions could be changed to not include tools :)
21:49:34barrywardellwell it was still working after my commit :)
21:49:43Nico_PMassa: hi
21:49:44barrywardell(I think)
21:50:00MassaNico_P: Do you know what the different drawmodes are for? What is the correct one to use for albumart bitmaps? I don't see any difference...
21:50:15Nico_PMassa: no, i have no idea
21:50:18*amiconn missed mirak
21:50:36CCcan anyone help with a reinstall issue i am having..
21:50:43Nico_Pbarrywardell: so which vesion should i use ?
21:50:45CCIpod no longer recognized as Disk in disk mode
21:50:55linuxstbNico_P: I've just checked, and I'm currently using 1.5
21:50:57Nico_Por maybe run mi4code
21:51:04barrywardellNico_P: try the version just before Bagder's commit
21:51:20Nico_Pbarrywardell: i just tried ver 1.7
21:51:24Nico_Pgot the same thing
21:51:56MassaI just checked with a h120 (simulator) build - it does also not matter which mode I use - it looks always the same????
21:51:57barrywardelloh. maybe it does need to be encrypted then
21:52:01linuxstbbarrywardell: Have you figured out the backlight on the H10?
21:52:17barrywardelllinuxstb: I haven't even looked at it yet
21:52:26linuxstbSo you can read the LCD without it?
21:52:45barrywardellthe backlight is turned on by the iriver booloader and stays on after that
21:52:55linuxstbAh :). The Elio turns it off...
21:53:07Nico_Pver 1.6 geav me this :
21:53:10Nico_Pmi4code v0.9.28-beta (c) by MrH 2006
21:53:10Nico_PEncrypting with key 20d36cc0 10e8c07d c0e7dcaa 107eb080
21:53:11Nico_PCRC32 corrected (00000000 -> dc6ca84e)
21:53:20Nico_Pi didn't have it before
21:53:23linuxstbYes - that's what I get. I think that will work.
21:53:42 Nick CC is now known as Cris_Cr0ss (
21:53:56StroggWoot! It boots! :)
21:54:02Nico_Pyep it works
21:54:04barrywardellmaybe encryption is necessary for 5GB. it isn't for 20GB
21:54:08StroggI think I found a bit of info that needs to be added to the wiki. hehe
21:54:11linuxstbCris_Cr0ss: Are you using Windows, a Mac, Linux?
21:54:22Nico_Pbarrywardell: i confirm that... the 5 GB need encryption
21:54:40barrywardellso maybe 0201 versions need encryption, 0301 ones don't
21:54:52barrywardellNico_P: I'll fix now
21:54:56merbananCRC32 corrected (00000000 -> dc6ca84e)
21:54:57 Join mirak [0] (
21:55:06 Quit mirak (Read error: 104 (Connection reset by peer))
21:55:11merbananisn't it just that the crc was wrongly calculated ?
21:55:19Cris_Cr0ssI had it wrkng fine yesterday, but i wanted to remove it because I knew it would take some time to learn and read
21:55:26Nico_Pis it supposed to be so long to shut down ?
21:55:30 Quit MarcoPolo (Read error: 104 (Connection reset by peer))
21:55:32Cris_Cr0ssand i needed it today so i decided i would give it another shot later
21:55:33barrywardellbtw. why is cvs so slow
21:55:52Nico_Pbarrywardell: the whole website is slow
21:55:55barrywardellNico_P: haha. it never shuts down. we haven't added that yet :)
21:55:56preglowfreqmod: yo
21:56:16Nico_Pbarrywardell: :)
21:56:35 Join _FireFly_ [0] (
21:56:37Cris_Cr0ss@linux any idea why it wont recognize as HardDisk mode anymore?
21:56:40freqmodhello, did you figure out anything about Matching constant doesn't allow a register?
21:56:43linuxstbCris_Cr0ss: Apart from suggesting you reboot (if you haven't done so already), I've got no ideas. But I recall people posting similar problems on the forums - you may find the solution there.
21:57:05barrywardellmerbanan: you might be right there
21:57:19Cris_Cr0ssby reboot u mean the ipod or my PC?
21:57:20*freqmod does not like the update of id3.c & playback.c, it breaks the speex patch
21:57:26linuxstbCris_Cr0ss: The PC.
21:58:23 Quit [San] (Read error: 104 (Connection reset by peer))
21:58:40 Join [San] [0] (
21:59:04preglowfreqmod: i really haven't had time to look at code
21:59:27preglowso if you want it commited any time soon, you'll have to fix it yourself :/
22:00:24Cris_Cr0ssalright i will try it i guess
22:00:24freqmodI might look at it when I don't have 100 other things I want to do, or want a break from those things.
22:02:09preglowgot quite a list building up myself
22:02:22 Quit [San] (Client Quit)
22:02:32barrywardellcrc only gets corrected when you encrypt
22:02:35 Join [San] [0] (
22:02:38preglowcan't risk doing rockbox stuff now, too distracting, heh
22:02:58barrywardellthat explains the problem with since it no longer encrypts
22:03:22 Quit CrisCr0ss (Read error: 110 (Connection timed out))
22:03:24 Join _hotwire_ [0] (
22:04:19 Quit Cris_Cr0ss ()
22:04:28 Quit solarflare ("Leaving")
22:04:46freqmodhehe... no, I have a "særemne", the main Norwegian project where I am going to make a ~5min animation of 2 poems in Blender before 2006.10.01 (Y.M.D) (will be licensed Creative Commons NC-ND-SA)
22:05:11barrywardelland the crc32 doesn't get checked for 010301 versions which explains why it worked for me
22:07:43Stroggdammit. I got it to work once and now I can't get it again
22:08:40preglowfreqmod: yeah, i remember doing the særemne some tons of years ago, heh
22:09:42linuxstbStrogg: Have you worked out which ipod you have yet?
22:10:02barrywardellNico_P: can you try this version of
22:10:09freqmodSpeex, with updated rockbox: *** glibc detected *** corrupted double-linked list: 0x00002b5f5513f8b8 *** :( :(
22:10:27Strogglinuxstb: I'm nearly certain it's a ipod color.. it's a 30gig photo version.
22:10:54Nico_Pbarrywardell: i'll try that
22:11:02StroggI think part of the problem is that windows is stupid. I have a feeling I may have better luck trying on my linux box at home.
22:11:27barrywardellthanks. i'll commit it if it works for you.
22:11:53barrywardelllinuxstb: can you try it on your elio?
22:11:55 Join lodesi_ [0] (
22:11:56linuxstbbarrywardell: It works for me on my Elio. Could you add an Elio option as well?
22:12:18barrywardellyeah, sure. same settings as h10_5gb, right?
22:12:28 Join mirak [0] (
22:13:08markunmirak: amiconn's looking for you
22:13:53linuxstbStrogg: If it's colour, and if it's not a Nano, then the easy way to identify is if the headphone socket is in the middle of the top (a Photo/Color) or on the far-right side of the top (a Video).
22:14:17Strogglinuxstb: Ooh. It's far right.
22:14:19mirakamiconn: I am there
22:14:28*Strogg tries once more with the video bootloader
22:14:32linuxstbStrogg: And the hold switch on the far left?
22:14:40Strogglinuxstb: Yeah. that's the one.
22:14:53linuxstbThen it's a video. You need to do extra steps when building the bootloader.
22:15:14Stroggwell damn. all this time I had a video..
22:15:23Stroggitunes has been holding back functionality on me.
22:15:31amiconnmirak: I wanted to give some hints regarding your asm endeavour
22:15:52amiconn(1) When dealing with an int array, move.w is the wrong instruction
22:16:04amiconnints are 32bit on coldfire, so you need move.w
22:16:12amiconnErm, move.l
22:16:16linuxstbbarrywardell: Yes, I've been using the h10_5gb switch successfully.
22:17:04mirakamiconn: the block array is an int16_t array
22:17:19amiconn(2) when move.l (%a0),%d0 accesses array[0], accessing array[1] needs move.l (4,%a0),%d0
22:17:29amiconn...i.e. the index is not scaled to data size
22:17:32 Quit _FireFly_ ("Leaving")
22:17:35mirakamiconn: even if it's int16_t ?
22:17:44amiconnOkay, for int16_t (aka short) move.w is correct
22:17:58StroggHah! that did it.
22:18:04amiconnBut then array[1] is move.w (2,%a0),%d0
22:18:18mirakamiconn: ok it's always in bytes ?
22:18:29mirakamiconn: the displacement value
22:18:32preglowoffsets are always in bytes
22:18:37preglowassembler always works with bytes
22:19:01Strogglinuxstb: Rocking.. the bootloader fires up now. :)
22:19:05Nico_Pbarrywardell: your script works
22:19:10mirakok I change that and give feedback
22:19:15amiconnIf you have an index register, you can select scaling. Literal displacements are always unscaled, i.e. in bytes, on coldfire
22:19:24amiconnpreglow: On coldfire, yes
22:19:32barrywardellNico_P, linuxstb: great, i'll commit the changes now
22:19:34amiconnSH is different
22:20:38 Quit [San] (Read error: 110 (Connection timed out))
22:21:10mirakamiconn: what exactly is the scale factor ?
22:21:23mirakamiconn: SF parmater on MAC instructions
22:22:02preglowamiconn: and most other assembly languages as well
22:22:37preglowmirak: scale factor is (%d1*4, %a0) for example
22:22:59preglowmirak: but don't do that unless strictly necessary, it's the slowest addressing mode there is
22:23:06amiconnEMAC scale factor...
22:23:21preglowemac has no scale factor
22:23:23preglowjust an optional shift
22:23:25preglowin integer mode
22:23:47preglowusually just used to offset fixed point formats
22:23:52amiconnThey call it scale factor Ry.{U,L},Rx.{U,L}SF,ACCx
22:23:57mirakpreglow: the shift is just one byte left or right
22:24:02preglowmirak: bit
22:24:07miraka bit
22:24:18mirakam not sure what it can be used for
22:24:39Nico_Pbarrywardell: are you aware the display is shifted on the h10 5 GB ?
22:25:03barrywardellyes, there seem to be several display glitches on both model lcds
22:25:15barrywardelldisabling the statusbar fixes them
22:25:17Nico_Pi get problems in the file browser
22:25:34barrywardelli'm trying to figure that one out at the moment
22:25:46 Quit lodesi (Read error: 110 (Connection timed out))
22:26:11Nico_Pit probably has to do with margins not getting set correctly
22:26:24Nico_Por maybe what i'm saying is obvious
22:26:38 Join [San] [0] (
22:26:41linuxstbbarrywardell: Am I right in thinking your latest mkmi4 doesn't encrypt, but calculates the CRC?
22:26:56barrywardellno, any suggestions are welcome. I'm still not familiar with a lot of rockbox
22:27:13barrywardellyes, that's what it is supposed to do anyway
22:27:20linuxstbBut the earlier versions (1.5, 1.6) did encrypt?
22:27:35Bagderlinuxstb: -pall makes the whole file "plain text"
22:27:55Bagderlinuxstb: the mi4 header has a length for amount of unencrypted data
22:28:15linuxstbOK. I've made a start on adding .mi4 support to scramble based on the info on Bagders web page (and looking at the output of I'll continue with that now that is working (I was creating non-working images...)
22:28:19Bagderh10 and sansa both work fine with the whole file marked as plain
22:29:10linuxstbBagder: I'm assuming that if I implement basic mi4 support in scramble without looking at mi4code.c, then we're all happy?
22:29:31preglowwhere'd mi4code.c come from?
22:29:35Bagderpreglow: MrH
22:29:54sharpegood day everyone.
22:30:08Bagderbarrywardell: checking the 6gb patch?
22:31:00 Join _Veseliq_ [0] (n=veseliq@
22:32:09barrywardellBagder: i think the new version of will work fine without the patch since we now use plaintext
22:32:45barrywardelllorijho has a 6gb, so maybe he can test it
22:32:47Bagderah, yes
22:35:16linuxstbBagder: BTW, is there a problem with - it seems slow this evening.
22:36:03BagderI don't think there's any other error then the limited bandwidth
22:36:18mirakamiconn: I did the changing but the video still displays bad.
22:36:30*Strogg faints over black background green foreground ipod display :)
22:36:50mirakamiconn: it's so bad that it can't just be some computation problems, I suspect some endianess problems
22:38:10amiconnpreglow: In fact you're right, asm seems to always use byte displacements
22:38:38mirakhum I forgot to modify the moves from registers to ram ...
22:39:03linuxstbBagder: Are the values at 0xe0 and 0xe4 documented? They seem to always be 0x00000100 and 0x000000ec respectively (Little-endian) for my Elio - in both the original firmware .mi4, and every time I've run
22:39:15amiconnI confused two things; SH displacements can only be multiples of the data width because offsets are scaled within the instruction encoding
22:39:16linuxstb(0xe0 and 0xe4 in the data portion of the file)
22:39:35 Quit uski (Read error: 110 (Connection timed out))
22:39:54merbananlinuxstb: do you mean the reset vector ?
22:39:57 Join rotator [0] (n=e@rockbox/developer/rotator)
22:39:59StroggDoes anyone have the url for the fonts page? The wiki references it by name, but not by url.
22:40:13amiconnMakes sense on a platform that throws an address error when trying unaligned accesses.
22:40:21linuxstbmerbanan: No, the values that mi4code.c puts in the actual firmware image itself at those offsets (not the header)
22:40:54StroggAha. Thanks
22:41:15merbananlinuxstb: do you want me to look in the source ?
22:41:23Bagderlinuxstb: it writes three 32 bit numbers
22:41:29Bagderthe two first are "magic"
22:41:55Bagderthe third is the datalength
22:42:13linuxstbBagder: That's what I thought.
22:43:41mirakok now it's better in the sens that I can reconise the shapes but colors are totally bad and saturated
22:45:09mirakmaybe signed or unsigned instructions problems
22:45:37preglowamiconn: even the archs that enfore dword alignment do
22:45:41preglowamiconn: at least the ones i've seen
22:45:46preglowwhich is a decent amount these days
22:46:25 Quit TeaSea (Remote closed the connection)
22:47:20 Join TeaSea [0] (
22:47:40 Join dan_a [0] (n=dan-mirc@
22:50:28mirakmove.l #0x0B190A74,%a0 a0l = 0A74 and a0u = 0B19 ?
22:52:42barrywardellBagder: MrH is on the right track. GPIOL definitely affects sound
22:52:54preglowmirak: yes
22:52:58Bagderbarrywardell: goodie!
22:53:22*linuxstb suspects MrH is sitting there with the source code to all these firmwares and is just teasing us...
22:53:33 Join Xerion [0] (
22:53:43preglowlinuxstb: ssssshhhh1h1h11!!
22:53:48preglowlinuxstb: let him :)
22:54:04BagderPortalPlayer bought all this code from him
22:54:12preglowperhaps he's just an old cracker
22:54:20preglowsome of those people are batshit _insane_ on reading disassemblies
22:54:31barrywardellenabling it for output and writing 1 causes the headphones to have noise - kinda like when you turn them up too loud
22:54:36barrywardellbut music doesn't work
22:54:49barrywardellso we're near, but not quite there yet
22:54:58mirak"move.w (%[block]),%d4\n\t" is there a possibility that the word at block[0] is swapped ?
22:55:05mirakbecause of endianess
22:55:35mirakis that a possiblity ?
22:57:11mirakhum in fact shouldn't I do a MVS instead of a MOVE ?
22:57:18mirakto preserve sign extension
22:58:09miraki guess a simple move doesn't fill the upper word of the register
22:58:34LearHrm, Cygwin is embarrasingly slow with the Rockbox makefiles. An "empty" make takes 50+ seconds here... Is that normal?
23:00:03LearI mean, on Linux, it takes a few seconds, afaicr...
23:00:50Bagdereven vmware on the same machine you run cygwin on is magnitudes faster
23:01:56LearI guess it doesn't spend ages just waiting (cpu load during build isn't very high)...
23:03:06amiconnmirak: mvs isn't supported on coldfire v2 (which our coldfires are)
23:03:09 Quit JoeBorn (Nick collision from services.)
23:03:15 Nick JB_away is now known as JoeBorn (
23:03:15*dan_a tries "recovery mode" on his Sansa
23:03:20mirakamiconn: gcc just told me
23:03:28amiconnIf you need to sign extend 16->32bit, you need a sequence
23:03:29mirakman that's horrible
23:03:39mirakamiconn: a what ?
23:03:41Bagderdan_a: a useful thing ;-)
23:03:48amiconnmove.w (%[block]), %d4
23:03:52amiconnext.l %d4
23:04:11amiconnthere are 3 ext* instructions
23:04:23amiconnext.w sign-extends 8->16bit
23:04:35amiconnext.l sign extends 16->32bit
23:04:41amiconnextb.l sign extends 8->32bit
23:04:58barrywardellwahey. proper sound on h10!
23:04:59dan_aBagder: Nothing changed on screen, no reset with my first test - I'm moving to just doing a device reset (after finishing recovery)
23:05:17lorijhobarrywardell: way to go !!! cool
23:05:23preglowdan_a: so you've got a sansa now?
23:05:32Bagderbarrywardell: yay!
23:05:46dan_apreglow: As well as a 3G iPod.
23:05:49preglowhow nice
23:05:54mirakamiconn: I think it was the last missing piece of the puzzle
23:06:07dan_abarrywardell: Well done!
23:06:17*barrywardell is very happy
23:06:20preglowbarrywardell: why wasn't it proper before?
23:07:04barrywardellthe volume was very low
23:07:07preglowah, right
23:07:09barrywardellalmost inaudible
23:07:09preglowwell, great
23:07:37barrywardellI had to enable a gpio port
23:09:07mirakamiconn: yes !!!
23:09:08 Join MarcoPolo [0] (
23:09:11mirakit works
23:10:33preglowwe've really got a bloody great bunch of new targets since we started anything non-archos at all
23:10:36 Quit Lear ("Chatzilla 0.9.74 [Firefox]")
23:10:37*lorijho bows before MrH, for he is the WIZARD !
23:12:16barrywardellis markun about? he was talking about a wmcodec patch earlier
23:13:09*Bagder prepares for the configure-restructure commit
23:13:10*lorijho bows before MrH, for he is Muad'Dib :) amazing !
23:13:43 Join scorche [0] (
23:15:14*barrywardell wonders if he should commit the h10 sound fix since it will break markun's patch again
23:15:31 Join itsmeguys [0] (
23:16:19barrywardellor should I commit markun's patch too?
23:16:31itsmeguyshey its the main man barry!
23:16:45 Join Drumr [0] (
23:16:53Bagderbarrywardell: if you deem his patch fine then I'd say commit
23:16:53itsmeguysanyhow im off
23:16:58itsmeguyslaters alll
23:17:01 Quit itsmeguys (Client Quit)
23:18:10markunlinuxstb: why did you buy the Tatung?
23:18:23markunbarrywardell: I'm here
23:18:24 Join Genre9mp3 [0] (n=yngwiejo@
23:18:30markunthink the patch needs some more work..
23:18:44barrywardellok, i'll leave it out for the moment then
23:19:26barrywardellwhat else does it need?
23:19:44barrywardellearlier you thought it was ok to commit, didn't you?
23:21:11markunYes, but I'm not so sure anymore
23:21:29linuxstbmarkun: I saw it in a local toy shop whilst there with my neices, and was tempted... It's not a particularly special player, but the idea of porting to another PP5020 player was interesting.
23:21:51lorijholinuxstb: true
23:22:32MassaNico_P, are you still here?
23:22:43Nico_PMassa: yes
23:23:16MassaNico_P: I just uploaded a new version of the albumart patch - I you have time, please test it :)
23:23:19Nico_Pbarrywardell, markun: what patch is it ?
23:23:31Nico_PMassa: i'll have a look
23:23:45Nico_Pis it with seperate tags, as we discussed ?
23:24:08barrywardellNico_P: it was a patch to unify some of the wmcodec stuff that is common to all pp targets
23:24:10 Quit freqmod (Read error: 104 (Connection reset by peer))
23:24:35scorcheugh....stay offline for a couple days, and the forums welcome me back with enough topics to last me a while...
23:24:48markunNico_P: and to make it possible to have non pp targets use the wolfson code
23:24:50MassaNico_P: yes, the tags have been splitted - and it uses the extended syntax of my v5.0.0; internally it still uses the image buffer for the bitmap...
23:26:58preglowlinuxstb: still got that dab dap?
23:27:08linuxstbpreglow: Yes...
23:27:51MassaDid anybody have a look at my first version of the token based WPS parser (file added to FS #4826)?
23:28:24Nico_PMassa: there's something i don't understand.. in this foruml thread :, someone describes WPS based on a conditional for album art
23:28:47Nico_Pso it was already possible, no ?
23:30:35MassaNico_P: not really - they used a trick (which worked partially); they just used the %?C conditional to enable and disable some text
23:31:03Nico_Pok I get it
23:31:05Massaand they used the case that the %C tag didn't produce an output if no bitmap has been found
23:31:05 Quit apo (Read error: 104 (Connection reset by peer))
23:31:08Nico_Pmuch better now
23:31:18 Quit dan_a ()
23:31:49 Join apo [0] (
23:31:58MassaNow it would be possible to switch an existing bitmap on and off with other conditionals (e.g. switch it on and off with the hold button ;)
23:32:02Nico_Pso you can now have a syntax like %?C<>, am I right ?
23:32:23Nico_Poops... %?C<>
23:32:30Nico_Pstill no good :(
23:32:45MassaNico_P: this was already possible.
23:32:54 Quit Drumr ()
23:33:07Nico_Pyes but i was trying to add a between the <>
23:33:20MassaNico_P: now you can do this: "%?mh<%C|>"
23:33:24 Quit Arathis ("[rl_bot quit]")
23:33:26mirakpreglow: there is ext.l after a move from ram, now to put back to ram is there something to shrink ?
23:33:33Nico_P'% C' without a space
23:33:37 Quit MarcoPolo (Read error: 104 (Connection reset by peer))
23:34:42 Join MarcoPolo [0] (
23:34:46Massathis would display the bitmap if the main devices hold button is switched on and doesn't display it when hold is off
23:34:49amiconnmirak: Just store the lower half; if you overflow the 16bits the result will wrap
23:35:04Massa(not that it makes sense - just as an example ;)
23:35:06preglowlinuxstb: know what chip is in it?
23:35:21amiconnIf wrapping can occur you would check for saturation, which means several extra instructions
23:35:22Nico_PMassa: it seems like a good system
23:35:28mirakamiconn: wrap ?
23:36:27MassaNico_P: I also tried to put everything into functions... hopefully better to read and change :)
23:36:36amiconn...but I doubt that the idct result will overflow 16 bits (depends on the scaling though) if the _final_ result is only 8 bits wide (the pixel values)
23:37:08 Quit _Veseliq_ (Remote closed the connection)
23:37:14linuxstbpreglow: It's a telechips - ARM940T-based
23:37:36mirakamiconn: my other try with abusing of MAC ressemble more to something now
23:38:03mirakI think if I put the matrix coef in register and don't get them from ram I can speed things a lot
23:38:35 Join dan_a [0] (n=dan-mirc@
23:39:31 Join Chris18192 [0] (
23:39:32 Part Chris18192
23:40:09mirakit's green saturated right now ...
23:42:04 Join _Veseliq_ [0] (n=veseliq@
23:42:11 Join uski [0] (
23:44:15preglowlinuxstb: ahh, right, remember now
23:47:30***Saving seen data "./dancer.seen"
23:47:46 Quit Siimors (Read error: 101 (Network is unreachable))
23:48:35 Join blx [0] (
23:48:47blx"Turn the X5 off, remove the USB cable and insert the charger. The X5 will automatically upgrade the flash with the Rockbox bootloader"
23:48:56blxthat step is not working for me
23:49:11blxi'm just getting the usual "Charging" screen
23:49:56Bagderblx: then you probably aren't having it named correctly
23:50:02vrit works for me
23:50:35blx /firmware/x5v_fw.bin
23:50:39MassaGood night everybody - I'm really tired and have to sleep now (S)
23:51:06pixelmatucoz: I'd like to give you the X5, updated H300 and Ondio drawings...
23:51:27tucozsounds great.
23:51:37 Quit Massa ("(A)bort, (R)etry, (I)nfluence with large hammer")
23:53:33bluebrotherhmm, are debug builds known to be broken?
23:53:43Bagderprepare to get confused when you reconfigure ;-)
23:53:47bluebrotheror am I doing something wrong?
23:54:17blxBagder, should i name it something else?
23:54:40Bagderblx: I don't remember, I just know it always worked fine for me and that naming it wrong is a common mistake
23:54:54blxi did not rename it
23:55:15markunpixelma: nice work on the rockbox logo!
23:55:15blxhowever. the documentation says FIRMWARE as directory name, the name on player is firmware
23:55:18Nico_Pbarrywardell: about markun's patch... why not commit your fix and update the patch too (as you know exactly what's changed...)
23:55:24Nico_Punless there are a lot of changes ?
23:55:27pixelmathanks :)
23:55:38 Join lodesi [0] (
23:56:24 Quit Rondom ("Ex-Chat")
23:56:52barrywardellNico_P: I have my changes applied to a clean cvs tree now so I'll commit after a little testing
23:57:18barrywardelland update markun's patch. it's only a small change anyway
23:58:11Nico_Pi'd love to help you with the port, but i suppose what you're doing is mostly reverse engineering ?
23:58:34barrywardellit's MrH doing all the reverse engineering

Previous day | Next day