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

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

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

#rockbox log for 2006-04-23

00:01:33preglowamiconn: btw, does the sample rate on the recording screen reflect the actual rate when using spdif on archos?
00:01:39preglowdoesn't look like it in the code
00:01:56amiconnHmm, dunno. I can check
00:02:06*amiconn doesn't record much)
00:02:19preglowi think it'd be best if it reflected the actual srate
00:03:31amiconnI agree
00:03:47amiconnIt's only lostlogic's box that throws DT_TEXTREL warnings
00:04:07amiconn...and it's the only box with a native gcc < 4
00:06:13 Quit |Unknown| ()
00:06:16sharpeoh noes!
00:07:16 Quit ender` (" There are two major products that come out of Berkeley: LSD and UNIX. We don't believe this to be a coincidence.")
00:07:19amiconnpreglow: Hmm, on archos the sample rate isn't displayed at all on the recording screen...
00:07:33amiconn...but iirc it was, and should be??
00:08:07preglowthe screen shot in the manual does display it
00:08:12preglowthat's what i based myself on
00:08:23amiconnAh, yes.
00:08:46amiconnSomehow the peakmeter now has twice the height as it used to have...
00:08:57preglowhow nice
00:09:19 Join midkay [0] (n=midkay@rockbox/developer/midkay)
00:13:10preglowi think we should rename tagcache to something a bit more friendly
00:13:43preglowit's not only for id3
00:13:47bluebrotherFriendlyTagCache ;-)
00:13:53preglowanother thing we should fix
00:14:03preglowrockbox currently says 'id3' about all tags
00:14:05midkaytag database or tagdb sounds best to me..
00:14:08preglowwhich isn't correct in the least
00:14:10bluebrotherall your base belongs to tag!
00:14:24sharpeall your tag are belong to databse
00:14:34bluebrotherbut pleeease nothing with "id3" in it.
00:14:36preglowtag database is better than tag cache, at least
00:14:44sharpetagbase sounds cool.
00:15:00bluebrotherwe have id3 tags, vorbis comments, ape tags ...
00:15:27sharpeit does... i mean... try saying it.
00:15:36amiconnpreglow: This change (double-height peakmeter in recording screen) was committed on 25 Mar 2006 by 'Dan'. He committed a patch by Martin Scarratt
00:15:51amiconnThe double height has to be taken back for archos.
00:16:03bluebrotherTag Database sounds good. At least it is a database.
00:16:14midkayamiconn, now that you seem to be back - can you tell me what the problem was with the keyboard update?
00:16:32amiconnI even saw that after looking at your diff for 2 minutes
00:16:55amiconnBefore, there was a button_read_w_tmo(HZ/2) , which nicely waits and yields meanwhile
00:17:14amiconnYou replaced it with a button_get(false) which busy polls...
00:17:26midkayi thought button_get(false) was the new/better way of doing it..
00:17:33preglowamiconn: badness
00:17:39midkayi had no trouble with it at the pitch screen.
00:18:12amiconnThen either the pitch screen yields at another place, or you didn't really test
00:18:22 Quit lee-qid_ ("Trillian (")
00:18:29*amiconn is somewhat annoyed by such things
00:18:39midkayamiconn, ah, it's button_get(true).
00:18:45amiconnThe feature freeze phase should stabilise things, not the opposite...
00:18:57midkayamiconn, well, i'm very sorry. :)
00:19:03amiconnbutton_get(true) wouldn't have helped you here, since you want to blink
00:19:13midkayamiconn, exactly, which is why i went for false..
00:19:59midkayamiconn, btw, shouldn't HZ/2 be extremely slow?
00:20:02amiconnWell, all 3 variants (button_get(true), button_get(false) and button_get_w_tmo(timeout) ) have their purpose
00:20:19amiconnIt's not any slower than your blink, even a little faster
00:20:49midkayi always use w_tmo in clock.c, but i see button_get false/true almost _everywhere_ else so i thought that was the smarter way of doing it, and w_tmo was supposed to be used only in certain places.
00:21:02midkayamiconn, i mean HZ/2 as far as button reading..
00:22:10midkaynow i'm remembering discussing this a while ago - HZ/2 seems to scan the buttons fine in Rockbox core but sort of limit code/lcd updates to that speed, but in plugins it slows down the button reading too - iirc that was deemed a bug but never hunted down?
00:22:18amiconnIt will return a pressed button not any slower than button_get(), but it will wait for a button if none is pressed atm, for a maximum of <timeout> ticks
00:22:55midkayand has it ever seemed that's different in plugins?
00:23:16amiconnThere is no bug in these functions I know of.
00:23:18midkayit only seems to return the button *after* <timeout> ticks, regardless of when it's pressed.. or something..
00:23:30midkaymaybe it's my imagination or maybe it was fixed a while back..
00:23:36 Quit Rondom (Read error: 113 (No route to host))
00:23:39amiconnIt was never broken
00:24:19midkayhm.. maybe i'm thinking of something else then. alright, thanks for the clarification :)
00:24:33amiconnbutton_get(false) just checks for a button and returns immediately. It returns BUTTON_NONE if none was pressed
00:24:36sharpewhat else could be a topic on the plague? :D
00:24:59amiconnbutton_get(true) waits indefinitely for a button to be pressed, and therefore will never return BUTTON_NONE
00:25:20midkayamiconn, so why would false kill playback if i may ask? scanning way more often than needed?
00:25:30amiconnbutton_get_w_tmo(timeout) also waits for a button to be pressed, but only <timeout> ticks at maximum
00:25:50amiconnIf still no button was pressed, it returns BUTTON_NONE
00:26:10amiconnWith a button already pressed, all 3 functions return immediately, returning the button value
00:26:21midkayamiconn, gotcha :)
00:26:35amiconnmidkay: There is a loop in keyboard.c that draws the layout, then evaluates the buttons
00:27:03amiconnIf you use button_get(false), this never waits, and hence never yields, so other threads have no chance to run
00:27:25midkayso overall button_get(false) should be avoided?
00:27:38amiconnSo the codec thread and playback thread can't run, and the pcm playback will run out of data
00:28:05amiconnmidkay: Neither. There are situations where it makes perfect sense, otherwise it wouldn't exist
00:28:24amiconnYou should just think about timing and threading, and test stuff
00:28:26midkayamiconn, i'll search the code for what situations, then, i'm quite curious..
00:29:10 Quit Acksaw ("I'm off, see ya later!")
00:29:11amiconnE.g. if you have a loop where you want constant delays, and also want to check buttons
00:29:29midkayamiconn, it wasn't clear to me that false would be a problem, i was under the impression it was 'true'.. unfortunately i didn't have enough time to test every possible situation.
00:29:31amiconnThen you'll use button_get(false) and sleep() in the loop
00:29:51amiconnPlayback is a pretty common situation
00:31:09midkayagreed, i'm not sure what you want me to say :) i could argue that playback was effectively broken for a couple weeks and you're biting my head off for a glitch involving a temporary pause in playback while you're at the not-so-often used keyboard screen.. the result of simply having the wrong impression of true/false in button_get..
00:32:20amiconnI'm only saying that changes should be tested before commit. And imho you cannot compare playback rework with the vkeyboard
00:33:05midkaythat was rather directed at your 'feature freeze should fix things, not break them' comment. not sure if you tested this on your ipod mini, but i found it to be very useful, possibly worth the couple seconds it took to fix. :)
00:33:31amiconnI don't see the advantage of a blinking cursor
00:33:51preglowi prefer it blinking
00:33:51midkayi mean inversing the line bar when in line-edit mode.
00:33:58amiconn..and the inverted line is a bit irritating
00:34:10midkayhaha. is it? seriously? i found it 400% more intuitive..
00:34:26midkaynow scrolling the cursor off the bottom of the keyboard actually makes sense.
00:34:46amiconnIt made sense before as well
00:35:08amiconnNo Ondio user has complained about the way vkeyboard works with a limited keypad
00:35:13midkaysomewhat, but there was absolutely no feedback and it confuses users every time i tell them how to do it until they actually manage to learn it..
00:41:04amiconnBah, this recording screen adaption is a hack....
00:41:28midkayamiconn, which?
00:42:10amiconnThe one committed on 25 Mar 2006, adapting the recording screen to the remote
00:42:26preglowworks ok enough
00:42:55amiconnpreglow: Maybe it works okay on iriver main LCD and ipod, but it's still a hack
00:43:26amiconn...and it brought that double-height peakmeter which has to go for archos, and should also go on the iriver remote
00:44:30preglowit does indeed
00:44:42amiconnThis would allow the sample frequency to be displayed on the remote
00:44:56amiconn...which happens to have 8 lines with the sysfont, same as archos
00:45:20amiconnThe double-height peakmeter expands this to 9, and makes the sample frequency display vanish
00:45:25sharpewhich sounds more alarming, the black death or anthrax?
00:46:03preglowanything with 'black' in it sounds alarming
00:49:30 Quit tvelocity ("Ex-Chat")
00:52:59sharpethe nursery rhyme, "ring around the rosie" is in reference to the plague
00:57:21 Join webguest13 [0] (
00:57:36webguest13Any 4G devs in?
00:59:30 Quit wehn ()
01:00:05 Join jeremy [0] (i=katzj@freenode/staff/jeremy)
01:10:01 Join fiftyfour123 [0] (
01:10:19preglowanyone here keen to test spdif recording?
01:10:47amiconnI can't, due to lack of proper optical cable
01:11:34sharpei can't... due to lack of... hardware?
01:12:16jeremyanyone know if someone is actually working on plugging the ipod accessory protocol stuff up before I start looking into it?
01:12:33sharpedon't think anyone is.
01:12:53amiconnI probably will, but not before the 3.0 release
01:13:04amiconnThe real problem is the serial comm afaiu
01:13:28preglowbut ok, if no one can test, i'll just commit it when it works here
01:13:50preglowi'm tired of this remote ssh coding
01:14:01amiconnipl doesn't have serial drivers for any of the newer ipods, so we need to find info about the serial port ourselves
01:14:32jeremyamiconn: how so? (... as I've only started glancing through the code and the info on the protocol itself)
01:15:10amiconnYeah, but you'll also need a driver for the port you want to run the protocol on...
01:15:28 Quit fiftyfour123 ("Chatzilla 0.9.71 [Firefox]")
01:16:35jeremy*nod* that could make things more "fun"
01:17:11preglownot only fun, but also not a complete waste of time
01:18:13kapheineamiconn: i was able to get gcc 4.0.2 to build by removing the assert that was causing the ICE. i really, really don't recommend doing that because i have no clue if the assert is wrong or if the code that is causing the assert to trigger is wrong. the only reason i am even trying it this way is because someone else ran into a case where an assert in do_SUBST was incorrect for his IA-32 platform..
01:19:16preglowworth a shot
01:20:03preglowtried building rockbox with it?
01:20:38kapheinei'm working on that now, but unfortunately i have no hardware to test it on.. i don't have any m68k-based players.
01:20:45amiconnHmm, someone wasn't fully alert when hacking the recording screen. We do have enough space to show the sample frequency within 8 lines and the double height peakmeter
01:20:51kapheinei have a ton of coldfire boards at work though, so i can try running something on those on monday
01:22:30amiconnWould make sense to compare the binary produced by m68k-elf-gcc running on amd64 with the binary produced by the same m68k-elf-gcc version running on i686
01:22:34 Join ronbo [0] (
01:22:49 Quit webguest13 ("CGI:IRC (EOF)")
01:23:12kapheineyeah, or even the .S file
01:23:46kapheineoh well in the case of rockbox i guess that wouldn't work, we'd want to compare the entire thing
01:23:48amiconnHmm, now I am confoozled. Obviously I am unable to count correctly :/
01:25:13***Saving seen data "./dancer.seen"
01:25:14 Part ronbo
01:25:25 Join Inc [0] (n=inc@arcane/developer/inc)
01:25:29IncThe requested URL /daily/ipodvideo/ was not found on this server.
01:25:35Incany ideas?>
01:27:30oboInc: try a CVS build instead
01:27:40 Join blindx [0] (
01:28:34oboLooks like all the daily ipod builds 404 - also the iaudio
01:29:03blindxHey, I get rockbox error: -2 when trying to boot to rockbox.. what did I do wrong?
01:32:31Incso I need to grab that one?
01:33:01obograb your build from the bottom of
01:33:46Incand this will still work with the installer right?
01:34:06bluebrotheryou can also try an older one:
01:35:17 Join Haykuro [0] (
01:35:31HaykuroHello! Anyone here active that can help me out?
01:36:27HaykuroThx, sharpe :)
01:36:35HaykuroOkay well, I have a 2GB nano
01:36:44Haykuroand I installed rockbox without a problem
01:36:49Haykurogot the JPEG viewer
01:37:06Haykurothen noticed the RVF extension for videos (after reading the manual for nano)
01:37:18Haykuroso i tried getting the video.rock file
01:37:21Haykurobut only found
01:37:29Haykuroerm.. hold on
01:37:31Haykurolet me get the link
01:37:44Haykurobut that link is dead
01:37:47Haykuroand no one else has it
01:37:50sharpeas far as i know, the rvf format is only supported on archoses
01:37:56Haykurohold on
01:38:11sharpeand it's only in grayscale.
01:38:27Haykurois there any other way to play videos?
01:38:43Haykuroscroll down a bit
01:38:50blindxWhat does "Rockbox error: -2" entail?
01:38:58sharpeon ipod videos, you have to boot into the apple firmware. but there aren't any ways for other players, except the archoses.
01:39:03HaykuroA.1. Supported File Formats
01:39:15Haykuro.rvf is one of them
01:39:25sharpei think that's just in general.
01:39:25midkaythen it's rather a mistake, Haykuro.
01:39:44midkayit doesn't work on anything but the bitmap LCD archos players.. supported ones.
01:39:45amiconnHaykuro: The manual is far from finished, especially for targets which won't be supported in the 3.0 release
01:39:45preglowi can't find the EBU1RCVCCHANNEL1 description in the coldfire docs
01:39:55amiconnThis is simply a leftover from the archos manual
01:40:59amiconnblindx: Where do you get this?
01:41:14blindxWhen turning on my iPod (color)
01:42:59amiconn-2 means the rockbox.ipod file is smaller than 4 bytes, i.e. corrupted
01:43:33blindxum. so i should download a new copy and swap em out?
01:44:00blindxheh. 0 bytes. wonder how that happened..
01:44:39Haykurookay, thx anyways.. just outa blight i'll mention wut i've accomplished.. i took the video.rock files from the archoses download cuz i didnt kno wut else to do, set the viewer properties and now it sais "Incompatible Module" which im guessing its been programmed to check which rockbox its for
01:45:11blindxThanks, amiconn. lets see if it boots now..
01:45:39blindxnow i got a -1 :\
01:45:52sharpewell, you're progressing :D
01:46:16amiconnIt's a step back
01:46:21sharpethat means there isn't rockbox.ipod
01:46:27sharpefor you, anyway.
01:46:29amiconn-1 means the bootloader can't find rockbox.ipod at all
01:47:05blindxit's not there.
01:47:18preglowi get random hangs when recording
01:47:35sharpeblindx: maybe you should like, not use whatever you currently use to unzip it...
01:47:36blindxAm I missing something here? :| I copy and paste it. It shows.. shows as the correct sizes, but when I unplug, it 'disappears'
01:48:31 Quit bluebrother ("Leaving")
01:48:51amiconnI guess you have write caching enabled for the ipod volume. Either disable that, or don't just unplug but safely remove the ipod
01:49:12blindxsafely remove :\
01:49:18blindxhow do you.. safely remove?
01:49:31Inc /awayyyyyyyyyyyy
01:49:34Haykurowuts the advantages of rockbox aside from using the plain old regular ipod software if it doesnt do anything different aside from skins? (im not hating against RB im just wondering is there some features im looking over?)
01:50:43Haykuroyea, but the plugins r to do useless things, screen savers (waste battery mode), jpeg viewers (ipod software views them already..), Doom (probably one of the only reasons i like RockBox compared to iPod Linux that the RB version has sound)
01:51:08Galois ?
01:51:22Galoisor how about the short version: Rockbox Does Gapless Playback
01:51:43Haykuror there documentations to code ur own plugins, maybe i can do my own video plugin (or any1 already try 2?)
01:52:27amiconnGapless playback, really flexible playlist system, doesn't depend in itunes, plays way more formats......
01:52:36Galois ...
01:52:46sharpethat's for twiki...
01:54:09Haykurowuts twiki? anywho, this seems enough if i ever require help is there someone that i can contact to inform me some extra info for this?
01:54:29sharpetwiki is the wiki that rockbox runs on
01:54:34sharpethe website that is
01:54:34Haykuroyea, thought so
01:55:23Haykuroim new to RB just found it 2day, and loved it for about 2 seconds till i noticed no video, its the only thing thats bugging me. aside that i like the interface and all. thx again guys, help was apreciated
01:55:30amiconnpreglow: Archos shows the selected recording sample rate even for spdif, not the actual one.
01:55:40preglowyeah, looked like it
01:55:53sharpeand you can also figure out more by looking at the source of plugins
01:55:56Galoisthe nano can't play videos anyway even without rockbox
01:57:02amiconnHaykuro: The newer rockbox targets will perhaps play video, one day. But it's certainly not a priority. First of all rockbox is a powerful dap firmware, and the primary purpose of a dap is to play music
01:57:44HaykuroGalois, i kno, but the ppl over at iPod linux got video working, and if i somehow accomplish video on RB i'd love RB and say its 10x better than iPod linux
01:57:58amiconnpreglow: Ideally this should be fixed, so that it displays the detected samplerate for spdif
01:58:34GaloisI use rockbox because ogg vorbis sounds better than AAC at low bitrates and nanos have so little disk space that sound quality matters
01:58:40sharpeHaykuro: ipodlinux is also specifically for one platform, rockbox coveres 14 different targets
01:58:42Shadowarrior13Is anyone IN the ipodlinux room?
01:58:53preglowamiconn: yeah, but i have no idea how to do that for archos
01:58:54Shadowarrior13I'm banned for some odd reason, considering I haven't been there for weeks.
01:59:04preglowon iriver i'd just need a simple feedback variable from pcm_recording.c
01:59:17preglowthat, or just put some low-level code in recording.c
01:59:27amiconnpreglow: Just add that for swcodec, and keep in mind not to tie it too closely to that
01:59:30amiconnOn archos, we'
01:59:52amiconnll need a feedback function somewhere in the recording code that reads the mas
02:00:21preglowwe'll see, i'll just keep it as it is in the first commit
02:00:44preglowif i ever get it to work completely, that is
02:02:58amiconnWe already have a function to read back the last encoded mpeg header from the mas
02:03:09amiconnIt's simple to extract the sample frequency from that
02:03:28blindxhey, i fixed it :D
02:04:11preglowamiconn: oh, right, so that's the only way of discovering it?
02:04:13amiconnFor simplicity, we can probably always use the readback value
02:04:33amiconnpreglow: Well, we have to ask the mas, or check the bitstream ourselves
02:04:37 Join whatboutbob [0] (
02:05:02blindxnow, where should i copy songs to to play them on rockbox?
02:05:10 Quit TCK (Read error: 110 (Connection timed out))
02:05:18sharpeto your ipod :)
02:05:23amiconnThe mas will tell us both the mpeg frame header bits, and the spdif input channel status bits
02:05:28Galoisanywhere except iPod_Control
02:05:29whatboutbobpreglow: still need a spdif tester?
02:05:42amiconnNot sure whether the channel status bits include the sample rate
02:06:03preglowwhatboutbob: sure
02:06:13preglowwhatboutbob: do you do patches?
02:06:26whatboutbobpreglow: sure
02:06:34preglowaight, i'll hook you up
02:08:27amiconnBah, it looks more and more like maybe-day instead of mayday
02:08:30preglowit'll bork your settings
02:08:47*amiconn just discovered that another long-standing bug is still there
02:08:48preglowamiconn: haha, you expected us to actually make that deadline?
02:08:53preglowthere's tons of stuff left to do
02:09:01preglowand it was an optimistic date from the start
02:09:15amiconnGood old panic: recfile: -1 still present on archos
02:10:12whatboutbobpreglow: will i need to reset my config or anything?
02:10:16 Quit stoffel (Remote closed the connection)
02:10:23preglowwhatboutbob: nah, that should happen automatically
02:10:36preglowi bumped the config since i removed some unused option
02:11:28 Part Haykuro
02:12:25 Join wehn [0] (
02:14:32whatboutbobergh...its been a while since i updated source (i've just been lazy and have used Mmmm's patched build) when i'm typing cvs up -dP I'm getting an error msg: No CVSROOT specified.
02:15:12whatboutbobwhat am i screwing up?
02:16:54preglowthe cvs setup
02:17:11sharpeseems logical
02:17:35 Quit PaulJ (".")
02:17:42preglowyou probably need to check the repo out again
02:23:00sharpeguess what i just finished? :D
02:24:02 Join ashridah [0] (
02:24:19 Join Haykuro [0] (
02:24:21sharpethat puncuation didn't go correctly.
02:24:53Haykurohey sharpe, i got a quick question, once i have a plugin programmed do i just compile like a normal C++ app or is there a special compiler?
02:24:53preglowwhatboutbob: so, any action?
02:25:14sharpeHaykuro: rockbox is written in c
02:25:34sharpeHaykuro: and you should also compile it with rockbox, as it uses functions referenced in it
02:25:35Haykurothat was my second guess
02:25:54Haykurohow wud i compile with rockbox?
02:26:08sharpedo you have a build enviornment already?
02:26:27Haykuroim sticking simply to notepad and a general c/c++ compiler (gcc, g++)
02:29:24scottderSo, any thoughts on Apple moving to sigmatel for their flash players?
02:29:44preglowHaykuro: doubt it
02:29:52scottderWill it impact rockbox much?
02:29:59preglowif they move to sigmatel, then hell yes
02:30:01sharpescottder: grr. it just makes me so angry. grr.
02:30:08whatboutbobpreglow: reinstalling cygwin :(
02:30:08preglowlet's just hope they move to something generic
02:30:15preglowand not some bloody dsp thing
02:30:53sharpewe could also hope for apple's open source mp3 player, but, you know, that's never going to happen. :D
02:30:58kapheineamiconn: i did a 32-bit and 64-bit build (with the modification) of 3.4.6 and built rockbox with both of them. the output is exactly the same. which doesn't mean there isn't a ticking time bomb, of course..
02:31:04scottderThat bad huh?
02:33:11amiconnkapheine: Sounds good so far.
02:34:00amiconnBuilding for all coldfire targets and comparing binaries would give a bit more backing. We could also compare plugin & codec binaries
02:34:08 Quit Haykuro ("Thanks alot RockBox crew ;)")
02:34:15amiconnCould you stick your patch somewhere?
02:35:16kapheinei was only able to get one coldfire target to build at all, the others complained about not being able to fit in PLUGIN_IRAM, but i didn't play around with it much to see why that was happening.
02:35:22kapheinesure, give me a sec
02:36:41preglownot enough room in iram? sounds weird
02:40:32 Quit Zetan ("Obviously got better things to do :P")
02:42:13 Join OnkelJonas [0] (
02:42:48kapheineamiconn: it won't apply to 4.0.x but by looking at the patch it should be obvious how to do it manually
02:44:26 Quit mikearthur (Read error: 104 (Connection reset by peer))
02:44:31kapheinehah. My original plan was to find something interested to work on for my iPod nano and instead I spent all of my time trying to build a cross-compiler for a platform I don't have any hardware for.
02:45:44preglowand god bless you for it
02:47:40 Quit Moos ("Glory to Rockbox !!!")
02:49:32 Quit OnkelJonas (Read error: 54 (Connection reset by peer))
02:50:44 Join RmnT [0] (
02:51:15preglowwhatboutbob: i need to sleep now, if you've got some feedback, just spill it here and i'll read the logs when i get up tomorrow
02:51:50whatboutbobwill do
02:51:54RmnThi all
02:51:57 Join quobl [0] (n=quobl@tor/session/external/x-fda72194dca205bc)
02:52:04RmnTis anyone here willing to help me with problems I'm having compiling rockbox for my 5G iPod?
02:52:36 Quit obo (Read error: 110 (Connection timed out))
02:52:48RmnTI am able to compile the code but when I run the "make zip" command I get some heap allocation errors
02:53:08RmnThas anyone seen this before?
02:53:36RmnTI've reinstalled cygwin on my WindowsXP but that did not help
02:53:42 Join Kyomi [0] (
02:53:44 Quit actionshrimp ("( :: NoNameScript 4.01 :: )")
02:53:48KyomiAlright.. I have a question for you all
02:54:16kapheineRmnT: what command is running when you get it?
02:54:17KyomiI got the latest build and I even SAW it copy over SID.codec and then when I try to play a SID file it says No Codec Found For...
02:54:38KyomiWhat would you all recommend?
02:57:34RmnTKyomi, I don't think there's anyone alive in here. I asked right before you did but I got no answer.
02:57:48kapheineRmnT: i just asked you what command was running :P
02:57:49Kyomikapheine gave you an answer :P
02:57:57 Quit muesli__ (Read error: 110 (Connection timed out))
02:58:01Kyomikapheine: Any suggestions for me? :)
02:58:12 Part bon
02:58:50KyomiI was thinking just delete all rockbox folders and start fresh
02:58:52RmnTkapheine: oh? I'm sorry. I do not see that.
02:59:39RmnTlet me try that again and I'll tell you
03:00:03 Join muesli__ [0] (n=muesli_t@
03:00:14KyomiHelp me :D
03:00:23kapheineKyomi: i've been using rockbox for about 2 days now, so i'm not really authorized to answer :P but for a sanity check, try browsing for the codec file on the device itself
03:00:40KyomiI got the latest build and I even SAW it copy over SID.codec and then when I try to play a SID file it says No Codec Found For <the SID file>
03:01:36kapheinedoes it match codecs with files by extention or by headers?
03:02:41sharpei had no idea there was an sid codec now
03:04:04 Join LordOfThePings [0] (
03:04:19 Join OnkelJonas [0] (
03:04:20KyomiExperimental build there is <3
03:04:55sharpesure you updated everything when you copied it over?
03:05:01kapheinei just did a CVS update and i don't see it, maybe i'm crazy?
03:05:16Kyomikapheine It's not in the cvs thingy
03:05:22KyomiIt's in the experimental build
03:05:23Kyomi<4303hp 2373m 1901mv 0aqt 208qps>
03:05:23KyomiAtlas OOCs 'silly'
03:05:37KyomiOk.. something hijacked my clipboard
03:05:51sharpe4303 horse power?
03:06:02kapheineit's a dinosaur
03:06:18kapheineis there anything like but more detailed?
03:06:38kapheineMore of an iPod TODO list, I suppose
03:06:55Kyomisharpe: It's a MUD
03:07:03Kyomi4303 is hitpoints.. aka health
03:07:11Kyomimana is next then move points
03:07:20Kyomithen the autoquest timer and quest points
03:07:27sharpei see.
03:08:00sharpei've never been much of a mud person
03:09:18 Quit OnkelJonas (Read error: 104 (Connection reset by peer))
03:09:18 Quit whatboutbob ("CGI:IRC (Ping timeout)")
03:09:33KyomiSo.. noone has any suggestions?
03:09:40KyomiI mean it worked fine before
03:10:07 Join MrSinister [0] (
03:10:22kapheinedid you try doing a clean build?
03:11:11ashridahKyomi: someone may have forgotten to update the file that associates codecs with songs.
03:11:21 Join OnkelJonas [0] (
03:11:29 Part OnkelJonas
03:11:49KyomiDo you need the patchset for midi stuff to play Doom II?
03:12:04KyomiAnd did anyone update the Doom for music or get the MIDI thing to work?
03:12:05ashridahi didn't think they'd added midi support for the doom plugin yet
03:12:23KyomiThere is for sound effects
03:12:43sharpeheh, my doom builds are like, broken. not sure why.
03:13:08kapheineKyomi: is there an entry for sid in metadata.c?
03:13:10MrSinisterany info on the "vintersorg"-torrent? been stuck for days with only .02 left...
03:13:21kapheineor tree.c i guess
03:13:26sharpeMrSinister: hmm?
03:13:41MrSinisterwrong chanek?
03:15:46MrSinisterprobably wrong server, you dont discuss things abut here?
03:16:30MrSinisterk, sorry
03:16:42sharpethis is for the digital audio player firmware,
03:17:16 Part MrSinister
03:19:13KyomiDo you think maybe it's my batch file that removes all the stuff I dont want? Like themes/wps/some rocks
03:19:21 Join MrSinister [0] (
03:19:23 Quit LordOfThePings ("CGI:IRC (EOF)")
03:19:53 Quit blindx ("Leaving")
03:19:55 Part MrSinister
03:20:26KyomiWhat are to codepages for?
03:22:12 Join MrSinister [0] (
03:22:53scorcheMrSinister: server,
03:23:53sharpei remember what i was going to do now
03:24:14scorcheget laid?
03:24:30sharpesadly, no.
03:24:34RmnTok I'm back with more info about my "make zip" problem
03:24:43RmnThere's the error I get
03:24:44RmnT4 [main] find 156480 fhandler_dev_zero::fixup_mmap_after_fork: requested 0x480000 != 0x0 mem alloc base 0x480000, state 0x2000, size 1040384, Win32 error 487
03:24:50MrSinisterschorche: thanks
03:25:08RmnTcan anyone help me?
03:25:14***Saving seen data "./dancer.seen"
03:25:28 Part MrSinister
03:25:30sharpethat sounds like a... personal problem(tm) to me? :D
03:26:15 Join Paul_The_Nerd [0] (
03:26:22kapheineRmnT: but what is reporting the error? make itself? a zip program?
03:27:28RmnTit's make itself. After I compile all the code I enter the command line "make zip", just like in the instructions in say
03:28:27kapheineah well i'm not familiar enough with windows to know what 'error 487' means :P but you should see what command make is issuing that causes the crash, and try to run that command manually
03:28:37sharpeit may be zip
03:28:41kapheineIt mentions a fork problem, so it may have to do with make trying to spawn off the other process
03:28:45sharpeas i'm guessing make is running fine...
03:28:54sharpeif you can build it
03:29:29RmnTyes It builds fine, although I need ti increase my memory heap size to compile
03:29:54RmnTI'll look up what's Win32 error 487
03:30:16 Join whatboutbob [0] (
03:30:17kapheineYeah that definitely sounds shady, unless you have a serious lack of ram :P
03:30:30RmnTI've been fighting with cygwin and this thing for 3 days now
03:30:38kapheineDid you build the cross compiler yourself?
03:31:10RmnTno. I donwloaded everything followinf the instructions
03:31:30whatboutbobergh...i'm such a clueless noob...i've reinstalled cygwin but now can't even see how to run it...
03:31:31RmnTand I do have 1GB or RAM on my system and nothing's running while I compile Rockbox
03:32:30kapheineI'm pretty clueless when it comes to Windows and even more so with Cygwin, but you could try googling for heap problems with cygwin or something like that
03:32:30RmnTwhatboutbob: there's a batch file in your installation folder. run that one
03:32:37sharpeRmnT: well... tried rebooting?
03:33:09RmnTsharpe: i rebooted many times, yes
03:33:21sharpewell, that solution's out then.
03:33:34RmnTdefinitely. that was my first try
03:34:15sharpetried reinstalling cygwin from a different server?
03:35:26Paul_The_NerdThe fact that he's the only person I've heard of who needs to increase his Cygwin heap size just to install makes me think the problems something a bit more obscure.
03:35:37Paul_The_NerdI mean "Just to compile" rather
03:36:10RmnTPaul: I agree with you
03:36:27RmnTI'm googleing the Win32 error 487
03:36:41whatboutbobcygwin.bat?rmnt: thanks, but i must've screwed the setup cos i just get a prompt that i can't seem to do anything with
03:37:24RmnTwhatboutbob: yeaj I hear you . I had to try the installation a million times before I got one that worked.
03:37:54 Quit jeremy (Connection timed out)
03:38:06whatboutbobi don't remember cygwin installation being this difficult...(or perhaps the wiki was just clearer 6 months back)
03:38:37kapheinePaul_The_Nerd: Yeah, that's true. Unless it is some interaction between Cygwin and some other setting in Windows.
03:40:35sharpewell, that kind of sucks. my friend who's father died recently, her cousin just died.
03:40:41 Join jeremy [0] (i=katzj@freenode/staff/jeremy)
03:43:46 Quit muesli__ (Read error: 110 (Connection timed out))
03:45:00RmnTthere's lots of lonks on the web referring to error 487 with cygwin....still reading
03:47:23sharpewoo! firefox has froze with using 138megs of memory, again. :)
03:48:00scorchei have gotten it to >500
03:48:19 Join TCK [0] (
03:49:00sharpeyou wait, i'll get it to use the page file!
03:49:33 Join parametr [0] (
03:49:40scorcheit is at 135 atm
03:49:59scorchethen again, i am using downthemall to dl about 500 images
03:50:15sharpestarting over from 23
03:50:35 Quit parametr (Client Quit)
03:50:55 Join Jungti1234 [0] (n=jungti12@
03:52:18scorchesharpe: you are such an upstanding young man
03:55:13scorcheJungti1234: ...
03:55:32scorcheand the purpose of that was....?
03:56:36midkayreason for linking to it
03:56:39Jungti1234It's advertisement of nike.
03:57:53Jungti1234Because it's nice.
03:58:00scorcheright then...
03:59:59 Nick BHSPitLappy2 is now known as BHSPitLappy (i=Steve-O@
04:02:52 Quit jeremy (Connection timed out)
04:04:15 Join jeremy [0] (i=katzj@freenode/staff/jeremy)
04:06:27sharpeheh, apparantly i made four errors in proofreading something.
04:07:06sharpei've never seen someone write a message so happily.
04:08:21sharpeand the fact if i could read japanese, would make it soo much better.
04:08:24 Part Paul_The_Nerd
04:08:37sharpebut, yeah, i realize that's the windows xp install screen...
04:08:37scorcheit isnt jap
04:09:09scorchewait...yes it is
04:09:22Mikachuno it isn't
04:09:29Mikachuit's korean
04:09:40scorchewait no it isnt
04:09:43sharpewell, it is in a country's language i haven't been to...
04:09:53 Quit RmnT ()
04:10:12scorcheMikachu: i thought i saw "end" somewhere in there, but i realize that i was mistaken
04:11:58 Join Haykuro [0] (n=haykuro@tor/session/external/x-1432a5016040ccb3)
04:12:29Jungti1234I believe that may use windows in my H300 sometime.
04:12:30scorchei dont think downThemAll is working...
04:14:37whatboutbobbah...i'm damn close to giving up on cygwin...i just followed the noob guide with screenshots and i still can't get it to work...
04:14:55 Quit TeaSea ("Leaving")
04:17:53whatboutbobwhen i run cygwin i'm just getting a prompt: bash-3.1$ that i can't seem to do anything with....
04:18:12sharpepath isn't set correctly...
04:19:15whatboutbobany ideas what i need to set it to?
04:19:48sharpei suddenly don't realize what i was talking about, hold on
04:20:12scorcheadd $cygwin\bin
04:21:08sharpeare you starting cygwin with the batch file ?
04:21:42whatboutbobwhen i get the prompt: no. When i run the batch file the screen pops up for a split second and closes.
04:22:29sharpewhat is in the batch file?
04:22:31whatboutbobbatch file is currently:
04:22:32whatboutbob@echo off
04:22:39whatboutbobchdir C:\Rockbox_Final3\bin
04:22:43whatboutbobbash −−login -i
04:22:56whatboutbobthat's it.
04:23:11sharpeis that where the executables are?
04:23:23sharpei'm asking him...
04:23:55whatboutbobthere's a whole bunch of exes in that dir, yeah...
04:24:27sharpetried reinstalling to a different directory? :D
04:25:17KyomiThis is strange
04:25:17whatboutbobyeah...see 'Final3'? ...there's been a 'final2' and a 'final' as well...
04:25:32sharpei see...
04:25:49sharpewant to make a Final4? :)
04:25:51KyomiI've completely re-installed rockbox and theres a SID.codec sitting right there
04:26:07KyomiAnd yet when I try to play a SID file it says no codec found
04:26:31whatboutbobfeels alot like i'm going around in circles. :(
04:27:11sharpewhatboutbob: only if you start to use final, then you're circular. otherwise, so far you're linear
04:27:33whatboutbobhehe...gee thanks...
04:27:39sharpeeh, i try.
04:27:47sharpei'm not very sure about your problem though.
04:27:54sharpehas that happened every time?
04:28:14sharpetried a different drive?
04:28:58whatboutbobnope. i've just installed vmware though.
04:29:22 Join JdGordon [0] (
04:29:45whatboutbobnow i've got no idea how to use this. *rolls eyes*
04:30:04sharpearen't there instructions on the wiki?
04:30:21whatboutbobnot to patch...
04:30:29whatboutbobhehe...i can't even see where the output files are...
04:31:26sharpe//Debian ?
04:31:37whatboutbob...see...that's what i would have thought....
04:32:06whatboutbob...but no.
04:33:08sharpei'm not sure what to tell you then
04:33:21whatboutbobscrew it...does some kinda sole feel like compiling a build with preglows's optical recording patch in it and sharing it w/ me?
04:33:34whatboutbobi just wanna test the bloody thing.
04:33:36sharpewhat target?
04:33:44 Quit jeremy (Connection timed out)
04:33:47sharpeeh, why not...
04:33:57whatboutbobbut i've spent 2 hours frigging around with cygwin and vmware.
04:34:00whatboutbobthanks mate.
04:34:15whatboutbobi just don't have the patience to deal with cygwin anymore today.
04:34:23 Join Paul_The_Nerd [0] (
04:34:38sharpewhere would be preglow's patch?
04:34:56whatboutbobhe posted a link earlier...lemme see...
04:35:04 Join jeremy [0] (i=katzj@freenode/staff/jeremy)
04:35:56whatboutbobwish i knew why my old cygwin stopped working...never had a prob w/ it...
04:36:10sharpenew fangled cygwin :D
04:36:15*Paul_The_Nerd suggests VMWare.
04:36:33sharpehe can't access //Debian ...
04:36:55whatboutbobPaul_The_Nerd: w/ VMWare, when I make a build, how the hell do i access it?
04:37:23Paul_The_Nerdwhatboutbob: Do you have a firewall, like Zone Alarm?
04:37:58Paul_The_NerdOkay, firewalls often prevent access. Try disabling it briefly to see if that's it.
04:38:03Paul_The_NerdAlso, its \\Debian not //Debian
04:38:07Paul_The_NerdThe direction is important
04:38:29sharpewell, we all know my luck with... something i don't have much luck with.
04:39:33sharpespreadable butter with canola oil! oh my god!
04:39:39sharpesuch an innovation!
04:39:55sharpebecause of canola oil! who would've though of that?
04:40:38whatboutbobi think i'm doing something stupid tho....i'm in vmware, in /home/user/rockbox-devel/build_h120...i can see how do i get that to the iriver?
04:41:09Paul_The_Nerdwhatboutbob: The VMWare image is seen as a virtual machine on your network. Start->Run-> \\Debian
04:41:22Paul_The_Nerdwhatboutbob: It's a samba share so you can then just browse to the file
04:42:12Paul_The_NerdOkay, I don't know what "u/p" means.
04:42:16Paul_The_NerdCan you speak in words maybe?
04:42:19sharpegot me too.
04:42:38Paul_The_NerdSame as logging into the VMWare image
04:42:46 Quit Shadowarrior13 (Read error: 110 (Connection timed out))
04:43:50whatboutbobthanks for talking slow for the ultra-noob guys.
04:44:28Paul_The_NerdThe VMWare thing is generally much better than the Cygwin thing.
04:44:53sharpeeven though i don't use it i've got a base build working...does patching in vmware work the same way as cygwin?
04:47:13whatboutboboh do i go back a dir *blushes*
04:47:28sharpecd ..
04:47:29KyomiWould you know anything about how to solve a Codec Not Found for <file> error?
04:47:54Kyomiwhatboutbob: To get to complete home dir just do cd\
04:47:57Paul_The_NerdKyomi: What kinda file is it?
04:47:58whatboutbobsee...that's what i would have thought...but it says command not found...
04:48:11sharpeyou think a lot, don't you?
04:48:22whatboutbobi try not to... :-/
04:49:55Paul_The_Nerdwhatboutbob: cd ~ gets to your home dir. cd .. goes up a level.
04:50:13KyomiPaul_The_Nerd: It's a SID file
04:50:23KyomiThe SID codec is ther... sitting in codecs
04:50:27KyomiAnd it worked before
04:50:34KyomiBut since the updates its... crapped out
04:51:12Paul_The_NerdKyomi: You're using the 1.2b patch?
04:51:18 Join Sinbios [0] (n=Sinbios@
04:51:37sharpei just got a bag of chips out of my refridgerator... that have been in there, unopened... since thursday..
04:51:53 Quit Kohlrabi (Read error: 104 (Connection reset by peer))
04:51:57 Quit ProgramZeta (Read error: 104 (Connection reset by peer))
04:52:37KyomiPaul_The_Nerd: I dunno how to check
04:52:55KyomiPaul_The_Nerd: I don't believe so.. KOSH didn't say anything about it
04:53:11KyomiThe last thing I saw on the SID codec is he/she removed 1.0 and added 1.1
04:55:57Paul_The_NerdKyomi: Well I can't help with the optimized build at all. That error would imply that the proper codec file wasn't found.
04:56:29KyomiBut it's sitting right there in the codecs dir
04:56:36KyomiI can see it in windows and on the h300
04:56:42Kyomih320 rather
04:56:51Paul_The_NerdKyomi: Yes, but are you user that's a NEW version of it, or is that left over from your previous install?
04:57:00Paul_The_NerdThat could be an outdated file
04:57:21whatboutbobduh...i was typing 'cd..' not 'cd ..'. Thanks for your extreme patience guys...and not openly laughing at me. :)
04:57:56KyomiPaul_The_Nerd: I've deleted my entire rockbox dir and re-installed it
04:58:07KyomiI'm pretty sure my batch file isn't the problem
04:58:14KyomiAll it does is remove the themes/wps folder
04:58:20KyomiAnd some of the rocks I never use
04:58:33Paul_The_NerdKyomi: Well, there's something wrong with the experimental build then probably
04:58:36whatboutbob3 hours later and i'm ready to start testing optical recording...
04:59:44Jungti1234Where is tag browser?
04:59:47Paul_The_NerdKyomi: You really shouldn't come asking for help unless you can verify the problem exists with the newest version of everything applicable (which would be a CVS Rockbox with the sid_1_2b.diff)
05:00:13KyomiPaul_The_Nerd: It's in sync with the newest CVS
05:00:33 Quit jeremy (Connection timed out)
05:00:56Paul_The_NerdKyomi: No, it's not. It's in sync with it from whenever he made the most recent update, which can often be a day or two ago. It also has about 50 other patches that could break SID.
05:01:21KyomiI can see this is a losing battle :P
05:01:28*Kyomi gives way
05:01:30 Join jeremy [0] (i=katzj@freenode/staff/jeremy)
05:01:32 Join Gibbed [0] (
05:01:36 Quit Rick (Nick collision from services.)
05:01:43Paul_The_NerdKyomi: I'm just saying that it's impossible to help with a problem that can't even be identified.
05:01:48 Nick Gibbed is now known as Rick (
05:01:50Jungti1234TagCache is ID3 tag?
05:01:53KyomiIt is identified
05:02:02KyomiSid codec can't be found even though its right there
05:02:06Paul_The_NerdKyomi: Alright. Tell me then, which version of the SID patch is he using?
05:02:16Kyomi1.1 as far as I know
05:02:24Paul_The_NerdKyomi: Which is outdated, and doesn't work period any more.
05:02:29KyomiUnless CVS has SID in it
05:02:39KyomiThere we go, problem solved :)
05:02:41Paul_The_NerdKyomi: CVS does not have SID. That's one you can check for yourself.
05:02:50Paul_The_NerdKyomi: You can't *build* Rockbox with the 1.1 patch.
05:03:10Paul_The_NerdSo, it can't be the 1.1, at least not without modifications.
05:03:15KyomiAlright then
05:03:18KyomiIt's the 1.2b :)
05:03:24KyomiSince it works for everything else :)
05:03:59Paul_The_NerdYou're now guessing though
05:04:14KyomiWell there isn't any newer SID codec
05:04:20KyomiAnd you can't build it with 1.1
05:04:24KyomiSo it's 1.2
05:04:27Paul_The_NerdOr 1.2b
05:04:37Paul_The_NerdOr he modified 1.1 to build
05:04:41Jungti1234hey, Tag Browser is "ID3 database" menu?
05:04:44KyomiAlright... I give up, I'm retarded :P
05:04:46Paul_The_NerdAnd because the 1.1 is out of date, it builds but does not actually run
05:04:57KyomiYou win ok? :
05:05:13Paul_The_NerdI'm just saying it's not a good idea to ask for help unless you can actually clearly state the conditions under which the problems occur.
05:05:30KyomiWell.. I did o.o
05:05:34Paul_The_NerdAs it is, 1.2b doesn't even build with the newest CVS, which is why I was 100% sure it wasn't the NEWEST CVS.
05:05:47Paul_The_NerdWhat was "clear" about "I don't know which version of the SID patch he's using, or how old the build is?"
05:05:56KyomiI didn't say that :P
05:06:33Paul_The_NerdWell, you couldn't answer me with a solid answer when I asked.
05:06:50Paul_The_NerdThat's the "conditions under which the problems occur" right there.
05:07:00KyomiAre you trying to torture me or something o.o
05:07:15Paul_The_NerdWell, you keep saying you *did* give the relevant information
05:07:19KyomiI give up and then it's like.. you continue then I start again and then you make me give up again
05:07:24Kyomibecause I did
05:07:27KyomiAccording to me :P
05:07:34Paul_The_NerdAccording to you
05:07:38Paul_The_NerdWho clearly doesn't know what "relevant" means
05:07:57KyomiAnd there you go again
05:07:59Paul_The_NerdIt's like saying "I run windows" and expecting someone to know whether that means XP, 2000, NT, ME, 98, 95, 3.1, 3.11, or 3.0
05:08:12KyomiWell I would assume it's xp
05:08:18Paul_The_NerdThen what do you do if it's not?
05:08:31KyomiBecause I doubt people still use any 3.x to get to the internet
05:08:36Kyomi95 probably not
05:08:38Kyomi98 maybe
05:08:50KyomiPaul_The_Nerd: Ask them?
05:08:58Paul_The_NerdKyomi: And when I asked you, you said "I don't know."
05:09:36KyomiWhat do you want from me? >_<
05:09:53KyomiIt's like you keep shoving my mistakes or whatever in my face
05:09:55sharpeanswers, apparantly.
05:10:15Paul_The_NerdWell you keep saying you didn't *make* a mistake
05:10:17KyomiI give up, you make me continue.. I give up again, you make me continue then insult me
05:11:06Paul_The_NerdAll I'm trying to do is press upon you the importance of actually identifying the full circumstances of a problem.
05:11:36KyomiI did.... but apparently not concise enough for you
05:11:42Paul_The_NerdNotice the word "full"
05:11:49KyomiWhich you fail to mention when I do and then just.... drill me
05:12:01Paul_The_NerdWhat about what you told me was anything near full?
05:12:23Paul_The_NerdYou couldn't tell me the date the build was made, or what version of the offending patch it used. You basically said "I don't know" to the only important questions I had about it.
05:12:31scorcheKyomi: windows crashes for me
05:12:33scorchefix it
05:13:03 Quit Haykuro ("i'm off everyone, good bye!")
05:13:17Paul_The_NerdI mean, you ask for help, then say I "drilled" you, when all I did was ask for the information I'd need to at least diagnose it.
05:14:42Paul_The_NerdYou even lied to me when you said "It's in sync with the latest CVS"
05:14:42 Join Gibbed [0] (
05:14:42 Quit Rick (Nick collision from services.)
05:14:54 Nick Gibbed is now known as Rick (
05:15:24sharpeeh, the love.
05:16:38Paul_The_Nerdsharpe: Well, a quick spot of research to me shows the latest Experimental is 14 hours out of date. I'm not a fan of people who ask for help, then won't actually *answer* questions, and then lie instead of actually checking their answer.
05:16:45 Join nave7693 [0] (
05:17:49KyomiYou are still going on about it?
05:17:59KyomiJesus man... let it go already o.o
05:18:22KyomiRight... I do that just to piss you off
05:18:42KyomiI answered your damn questions... maybe not the answers you wanted but I answered them to the best of my ability
05:18:53sharpeoh noes.
05:18:54 Join RmnT [0] (
05:19:06 Part ashridah ("whoa.")
05:19:09Paul_The_NerdKyomi: So your ability does not include "Going back to see in the thread what date he last updated to CVS?"
05:19:28Paul_The_NerdInstead claiming that it's up to date with no evidence to back that up at all.
05:19:36RmnTI'm back....with a bit more info about my heap memory problem.
05:19:43Paul_The_NerdRmnT: Whatcha got?\
05:19:51RmnTI found out that the find command is causeing the error.
05:20:08RmnTthere's a perl script that uses the command "find"
05:20:19RmnTthis is the command line that causes the proble,
05:20:21RmnTfind apps "(" -name "*.rock" -o -name "*.ovl" ")" ! -empty -exec cp {} .rockbox/rocks/ \;
05:20:31 Join ProgramZeta [0] (
05:20:51RmnTif I run that command line it I get the same error I get when I run "make zip"
05:21:16sharpesure it isn't cp ?
05:21:55RmnTwell, the erro says it's "find", not "cp"
05:22:01RmnT5 [main] find 2988 fhandler_dev_zero::fixup_mmap_after_fork: requested 0x480000 != 0x0 mem alloc base 0x480000, state 0x2000, size 1040384, Win32 error 487
05:23:40Paul_The_NerdKyomi: Look, I'm sorry for getting snappish at you. I've just been getting frustrated because people keep deciding my questions are unnecessary, and answering with what they think I'm asking, instead of simply giving me exactly what I ask. It makes it vastly harder to help people if they decide they know better than the person they're asking for help.
05:24:09Paul_The_NerdAnyway, off to dinner.
05:24:10 Part Paul_The_Nerd
05:25:15***Saving seen data "./dancer.seen"
05:29:44scorcheim out
05:29:51scorchesharpe: wanna come with me?
05:29:58sharpewhy not, still nothing else to do.
05:30:16scorcheyou can probably get laid
05:30:23scorchewell, i take that back
05:30:38scorchemaybe if someone gets really drunk you can =0
05:30:50sharpethat sounds like a personal insult
05:31:01scorchewell, there wont be anyone your age there
05:31:13sharpei know
05:31:15scorcheyou can still come though
05:31:20sharpeeh, it's okay
05:31:27scorchefine...i see how you are
05:46:00lostlogicgood evening!
05:47:26lostlogicwow, hoppin' place.
05:48:05Mikachui meant to say hi but apparently i forgot to actually type it
05:52:15 Part nave7693
05:52:44RmnTand...for those that are still interested in my memory heap problem: I have a workaround!! :)
05:53:02RmnTI replaced teh find with a straight cp
05:53:13RmnTvery simple
05:53:23RmnTnot really solverd the problem, just worked around it
05:53:41RmnTbut I can do a full rockbox build now
05:53:49sharpethat's good
05:53:59RmnTyep. that's all I wanted anyway
05:54:14RmnTthanks for your interest, sharpe
05:54:51sharpeheh, you're welcome
05:56:13 Quit Kyomi ()
05:56:43RmnTI can go cook dinner now. Take care y'all.
05:57:22 Quit RmnT ()
06:00:21 Join amiconn_ [0] (n=jens@rockbox/developer/amiconn)
06:05:58sharpedo i really, want to try to get multicolor text mode working...
06:07:48 Join Zzompp [0] (
06:07:49lostlogicof course you do!
06:08:00lostlogicWhat's a C64 without blinknig yellow on red text!
06:08:16sharpeyou want to write it ?
06:09:03lostlogichmmm... got enough trouble with playback currently :-P
06:09:21sharpewhat about the code for the bitmap modes?
06:09:23lostlogicplus, after 3.0, I think I'm going to take a rockbox break and work on writing an SCM or something.
06:09:53sharpepretty acronym. :D
06:10:40lostlogicSource Code Management system or Version Control System
06:10:47lostlogicwhat CVS is, sorta
06:10:55kapheineWhy bother, darcs already exists :P
06:11:01sharpewhere did vcs come from?
06:11:08lostlogickapheine: whadat?
06:11:43kapheinelostlogic: I couldn't live without it
06:12:56lostlogicnope, darcs doesn't do what my VCS will do.
06:13:12kapheineSolve world hunger?
06:13:14sharpewhat will your VCS do?
06:13:45lostlogicnot telling. I might try to sell it and I don't want you bastards beating me to market (even though it will probably end up open source)
06:13:59sharpeyeah, i live to thrive off your ideas. ;)
06:14:15kapheineWell, as great as darcs is, SCM is hard and there is always room for improvement, so best of luck :P
06:14:35Galoismmm bitkeeper
06:14:35kapheineIt's one of those things that seems impossible to make perfect
06:14:40kapheineGalois: do you use bitkeeper?
06:15:23kapheineOh no, I hope I didn't bring back some childhood memories
06:15:31sharpelostlogic: can i help? pretty please? heheh
06:16:54sharpeit's funny because, it's me :D
06:17:51lostlogickapheine: darcs feels a lot like Git. Why darcs over git?
06:18:57 Quit amiconn (Read error: 110 (Connection timed out))
06:18:57 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
06:19:11kapheinelostlogic: well, for one git isn't an SCM, it's a layer down. In fact, darcs has some initial support for using git as a backend. But beyond that, the major feature of darcs is patch theory. Which, in very simple terms, means it is meant for pulling in individual changes, rather than a state of a given tree.
06:19:40kapheineThat is a very simplified explanation of it, but also I have about 8 or 9 beers in me, so don't expect me to explain the intricate details of patch theory :P
06:19:43lostlogicokokok, why darcs over cogito then :-P
06:20:19lostlogicah, that does sound nice
06:20:31kapheineBut I will point out that cogito still wins for large source trees. Darcs is still young and patch theory is processor intensive. It still has trouble with trees as large as the Linux kernel.
06:20:44lostlogicso you can say "Hey, give me the cahnges from transaction X" and supposedly get just those changes applied in a Best Guess way to your current tree
06:21:43sharpehmm. why is it i'm writing an emulator?
06:22:06sharpeoh i'm missing a page here.
06:22:10lostlogicIf I didn't have zero interest in learning haskell, I would like to incorporate my ideas with darcs. In theory, what I want to do is make an VCS that is _parse tree_ rather than _text_ based.
06:22:12kapheineRight. Which can be very handy when you have a stable tree and an unstable tree, and you want to pull bugfixes from unstable into stable, but not other changes
06:22:20lostlogickapheine: yep
06:22:59kapheineDarcs allows different patch types, but currently there is only the text based and searc/replace type changes. Nobody has implemented anything else, but thigns liek parse tree changes would be great.
06:23:25kapheineIf you do start a VCS, you should try darcs for a bit. It's very possible it doesn't do all of what you want, but it should at least give you ideas to incorporate into your own system.
06:23:48sharpeyou should name it something catchy.
06:23:54sharpelike, stick.
06:24:16lostlogic*nod* Knowing myself, I won't actually end up wtih anything useable, but I'll learn a lot in the process of not making anything useable :)
06:24:21kapheinepancake sausages
06:24:45sharpeor gumby
06:24:54sharpeor pokey
06:25:02lostlogicas of right now, my working name is STYLE
06:25:06lostlogicdon't know what it stands for yet.
06:25:24sharpeshall try yelling less, everyone.
06:25:47sharpeit works...
06:25:54kapheineStyle Tyle Yle Le E
06:26:06whatboutbobpreglow: Sorry boss, but I can't get spdif recording working.
06:27:05whatboutbobfirst my rig: at853 mics > 3wire battery box > denecke ad20 > iriver H120
06:27:19 Quit axion_ (Read error: 110 (Connection timed out))
06:27:31whatboutbobthe denecke ad20 outputs 44.1Kz/20 bits
06:28:13whatboutbobFirst step was tp confirm that the rig worked in the iriver f/w (the ad20's been in the cupboard for a while). It worked fine.
06:28:42 Quit ProgramZeta (Read error: 104 (Connection reset by peer))
06:29:30JdGordonhas any1 tried recording live gigs on the h300? is it pointless?
06:29:35whatboutbobThen I booted into rockbox, selected freq: 44.1, Source: digital, and left all other settings untouched.
06:29:56whatboutbob(will answer that in a minute jdgordon)
06:31:31whatboutbobAttempted to record: Size got to 8.00KB then stopped. Time kept running. I left it for a minute and then pressed stop. Nothing happened so I eventually reset. I attempted to salvage the file to no appears nothing was saved.
06:31:54sharpesuch a tragedy...
06:32:16 Quit Jungti1234 ()
06:34:09whatboutbobI changed Freq to 48khz, attempted to record again. Everything appeared to be working (well, the file size and time kept ticking), but when i stopped after 3 minutes and opened the file in adobe audition i'd recorded 6(!) minutes of silence
06:35:03 Join Paul_The_Nerd [0] (
06:35:41 Join ProgramZeta [0] (
06:39:23 Quit whatboutbob ("CGI:IRC (Ping timeout)")
06:47:21 Join whatboutbob [0] (
06:51:32whatboutbobpreglow: so that's where i'm going to leave testing for now. I'm guessing (and what the hell would i know?) that its not liking the sample rate.
06:53:23whatboutboboh, i forgot to mention that after successfully saving the 48khz file i shut down the iriver, plugged in the usb cable and restarted - but got an ata 11 error. This was replicable.
06:54:07 Join quobl_ [0] (n=quobl@tor/session/external/x-d3d9e2024fa7346f)
06:54:23whatboutboband from comments in your code it sounds like Freq selection shouldn't change anything, so I tried recording w/ 44.1khz multiple times to no avail.
06:54:58whatboutbobJdGordon: I've not recorded w/ the H300 but I've mates that have quite successfully.
06:55:31JdGordoni tried it once and i was on the back wall and pretty much all i got was the bass..
06:55:58*JdGordon is going to a concert tomorow night and wants to record it :D
06:56:38whatboutboboooh...are you talking about just using the internal mic?
06:57:13 Join MusiFreq [0] (
06:57:25whatboutbobyeah, i wouldn't bother.
06:57:39whatboutbobyou need external mics tomake it worthwhile.
06:57:53JdGordonthats what i guessed
06:58:12whatboutbobyou can pick up budget mics for ~AU$100....
06:58:28 Join Arrogant [0] (i=Scott@
06:58:44JdGordonhahahaha budget!
06:59:18whatboutbobwhich gig are you hitting up?
06:59:49JdGordonopeth tomorow night
07:00 wonder you just get bass if that's the stuff you're recording. :)
07:00:48 Quit quobl (Remote closed the connection)
07:01:09whatboutbobyou can borrow my mics if you really want to...
07:01:38JdGordonna, i like singing along and it would rec the recording :D
07:01:48whatboutbobhehe...that's true. :)
07:02:07 Join RoC_MM [0] (
07:02:14whatboutbobthe forum's one of my fave venues.
07:02:20 Quit MusiFreq (Remote closed the connection)
07:02:41 Join MusiFreq [0] (
07:11:20 Quit whatboutbob ("CGI:IRC (EOF)")
07:14:43 Quit TCK (Read error: 110 (Connection timed out))
07:18:34 Quit RoC_MM ("Leaving")
07:23:57 Quit MusiFreq ("Lost terminal")
07:24:02 Join damaki_ [0] (
07:25:17***Saving seen data "./dancer.seen"
07:30:07 Quit RotAtoR ("zzzzzzzzz")
07:31:21lostlogiccommit and run!
07:33:58Paul_The_NerdBah, if you manage to break something with *that* I'll be impressed.
07:36:12 Join KlrSpz [0] (
07:39:38 Quit damaki__ (Read error: 110 (Connection timed out))
08:00:32 Join mymomthelush [0] (
08:01:58 Quit JBGood25 (Read error: 110 (Connection timed out))
08:18:24amiconngood morning
08:20:07KlrSpzanyone have a copy of the WPS Simulator?"> shows that he's no longer hosting them
08:21:15Paul_The_NerdWell, you could always just build one.
08:21:43Paul_The_NerdThose ones were likely quite out of date anyway.
08:22:17KlrSpzon win32?
08:23:10Paul_The_NerdDoes it not work any more?
08:25:40KlrSpzsupposedly not.. i mean he doesn't eve have links anymore
08:26:06Paul_The_NerdKlrSpz: So, you haven't tried for yourself
08:26:15Paul_The_NerdHe says he can't build a Win32 SDL sim *from Linux*
08:26:32Paul_The_NerdThat note is also at least 23 days old.
08:27:19KlrSpzwell i can't find it to build it anyway... i must be missing the link somewhere
08:27:48Paul_The_NerdWhen you run "make" in your build directory, one of the choices is "Simulator"
08:29:18amiconnconfigure, not make
08:29:24Paul_The_NerdOh, right
08:29:41Paul_The_Nerd../tools/configure, pick a target, then S for sim instead of N for normal
08:30:03Paul_The_NerdThen make, and make install (instead of make zip)
08:30:25amiconnThe cross-compilation of a win32 sdl sim from windows should be possible, but has to be added to configure
08:30:38amiconnErm, from linux of course
08:31:14Paul_The_NerdI think the site's a good bit out of date.
08:32:03amiconnYeah, rashers seems to have little time and/or interest to do rockbox work lately
08:32:56amiconnThe fonts are also a bit outdated; the .fnt files are made with the buggy convbdf that made some glyphs appear half-height
08:33:28amiconnThe .bdf files are fine though
08:34:15 Quit mymomthelush ("Trillian (")
08:36:08KlrSpzanyone have a nanopod skin/protector they can suggest?
08:36:40sharpethink plastic bag!
08:36:51Paul_The_NerdI'm a fan of the iSkins one. It makes it a bit bulkier, but I rather like it.
08:37:21KlrSpzi was thinkin of this one:
08:37:29Paul_The_NerdIt's also really good at keeping the scrollwheel from getting bumped if I have hold off, and makes it easy to press any of the 4 buttons without accidentally scrolling.
08:37:32amiconnHaha, imho that's one off their weird things about most ipods, especially the nano
08:38:19Paul_The_NerdKlrSpz: They use "dirt" as a verb. I don't like it. :-P
08:38:54amiconnWhat's the point in making such a small device but at the same time making it so scratchy that you need to make it larger by adding some protector??
08:38:55Paul_The_Nerdamiconn: The scrollwheel protector has a sorta folded ridge around the outside edge that lets you press the buttons without getting close enough for it to pick up your finger contact.
08:39:09KlrSpzgotta link?
08:39:13*Paul_The_Nerd felt it was too small anyway'
08:39:28Paul_The_NerdI have large, clumsy hands. I don't mix well with small devices.
08:39:47Paul_The_NerdKlrSpz: This is the one I use
08:40:07 Join JBGood25 [0] (
08:40:10sharpei are graphically editing.
08:40:22sharpeand am also incoherent.
08:41:08KlrSpzwhich color do you have paul?
08:41:29KlrSpzthat looks tight
08:41:30 Quit ap0calypse (Read error: 104 (Connection reset by peer))
08:41:32KlrSpzthe black and red one
08:41:38KlrSpzor the orange one
08:41:41Paul_The_NerdKlrSpz: I have the black and green.
08:41:49KlrSpzacid groove?
08:41:55KlrSpzhow does the glowing look?
08:42:03Paul_The_NerdIt's not a glowing one.
08:42:07Paul_The_NerdAt least I don't think it is.
08:42:34KlrSpzah, yeah only 3 of them are
08:42:38KlrSpzthe rest look like they glow though
08:42:46KlrSpzor is there a thin protector on it?
08:43:03amiconnlostlogic: Regarding voice: how much delay to expect from the mix buffer when playback is stopped?
08:44:20amiconnVoice got tons better on swcodec thanks to your rework, but clip preemption still seems to have problems
08:44:55 Quit ProgramZeta (Read error: 104 (Connection reset by peer))
08:45:23KlrSpzPaul_The_Nerd, nm.. just trying to figure which i like better
08:45:28KlrSpzcarbon blast or the black/red one
08:46:07Paul_The_NerdKlrSpz: Honestly, the main reason I ordered an iSkin was I wanted a better case for my H120, and those were on sale at the time, and I thought "Hmm... why not just order cases for both from the same place."
08:46:28KlrSpzbut you would recommend it right?
08:46:35Paul_The_NerdAbsolutely even
08:46:54amiconnlostlogic: I have an idea what might cause this. Is it correct that you didn't touch talk.c, or did I overlook something?
08:47:31amiconnAlso, would talk.c be allowed to preempt a clip immediately on swcodec, causing incomplete frames?
08:51:16HotfusionI wonder. If I take a H3xx rockbox wps theme and make it fit to a x5l would be the best way to adjust the pic sizes the quickest
08:53:13Paul_The_NerdHotfusion: Scale them all to 72% of their size, then scale the pixel coordinates to 72% of their values, then pick a font 72% smaller, then add a padding pixel or two to the backdrop image if it's a wee bit too small as it may be. 72.72727272 is a more valid value, but 73% may result in a too large backdrop.
08:54:53Hotfusionyeah I'll have to play with the font size
08:55:05Hotfusiongot to sleep
08:55:15Paul_The_NerdOoops, not "72% smaller" but "72% of the original size"
08:56:02Hotfusionyeah, fire up photoshop. Cant remember the free one though.
08:56:14Hotfusionname keeps slipping
08:56:31Hotfusionthere is that but there is another
08:57:00Hotfusionit wans't very advanced but you could save multiple formats and very easy to resize stuff
08:57:05Hotfusiondidn't take up much memory
08:57:46 Join TCK [0] (
09:02:21 Quit JBGood25 (Read error: 110 (Connection timed out))
09:04:17 Part Paul_The_Nerd
09:04:26Hotfusionah I remember now
09:04:50 Quit macdonalder ("bye bye ;D")
09:08:32sharpewoo! i did it again.
09:08:38 Join dj-fu [0] (
09:11:28sharpewho wants to see a badly done planet graphic... heh...
09:14:46 Part taz
09:19:15KlrSpzmy last planet endeavour:
09:19:26KlrSpzthat's not too bad
09:19:45sharpedone in gimp without any tutorial...
09:19:57KlrSpzif you take the part that's white and color it to blue or something, then repeat it and make it green, you can have a habital planet
09:19:58sharpei know i'm going to forget how i did it in a few hours
09:20:14sharpei know, it's supposed to be more of a moonish type thing.
09:20:28KlrSpzahh yeah
09:20:53sharpeand supposed to be smaller than that
09:20:53 Join stoffel [0] (
09:20:58KlrSpz!layout/layout2.jpg you can kinda see a moon i have in the top left there (nsfw)
09:21:13KlrSpzit looks like the exact same technique
09:21:30sharpedid you use gimp? :D
09:21:43sharpeahah, one up.
09:22:20sharpei probably could've done a grunge brush type thing...
09:22:30sharpefor a better looking surface.
09:23:13KlrSpzi'm gunna pass out
09:23:19sharpethat's nice
09:23:23KlrSpzgotta dream up my wonder-wps
09:23:47sharpehah, mine's text based.
09:24:03KlrSpzi want mien to be cool so my friends will drool :).. c'ya tomorrow
09:24:06 Quit KlrSpz ()
09:25:18***Saving seen data "./dancer.seen"
09:25:51 Join MarcoPolo [0] (
09:26:59 Part amiconn
09:28:06 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
09:28:45 Join Acksaw [0] (
09:34:32 Quit MarcoPolo (Remote closed the connection)
09:40:31 Join powr-toc [0] (
09:40:40 Quit `3nergy (Read error: 104 (Connection reset by peer))
09:45:27 Join Ribs [0] (
09:47:47 Join mikearthur [0] (
09:51:55 Quit TCK ("well, if you say so.")
10:12:26 Join SereRokR [0] (
10:17:16 Join ender` [0] (i=ychat@
10:21:56 Quit SereR0KR (Read error: 110 (Connection timed out))
10:27:21Slasherihahaa, found a solution to do the fast tagcache committing: two-way lookup lists :)
10:27:54 Quit SereRokR ("XChat Aqua")
10:28:21 Join SereRokR [0] (
10:29:09 Join ProgramZeta [0] (
10:38:39 Quit jeremy (Connection timed out)
10:39:09 Join TeaSea [0] (
10:39:38 Join MarcoPolo [0] (
10:39:50amiconnomg. The iriver retail firmware seems to be unoptimised code...
10:40:16JdGordonwell... duh!
10:40:30 Join bluebrother [0] (
10:40:32 Join jeremy [0] (i=katzj@freenode/staff/jeremy)
10:40:53amiconnOf course the codecs can't be unoptimised, but at least the isp1362 driver code is
10:44:39 Join muesli__ [0] (n=muesli_t@
10:53:31 Join PaulJ [0] (
11:01:28 Join lodesi [0] (
11:05:15 Join `3nergy [0] (
11:11:38Slasherilooks good, now the fast committing seems to be actually working
11:12:12 Join jbauman [0] (
11:16:48 Quit MarcoPolo (Remote closed the connection)
11:19:04 Join Paul_The_Nerd [0] (
11:19:29 Join Hansmaulwurf [0] (
11:19:56 Quit Arrogant ("Leaving")
11:20:14 Quit Hansmaulwurf (Client Quit)
11:22:59 Quit `3nergy (Read error: 104 (Connection reset by peer))
11:25:07 Join Hansmaulwurf [0] (
11:25:15 Part Paul_The_Nerd
11:25:21***Saving seen data "./dancer.seen"
11:32:02 Part amiconn
11:33:01 Join speacial_ed [0] (
11:38:27 Join ep0ch [0] (n=ep0ch@
11:41:32ep0chi fancy doing some testing with the Voice files, which English one sounds best?
11:43:31 Join powers [0] (
11:44:35powershi everyone! I was wondering if i can have both rockbox and ipodlinux on my nano..
11:45:15powersu know how? i would really appreciate your help
11:45:56powersor just point me a link
11:46:11JdGordoni dont have an ipod.. but i think u need to format the disk for an ext2 part and fat32 and use ipodloader2 or something
11:47:02JdGordonsomehwere on that page should be able to help u
11:47:16powersu mean i have to make different partition for linux?
11:47:37JdGordoni dont know...
11:47:41powersnevermind, i think i found something
11:47:45JdGordonpaul_the_nerd is the guy to ask..
11:47:47powersthanks :)
11:48:25powersyeah i remember paul
11:48:47powersu know what times he hangs in here?
11:49:28powersmight need some help if i screw up my pod
11:49:46JdGordonhe's in the us iirc, so ye...
11:50:01JdGordonhe is normally online when i am.. left 20min ago
11:50:22powersanyway.. thanx again :
11:50:29 Quit powers ()
11:51:54ep0chlostlogic, linuxstb: fancy updating the SID patch to work with CVS?
12:03:46 Join pondlife [0] (n=pondlife@
12:04:47 Quit stoffel (Remote closed the connection)
12:04:50 Join matsl [0] (
12:07:21 Join obo [0] (
12:08:49 Join lee-qid_ [0] (
12:09:54 Nick lee-qid_ is now known as lee-qid (
12:10:31 Quit speacial_ed ("Bye")
12:12:35 Join JBGood [0] (
12:13:07 Join Kohlrabi [0] (
12:14:11 Join lodesi_ [0] (
12:15:41 Quit lodesi (Read error: 110 (Connection timed out))
12:15:51 Join powers [0] (
12:16:29 Quit powers (Client Quit)
12:18:18Slasherihmm, it seems tagcache is currently allocating ram much more than it needs when loading to ram
12:18:57Slasheriabout 400 kB too much at least
12:21:54ep0chpreglow: do you know why mp1 and mp3 play with no boost (on ihp) whereas mp2 requires 50% boost?
12:22:43 Quit jbauman (Read error: 101 (Network is unreachable))
12:25:53 Quit jeremy (Connection timed out)
12:27:26 Join jeremy [0] (i=katzj@freenode/staff/jeremy)
12:28:48 Quit matsl (Read error: 110 (Connection timed out))
12:29:09 Join matsl [0] (
12:35:32 Join JBGood25 [0] (
12:47:44 Quit Kohlrabi (Nick collision from services.)
12:47:55 Join Kohlriba [0] (
12:52:21 Quit JBGood (Read error: 110 (Connection timed out))
13:01:52Mikachulinuxstb_: there's a silly bug in sudoku, if you open a saved .ss that isn't in a legal state, you can't exit the plugin because you can't open the menu while the board is illegal
13:04:10Mikachulinuxstb_: i also can't figure out why you return PLUGIN_USB_CONNECTED when you quit from the menu
13:08:06preglowep0ch: no
13:08:22preglowso recording hangs with other people as well
13:14:07 Quit jeremy (Connection timed out)
13:15:06 Join jeremy [0] (i=katzj@freenode/staff/jeremy)
13:16:45 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
13:17:44 Join darkless [0] (n=darkless@
13:21:28linuxstb_Mikachu: Thanks for the bug report - I'll have a look. I seem to recall that I returned PLUGIN_USB_CONNECTED to force Rockbox to rescan the disk because the plugin might have created a new file. But it probably shouldn't always return that...
13:21:33 Nick linuxstb_ is now known as linuxstb (
13:21:43 Quit lodesi_ (Read error: 110 (Connection timed out))
13:22:36 Quit obo ("bye")
13:25:22***Saving seen data "./dancer.seen"
13:26:06 Join lodesi_ [0] (
13:29:18amiconnlinuxstb: Another sudoku quirk: If you start sudoku as a plugin, it always starts with generating a new game
13:29:22Mikachulinuxstb: i also don't think the quit menu entry should save, or rename it Save & Quit
13:29:52Mikachulinuxstb: i solved it like this
13:30:02Mikachu- return(true);
13:30:02Mikachu+ return(!check_status(state));
13:30:05Mikachuin load_sudoku
13:30:13amiconnThis is a lenghty operation esp. on archos, and it is annoying if you don't want that, but e.g. want to build yor own game
13:30:17preglowanyone got any idea what might cause recording to just hang?
13:30:55 Quit wehn ()
13:31:45linuxstbMikachu: I'm not sure if that's enough - check_status() doesn't check the whole board, it just checks the part of the board affected by the number at the cursor position.
13:32:05Mikachulinuxstb: hm yeah
13:32:22Mikachuyou could still move the cursor into a space that doesn't let you exit
13:36:10linuxstbamiconn: If the default "" file exists, then that will be loaded when you start Sudoku as a plugin. But as you say, if that doesn't exist it will generate a game randomly. I'm not sure I would want to change that.
13:36:51Mikachuwhat happens if you open a .ss file, generate, and save?
13:36:52 Join webguest15 [0] (
13:37:17linuxstbThe generate function sets the filename to
13:37:32 Join Moos [0] (
13:37:35 Join mborus [0] (
13:37:56mborusHi - just testing spdif in :)
13:38:40 Quit webguest15 (Client Quit)
13:38:55 Join Paul_The_Nerd [0] (
13:39:21amiconnlinuxstb: Generating takes 15..30 seconds on Ondio...
13:40:10 Part Paul_The_Nerd
13:40:34mborusRegarding ideas why recording hangs: Is it possible that the Iriver hangs when no signal is found?
13:40:58linuxstbamiconn: I still think that the majority of users will start Sudoku in order to play a random game, rather than enter one manually.
13:41:38linuxstbAnd I think that should only happen the first time you start the plugin. The next time, you will have a file which will be loaded.
13:41:48preglowmborus: doubt it
13:41:58preglowi haven't done much to the code either
13:42:03amiconnlinuxstb: It happens everytime as long as there is no
13:42:09preglowso it's probably some braindead hardware thing
13:42:29mborusWhen I go into debug - S/PDIF analyser to signal is shown. I tried both 44.1kHz and 48 kHz signals and with the same setup I can record 44.1 using the Iriver firmware.
13:42:38amiconn...which won't exist unless you explicitly save a game with that name
13:42:55preglowmborus: i can record both 44.1khz and 48khz here with no problems
13:43:05linuxstbamiconn: Then something is going wrong - a randomly generated game should be given that name, and quitting should cause it to be saved.
13:43:09preglowi have only tried 16 bit sound, thoigh
13:43:27amiconnlinuxstb: It's definitely not saved on Ondio
13:43:51 Quit Moos (Read error: 104 (Connection reset by peer))
13:44:14 Join mborus_2nd [0] (
13:44:52 Quit mborus ("CGI:IRC (EOF)")
13:45:15 Join Paul_The_Nerd [0] (
13:46:30 Quit matsl (Read error: 110 (Connection timed out))
13:46:34preglowi can only make recording hang after having switched from either mic/line in to digital
13:46:38mborus_2ndpreglow, just to make sure it's not a hardware problem I'm going to do a digital recording on the Iriver firmware right now and try Rockbox again afterwards.
13:46:59linuxstbamiconn: I'm guessing you're quitting using the OFF button, rather than selecting Quit in the menu? The game only seems to be saved when you quit from the menu, and ipods don't have a quit button defined....
13:47:06 Join matsl [0] (
13:47:27amiconnlinuxstb: Yes of course I'm using the Off button. Using the menu would be cumbersome
13:47:47Mikachuso create an empty
13:48:05preglowmborus_2nd: oh, i'm almost certain it's not a hardware problem
13:48:19amiconnImho all plugins should have a quit button. Showing the menu when the intended operation is to quit is counter-intuitive
13:48:37amiconn(like the current behaviour of the jpeg viewer and jewels)
13:48:40preglowand a separate button for quit is also quite a waste
13:49:07 Nick lee-qid is now known as lee-qid||0xf00d (
13:49:46amiconnImho it's not a waste, but rather ensures consistent operation
13:50:06amiconnI like the jewels game, but exiting it has become a button orgy recently...
13:50:07preglowi'm ok with it if there's enough buttons
13:50:17preglowbut how would you fit a quit button in ipod doom?
13:50:39preglowexiting it has become a case of entering the menu and pressing quit
13:50:42preglowhardly a big bother
13:50:48amiconnDoom is different in that it has its own in-game menus, which do have a quit option
13:50:49linuxstbamiconn: I agree with you, but in practice we don't have enough buttons to give everything a consistent purpose.
13:51:03preglowi don't feel entering a menu and quitting is a big bother at all
13:53:05amiconnQuitting jewels used to be as simple as pressing Off. Now it is: Press Off, select 'End Game' from the in-game menu (3x down, right), and select 'Quit' from the outer menu (4x down, right)
13:53:20 Quit bluebrother ("brb ...")
13:53:30amiconnAn operation that used to be 1 keypress extended to 10 keypresses...
13:53:33preglowput quit in the first menu as well
13:54:08preglowbut ok, what suggestion do you have for a quit button on ipod?
13:54:09Paul_The_Nerdsid.c doesn't seem to compile anymore. =/
13:54:25amiconnAn the in game menu isn't a standard rockbox menu; on ipod it doesn't react on scroll-repeats...
13:54:56mborus_2ndBack. Here's the test result just to make sure. Iriver firmware records digitally both of 48 and 44.1 kHz - so my setup works. Rockbox creates a 0-Byte sized
13:55:03Paul_The_Nerd10 pressed does seem like a bit much.
13:55:18Mikachuamiconn: won't it wrap if you press up instead of down?
13:55:19Paul_The_NerdIt seems that "Quit" should be standardized to "Pull up the menu, go up one (looping to the bottom) and press."
13:55:20mborus_2ndfile - shows 8k in the display and the time counts up
13:55:36preglowmborus_2nd: i can record just fine here
13:55:36linuxstbamiconn: What's the normal button to bring up the main menu on the Ondio? Is that free in Jewels?
13:55:40amiconnWell, you can use Up to wrap around, but with the non-standard menus you can't be sure it works
13:55:54preglowmborus_2nd: what format does it send in?
13:55:56amiconnMain menu on Ondio is long Mode
13:55:57preglowmborus_2nd: 16 bits?
13:56:11preglowamiconn: what would you suggest we use as a quit button on ipod?
13:56:14mborus_2ndyes - it should be 16bit stereo 44.1 kHz
13:56:28preglowmborus_2nd: btw, do a proper reboot
13:56:34preglowmborus_2nd: then try again
13:57:25amiconnpreglow: On ipod there isn't a real standard yet, seems to be common to use just Menu if the plugin doesn't have a menu or otherwise needs many buttons. In the latter case Select+Menu is common
13:57:56linuxstbamiconn: I totally agree with you about Jewels - it seems there are enough free buttons (not including the ipod) to keep the standard behaviour for menu and quit.
13:58:52amiconnPaul_The_Nerd: In the 'Main menu' that some game plugins have, the quit button should always quit the game instead of just the menu
13:58:58preglowand in the case of jewels we have to use some weird button combo for quit
13:59:51Paul_The_Nerdamiconn: So, menu once for menu, then again to leave the game?
13:59:53amiconnMy suggestion for better consistency on ipod is that Select+Menu should always act as quit, even if just Menu can also be used for quit
14:00:20Paul_The_NerdI agree with that.
14:00:21linuxstbSounds sensible.
14:00:49Mikachuthe cube is confusing, select+menu changes some parameter but just menu quits it
14:00:59Mikachunot that the cube is the most important plugin
14:01:08Paul_The_NerdI just realized something... The front page shows actual names, but the since 2.5 page shows things like "Lostlogic" and "Kkurbjun"
14:01:24mborus_2ndI'm back. If I reboot, select digital, start a recording the Iriver hangs. I I do a mic recording after rebooting and start a digital recording, it does not hang...
14:01:24amiconnImportance doesn't matter here. If cube is inconsistent with the rest, it needs to be fixed
14:01:32preglowyeah, just like the commit mails and front page as well
14:01:45mborus_2nd.. however the recorded file is filled with silence only.
14:01:52*Paul_The_Nerd was confused when he started scanning for "Brandon"
14:01:53amiconn(and the parameter setting in cube is a bit odd anyway. It could use a similar scheme as bounce)
14:01:53lostlogicamiconn: When pcm is stopped, the delay is lowever long it takes the voice codec to decode ~1s of audio, or the whole clip, whichever is less.
14:02:10linuxstbAlthough that means we need to make sure that no plugin uses the SELECT press event - only the release events.
14:02:16lostlogicamiconn: I did not touch talk.c, and preemption _when pcm is stopped_ can fairly easily be improved, but much harder when pcm is playing.
14:02:43preglowlostlogic: yo, still no idea why the skipping tracks backwards across dirs hang?
14:02:55preglowmborus_2nd: i have no idea, i can't make it hang that easily here
14:03:08amiconnThe preemption doesn't seem to work at all sometimes.
14:03:08Paul_The_Nerdlostlogic: Did you remove "CODEC_DSP_ENABLE" at some point recently? (I just started looking for it.)
14:03:18preglowhe did
14:03:20preglowwhich was good
14:03:33Paul_The_Nerdpreglow: It's still in the SID patch. Should it be safe to just comment out that line?
14:03:41preglowremove it
14:03:44preglowit's not legal anymore
14:04:05amiconnNot only doesn't it preempt the current clip, but it also keeps clips in the queue which should have been dropped completely...
14:04:05preglowcodecs should not be able to decide if they want dsp or not anyway
14:04:07lostlogicpreglow: glad you agree :)
14:04:16amiconn...which is strange, since it works on archos...
14:04:18lostlogicpreglow: no, haven't looked at it yet, was working only on that voice rework
14:04:25mborus_2ndIf the patch I used still current? (
14:04:35mborus_2ndIs... (not if)
14:04:38preglowshould be, i haven't done anything since last night
14:04:40lostlogicamiconn: that's because the clips are already decodec to the pcm buffer then
14:05:21amiconnlostlogic: Then something is wrong with how voice clips are decoded. They need to be kept in the voice queue as long as possible, in order to keep the latency down
14:05:51lostlogicamiconn: it's too hard to calculate the insertion point and stuff, IMO
14:06:16amiconnIirc you told me that the pcm buffer is no longer than 1/4 sec for voice
14:06:48lostlogicamiconn: when playback is stopped, voice uses the 'normal' pcm buffer
14:07:06lostlogicfor mixing, it is inserted starting about 1/4 second back from the music playing point
14:07:08amiconnHmm, why is that?
14:07:25lostlogicamiconn: they are two totally different beasts, playing voice on its own vs. mixing
14:08:02amiconnHmm, but the normal pcm buffer can be adjusted in length?
14:08:13lostlogicbut either way, up-to the whole pcm buffer may be decoded ahead, and I haven't come up with a way to reduce this without introducing a lot of skippiness when mixing
14:08:47 Quit mirak (Read error: 110 (Connection timed out))
14:08:52preglowideally you should have separate decode buffer for voice that you just flush whenever what's in it is invalidated
14:08:56lostlogicfor non-mix voice, it would be almost trivial to flush the buffer for a preempt.
14:09:14preglowmixing should happen as late as possible
14:09:18lostlogicpreglow: we could do that, but there's still the problem of which thread and how far back to mix.
14:09:47lostlogicthe only 'person' who is guaranteed to know when more data needs mxiing is the DMA, and we sure as hella ren't mixing on the DMA interrupt.
14:10:16amiconnHmm, iiuc there could be a secondary pcm buffer for voice. This one is where decoded voice is buffered, and it would offer a flush function
14:10:38lostlogicyes, we could do that.
14:10:49lostlogicI gotta work on other playback bugs for a while though.
14:10:50amiconnThen we could mix real close to the playback point
14:11:10amiconnThe voice pcm buffer could be mono of course, saving ram
14:11:28preglowno worries
14:11:32lostlogichmm, interesting, that would involve DSP and voice codec adjustments.
14:11:34preglowwe'll never hit the date anyway
14:12:03*amiconn needs to work on remote recognition and remote lcd init
14:12:04lostlogicwell I need a break from voice any way, it's not a feature I use, so it was pretty brutal to work on it a lot.
14:12:19preglowi'm still looking for bug reports on the resampler
14:12:24preglowi can't make it clicky anymore
14:12:51lostlogicpreglow: I have bugs that relate to resampling (causing misscalculation of playing points) but none in the resampler itself.
14:13:07preglowi'd be interested with the resampler based ones, yes
14:13:17preglowi don't have time to come up with anything that sounds better
14:13:22preglowso i have to go for bug fixing/optimising now
14:14:05lostlogicpreglow: seems like seeking and a/b repeat with 48khz audio causes the playback point reported not to match at all the real playing point.
14:14:12lostlogicaccording to forums
14:14:15 Quit matsl (Read error: 110 (Connection timed out))
14:14:32lostlogicbelieve so, lemme try and find it.
14:14:59mborus_2ndI can second that 48kHz problem. Have had that all week
14:15:15lostlogicmborus_2nd: did it only start this week?
14:15:25lostlogicmborus_2nd: and is it only mp3 or any 48khz stream?
14:15:45mborus_2ndI'm testing this with 48kHz mp2 streams
14:16:12mborus_2ndthese are radio shows, so they normally run three hours
14:16:17amiconnlostlogic: Hehe, similar here. I don't use the remote for real, but I am working on improving remote detection...
14:16:26 Join ronbo [0] (
14:17:14mborus_2ndI haven't tested AB repeat on those lately, so I can't say when this started. But I noticed two bugs over this week
14:17:33*amiconn wonders whether it would be better to add yet another thread, or make the remote_scroll thread a bit more complex to save an extra thread
14:17:36preglowi never record, and am working on spdif recording
14:17:56 Part ronbo
14:18:05mborus_2ndthe first one was, that bookmarks towards the end of the file didn't play. This now works again
14:18:43mborus_2ndThe second problem is AB repeat. When this failes, afterwards forwarding or rewinding can freeze the Iriver.
14:19:04amiconnlostlogic: Btw, speaking about playback. Now that you fixed the bug that a resume position was never cleared at the end of a playlist, there's a new bug
14:19:27preglowmborus_2nd: so, you always get silent recordings with spdif?
14:19:29amiconnNow the resume position gets sometimes cleared prematurely
14:19:56amiconnI.e. if you stop a playlist, you can't be sure that it's possible to resume it later
14:20:23mborus_2ndpreglow, yes - either crash=0 bytes or silent.
14:20:41preglowi can make it record just fine here
14:20:54amiconnHappened at least 3 times on my mini2g
14:21:29lostlogicamiconn: :( ok
14:21:31 Quit Rob2222_ ()
14:21:32amiconnEverytime *towards* the end of the playlist
14:21:46amiconn(like in the 2nd-last or 3rd-last track)
14:22:21amiconnMaybe it happens if the whole rest of the playlist is already buffered?
14:22:25lostlogicamiconn: ah, I think that makes sense −− my if() might have been on playlist end.
14:22:41lostlogicI prefer bugs that at least make sense :)
14:23:09lostlogicah, the joys of releases, we all get to work on not our favorite things but things that will make the 'product' 'better' :-\
14:23:45preglowthink of the fame!
14:23:53lostlogic::blink blink:: riiiight'
14:23:55crashdand the money and bitchazzzz
14:24:49lostlogicpondlife: wait... something changed with crossfade!? I mean I changed beep... but I don't think I changed crossfade :-\
14:31:13 Join bluebrother [0] (
14:32:15 Join MarcoPolo [0] (
14:34:27Paul_The_NerdY'know, putting builds on your player that should bump the config block but don't can have some funny side effects.
14:34:45Paul_The_NerdFor example, the iPod Nano doesn't have a screen flip, but apparently the option got enabled because all controls were reversed.
14:36:32amiconnPaul_The_Nerd: Someone needs to implement screen flip for the ipod colour lcds
14:36:59amiconnIt should be perfectly possible since it is already implemented for H300, and the lcd controller is the same
14:39:21 Join Lear [0] (
14:41:15lostlogicouch. just turned on my ipod and had it at vol. -11 from using it in car stereo
14:43:53Paul_The_NerdI usually can't take over about -20 or so unless it's one of a few specific songs.
14:46:21mborus_2ndlostlogic: before I leave, I just tested the 48kHz AB problem again with the build I made to try digital recording. Here the AB repeat fails as soon as I set the "B" point. Instead of jumping to "A", the display shows the time for "A", but audio from a point in the file even after "B" is played.
14:49:00 Join twisted` [0] (
14:49:11twisted`sharpe: *poke*
14:49:34lostlogicmborus_2nd: was there a long (> 30 meg) distance between the two points?
14:52:11mborus_2ndNo. - aproximately 10 minutes of 192k audio. Maybe there's a problem in how I set the points: I set point A, then I forward a few minutes and then set point B.
14:56:57mborus_2ndHmm, I'll try this now without forwarding when setting the points.
14:57:03 Nick lee-qid||0xf00d is now known as lee-qid (
14:59:27 Join ashridah [0] (
15:00:31mborus_2ndWhile I wait for point B - is there a reason why A-B points get lost when I turn off the Iriver? Currently I have to re-set every time I start the Iriver...
15:02:03pondlifelostlogic: Well, crossfade with track skipping seems fixed now.
15:02:39pondlifelostlogic: But navigating to another track (maybe the next track) is broken, on my H300.
15:03:35pondlifelostlogic: Am also getting fairly regular crashes that are probably resume related, but haven't come up with a simple recipe yet.
15:04:02*pondlife is using voice+crossfade+beep and really trying to Break Things
15:04:09mborus_2ndPoint B reached and set without forwarding to it. It was only 6 minutes. The details: File length (3:09:57) Point A 2:45:00 Point B 2:51:11. The jump back to A worked. Now let's see what happens, when B is reached.
15:06:14novimondo other get superslow transfer rates with their ipod nano after installing rockbox?
15:06:37Paul_The_Nerdnovimon: Try booting into the Retail OS if you're planning on transferring a lot of files.
15:07:00novimonis there some know problem?
15:07:17Galoisnovimon: I have decent transfer rates, but I seem to be the only one without a problem
15:07:41Paul_The_NerdGalois: I find your 2 second transfer somewhat doubtful, personally
15:08:05Galoiswell I can't prove it to you short of flying you to washington
15:08:32 Quit Lear ("Chatzilla 0.9.72 [Firefox]")
15:08:34*Paul_The_Nerd shrugs
15:08:50mborus_2ndPoint B reached. Jump back to A worked again. Hmm. Could the forwarding in between setting point have caused the problem? I'll reboot my Iriver and try again.
15:18:08novimonGalois, thanks for the tip, transferring is now super fast :)
15:18:20mborus_2ndMore test results: There is a forwarding problem. I have a 15 second minimum forwarding step defined (so I can move quickly within radio shows) and in trying to navigate to the A 2:45:00 point I managed to misplace the audio also. So maybe AB is working OK
15:18:24Galoishey it wasn't my tip
15:18:39 Quit markun (Remote closed the connection)
15:18:47novimonsorry, Paul_The_Nerd :)
15:19:08 Quit HCl (Remote closed the connection)
15:19:38Galoisbut, I'm telling you, I get the same transfer rates in the hardware disk mode and the retail firmware, whether you believe it or not
15:19:55 Join HCl [0] (
15:20:00 Join markun [0] (
15:20:13novimonis "hardware disk mode" something special that you have to enable
15:20:28Galoisyou get it just by plugging in the usb cable while rockbox is running
15:20:42novimonfor me it was ten times slower :)
15:21:21novimoncan you music by folders in itunes?
15:21:37GaloisI have no idea. I don't use itunes.
15:21:45 Join Moos [0] (
15:21:46novimonits hideous :)
15:21:56Galoisitunes is an abomination
15:23:50 Join obo [0] (
15:24:08 Quit mborus_2nd ("CGI:IRC (EOF)")
15:25:23 Quit JdGordon ("Miranda IM! Smaller, Faster, Easier.")
15:25:26***Saving seen data "./dancer.seen"
15:28:24Incanyone have the 5g countour showcase?
15:28:46 Quit ashridah ("Leaving")
15:41:42 Quit pondlife ("Dinner time")
15:41:53 Part Paul_The_Nerd
15:43:47 Quit SereRokR ("XChat Aqua")
15:45:56 Join SereR0KR [0] (
15:47:27 Quit MarcoPolo (Read error: 104 (Connection reset by peer))
15:50:30 Quit twisted` (Remote closed the connection)
15:54:17 Join lodesi [0] (
15:57:32 Join twisted` [0] (
16:00:44preglowamiconn: i can't build gcc 3.4.6 for coldfire even with the patch
16:01:16preglowstupid old -mc68020 option unrecognized thing
16:01:20preglowi can't remember how to fix it :/
16:04:01 Join whatboutbob [0] (
16:07:41 Quit lodesi_ (Read error: 110 (Connection timed out))
16:12:44 Join Rob2222 [0] (
16:14:32whatboutbobpreglow: having fun with spdif? :-)
16:17:14 Quit twisted` (Remote closed the connection)
16:18:23Hansmaulwurfrecording with optical in works?
16:19:45whatboutbobhansmaulwurf: not yet.
16:23:11whatboutbob...but preglow's giving it a shot.. :-)
16:24:06whatboutbob*fingers and toes crossed*
16:25:25 Join stoffel [0] (
16:25:42whatboutbobThis evening I managed to record the first song of one of the best concerts I've been to for aaaages, before the iriver crapped out on me with a hdd full message when there was 18gig free.
16:26:14whatboutbobTo add salt to the wound the recording of the one song sounds amazing.
16:26:37Hansmaulwurfat the moment my 20GB disk have only 512mb free
16:26:38Hansmaulwurfsucks :(
16:26:53Hansmaulwurfneed a bigger disk or a new "rockbox ready" player
16:28:01whatboutbobi'd made a point of formatting the drive today so i'd have plenty of room free. Got home and read about the issue w/ the fake full hdd msg. doh!
16:28:25preglowwhatboutbob: well, the problem is that spdif recording does work
16:28:40 Quit stoffel (Client Quit)
16:28:42whatboutbobpreglow: what sort of rig are you recording with?
16:29:05preglowtesting with standard spdif from the cd player (44.1khz) and pc (48kgz)
16:29:22whatboutbobboth 16bit?
16:29:44 Quit dj-fu (Read error: 110 (Connection timed out))
16:29:49 Join twisted` [0] (
16:29:53twisted`whoa, my iPod fixed itself
16:30:40whatboutbobtwisted: Did it give itself a pat on the back when it was done?
16:30:53preglowwhatboutbob: yes
16:30:59twisted`whatboutbob: I did that for him
16:30:59preglowbut i don't think the bit rate should matter
16:31:01twisted`it got no arms :(
16:31:30whatboutbob...and it fixed itself?! Its a miracle!
16:31:39preglowbut no, there's something else wrong here too
16:31:43preglowbut i can record just fine
16:31:46preglowuntil it hangs, heh
16:32:10whatboutbobpreglow: My preamp outputs a funny sample-rate...
16:32:55twisted`whatboutbob: LOL, no I formatted it's partition to ext2 to backup my files... but I copied it back... so I formatted it back to fat32 and it reinstated itself...
16:33:33whatboutbobIt says it outputs 20bits...but then it does some weird-arse sampling down...not sure if that would cause the issues i'm seeing, but thought it may be worth mentioning.
16:34:50 Join blindx [0] (
16:35:22blindxhaving some difficulties browsing through stuff with rockbox. I'm viewing an image, how do I get back to the filelist?
16:36:55linuxstbWhat player do you have?
16:38:03blindxiPod color.
16:38:16linuxstbHold select+menu
16:38:23blindxi have to reset it?
16:38:29blindxoh ok
16:38:36blindxoh cool. thanks.
16:39:29 Quit powr-toc (Remote closed the connection)
16:40:18blindxis there a list of commands somewhere?
16:43:28 Quit jeremy (Success)
16:43:34 Join mirak [0] (
16:43:46 Quit lee-qid ("Trillian (")
16:46:25 Join new5guser [0] (
16:47:58preglowwhatboutbob: what sample rate?
16:48:13preglowthe spdif debug screen should be able to tell you
16:52:29 Join stoffel [0] (
16:52:33whatboutbobpreglow: s/pdif analyzer?
16:52:40preglowwhatboutbob: in the debug menu
16:53:08whatboutboball 0's.
16:53:25whatboutboboh...wait...sorry...wrong build.
16:53:34whatboutboball throw yours back on now.
16:53:47preglowthe spdig analyzer should work anyway
16:53:50preglowit's been there for a long time
16:54:09whatboutboboh...well...its giving me all 0's.
16:54:31preglowiwhat about the sample rate?
16:54:33preglowat the bottom
16:54:53whatboutbobclock accuracy: 0
16:55:08preglowabove that
16:55:33twisted`ok this is, really bad
16:55:39whatboutbobfrequency: 0 (44.1kHz)
16:55:40twisted`I can't restore my ipod :|
16:55:51twisted`windows doesn't reconise it
16:55:55preglowtwisted`: reformat it
16:56:09twisted`preglow: it's already fat32
16:56:23preglowyeah, but windows obviously can't see the drive
16:56:32preglowsometimes before you can restore it, you need to just clean reformat it
16:56:35twisted`preglow: when I plug it in Windows goes like, cannot reconise...
16:56:42twisted`format everything or...
16:56:45twisted`only... that one partition
16:57:03preglowjust try until something works, i have no idea
16:57:10preglowsome people need to repartition, even
16:57:24twisted`hmm doesn't the first partition hold the original firmware etc?
16:57:58preglowwhatboutbob: not exactly what i'd call a weird sample rate
16:59:46crashdtwisted`: what you want to do is kill all the partitios (either using windows/linux or the HP reformatting tool that's linked on the ipl wiki)
17:00:02crashdand then, put the ipod into disk mode, and use the apple updater to reset it to factory defaults
17:00:35twisted`crashd: how can I put it into disk mode? doesn't killing all partitions kill the firmware?
17:01:14crashdwhen you're booting, hold down menu and select to reboot, then immediatly hold select+play
17:01:30crashdwell, it removes the retailos version you have, but the disk mode is still there in flash
17:01:36twisted`crashd: ya original firmware?
17:01:38crashdyou'd have to pretty clever to actually delete that
17:01:52crashdtwisted`: it doesnt matter, the apple updater will restore that
17:02:49twisted`crashd: hmm... ok me searchin for the setting where I can set it to diskmode
17:02:55crashddude, i just told you
17:03:01crashd:01 < crashd> when you're booting, hold down menu and select to reboot, then
17:03:01crashd immediatly hold select+play
17:03:08 Join jeremy [0] (i=katzj@freenode/staff/jeremy)
17:03:12crashdsigh. now look what you made me do
17:03:24preglowyou can't fuck up your ipod
17:03:28whatboutbob:preglow: *slaps forehead* i meant resolution...(tis 1am here and i got 2hrs sleep last night)
17:03:29preglowit's damned close to impossible
17:03:39preglowwhatboutbob: i don't think resolution should matter at all
17:03:54preglowtwisted`: you can erase the entire bloody disk with zeroes, and you'll still be fine
17:04:09preglowdisk contents doesn't matter
17:04:17twisted`I pressed that and now I see a big Check and it says OK to disconnect
17:04:27crashdhave you plugged it into something, ie: usb?
17:04:33twisted`I have it in my hands
17:04:36twisted`it says Diskmode on top tho
17:04:37crashdwell, let's try that eh? they dont have wireless connectivity
17:04:49crashdright, that's diskmode, so now connect it to something and try using the apple updater
17:04:52whatboutbobpreglow: can you think of anything else you want me to try before i hit the hay?
17:05:14twisted`crashd: ah ok, let's kick my lil brother off his windows pc again
17:05:17twisted`he'll hate me
17:09:44 Join actionshrimp [0] (
17:09:57whatboutbobwell, g'night then folks.
17:10:38whatboutbobgood luck preglow. :-) i'll be checking the logs if you need anything.
17:11:29 Part whatboutbob
17:21:05twisted`crashd: hmm, tried another usb port and it suddenly reconised it... it's now formatting I think... it's taking _long_
17:24:02amiconnpreglow: If you mean the multilibs problem (but that fails for 68040), see
17:24:37amiconnNote that I didn't try the m68k build patch for amd64 yet
17:25:27***Saving seen data "./dancer.seen"
17:29:02preglowamiconn: this is just my m68k-elf-as not recognizing the option -mc68020
17:29:10preglowit's the newest version and all, so don't ask why
17:29:35amiconnThat's strange
17:29:56amiconnYou mention it being an old problem, but I actually never heard it before...
17:30:10preglowi might be confusing it with the multilib thing
17:30:35 Join solexx_ [0] (
17:31:09*amiconn should try to build an m68k-elf-gcc ...
17:32:30 Join RotAtoR [0] (n=e@rockbox/developer/rotator)
17:34:16preglowamiconn: please do
17:34:33preglowthe patch is just commenting out the trigger if line, so...
17:34:40amiconnI know.
17:34:57amiconnI think about adding some more code there, that outputs the exact values
17:39:26Bjoern-ErikHi, I'm wondering, does Rockbox support ID3v2.4-tags (encoded in UTF-8 charset)?
17:41:04 Join petur [0] (n=petur@rockbox/developer/petur)
17:41:14amiconnpreglow: Build started...
17:41:43bluebrotherfile `which latex`
17:41:43bluebrother/usr/bin/latex: symbolic link to `pdfetex'
17:41:52amiconnssh is sometimes useful, 'cause there's no rdp for linux...
17:42:13crashdremote x over ssh
17:42:17crashdget with it amiconn ;)
17:42:27amiconncrashd: Ahahahaha
17:42:38crashdand so ends my comedy japes for the day
17:42:42amiconnx over ssh isn't even remotely comparable with rdp
17:42:47crashdno, it isnt
17:42:58crashdand linux's desktop model and windows arent remotely comparable either
17:43:01amiconnX over ssh is sluggish even over lan
17:43:40amiconnMaybe NX. I still want to try it, but I need packages for debian-amd64...
17:43:59bluebrotherKDE has something rdp-like.
17:44:15preglowwhat is rdp??
17:44:27bluebrotherremote desktop protocol iirc.
17:44:32crashdremote desktop protocall
17:44:34amiconnMicrosoft remote desktop protocol
17:44:34preglowwhat about vnc?
17:44:42crashdmicrosoft's 'vnc'-a-like stolen from citrix
17:44:49crashdwell, stolen, they paid for it, but still
17:44:50amiconnrdp is significantly faster than vnc
17:45:15preglowdoesn't x support functionality like that out of the box anyway?
17:45:52bluebrotheryou mean xdmcp?
17:46:35bluebrotheryes, if the login manager is configured to allow it.
17:47:01bluebrotherI used it on LAN connections and it was like sitting in front of the remote machine.
17:47:06*amiconn now has a shiny new m68k-elf-gcc for amd64 :-)
17:48:17 Join piroko [0] (
17:48:52RotAtoRi see there was an earlier discussion about exiting plugins, specifically jewels
17:48:58RotAtoRso is it the general consensus that there should be a standard exit button to unconditionally exit the plugin?
17:49:09amiconnIf possible, yes
17:49:27*BHSPitLappy thinks that there should be a global failsafe in there, like a "ctrl+alt+delete" if you will
17:49:28RotAtoRok, it's just a quck change then
17:49:41amiconnPlugins which have some kind of main menu should drop to that main menu with the first exit button press
17:49:58RotAtoRsure, sounds good
17:50:12RotAtoRthen i'll change it back to the two button presses to exit
17:50:17 Quit solexx (Read error: 110 (Connection timed out))
17:52:05preglowamiconn: you didn't get the problem?
17:52:09amiconnOk, this gcc builds all coldfire targets fine
17:52:30amiconnpreglow: No problems, apart from a ton of warnings, but that seems to be normal when building gcc
17:53:05preglowjust target=m68k-elf ?
17:53:06preglownothing more?
17:54:25kapheineamiconn: are you using 4.0.3?
17:55:00amiconnkapheine: Native gcc is 4.0.3, and I was building our recommended m68k-elf-gcc (3.4.6)
17:55:28kapheineAhh okay, and you made the change to combine.c I assume?
17:55:43amiconnI just applied your patch
17:56:03 Quit mirak (Read error: 104 (Connection reset by peer))
17:56:20kapheineOkay.. if I get a chance I'm going to look at the function that was causing the assertion to fail and see if I can whip up some code to trigger it.. I want to see if I can break it or not :P
17:56:27amiconnpreglow: configure −−target=m68k-elf −−prefix=/opt/m68k for building binutils,
17:56:49amiconnand configure −−target=m68k-elf −−prefix=/opt/m68k −−enable-laguages=c for building gcc
17:57:27amiconnOf course with adding /opt/m68k/bin to the path inbetween
17:57:46preglowexactly what i did
17:57:47 Join mirak [0] (
17:58:47amiconnThis is debian testing (with some remaining packages from unstable), pretty current (updated a few days ago)
17:59:10preglowstill running ubuntu breezy
17:59:25amiconnUbuntu didn't work for me at all
17:59:34amiconnHung while booting the installer...
17:59:43crashdamiconn: any special hardware?
18:00:19amiconnThe culprit is that ubuntu still uses kernel 2.6.12, and that doesn't cope with certain bios'es throwing apic errors
18:00:54amiconnHad the same problem with debian testing back then; trying unstable (kernel 2.6.15) fixed the problem
18:01:18amiconndmesg shows many lines like:
18:01:20amiconnAPIC error on CPU0: 40(40)
18:01:49preglowno problems like that here
18:01:59amiconnNo really special hardware, it's a HP DX5150 business microtower
18:02:36preglowcan't remember what i used to build this box anymore
18:02:49preglowi remember it's got 1 gig of ram and a athlon64 3200+, that's that :)
18:02:59crashdhehe, im on a similar setup preglow
18:03:09 Part ep0ch ("Kopete 0.11.1 :")
18:03:10crashdtasty athlon64s
18:03:17preglowtasty indeed
18:03:41amiconnAthlon64 3800+, ATI XS200 chipset (integrated graphics), 1GB RAM, 2x 160GB SATA HDs, Broadcom GBit LAN
18:03:42kapheineMy processor tasks like salt water taffy
18:04:24BHSPitLappyI wish I understood the reasoning behind AMD's processor naming
18:04:33BHSPitLappy(the numeric aspect)
18:04:41preglowthe reason was intel playing mhz whores a while ago
18:04:47humulusamiconn: just a question, does the installer recognizes the NIC?
18:04:59preglowso amd felt they needed a figure to make the different processor revs comparable from the names alone
18:05:01 Join `3nergy [0] (
18:05:08amiconnI've installed using the businesscard installer
18:05:14humulusamiconn: tg3 driver? do you know that?
18:05:20amiconnyes, tg3
18:05:26humulusbusinesscard installer?
18:05:29humuluswhats that
18:05:56amiconnIt's the mini installer (~40MB) that fits on a businesscard-sized CD-R
18:06:16amiconnThe standard netinstaller is ~120MB, and already contains some basic packages
18:06:28preglowså the mini installers is almost a pure netinstall?
18:06:32humulusamiconn: which distro? or is it shipped with the hardware
18:06:43amiconndebian testing
18:06:59amiconnHP ships a 32bit distro which I didn't want
18:07:05amiconn(SuSE 9.3)
18:07:20humulusamiconn: well sarge netinst. does not recognize the NIC
18:07:34 Quit _Lucretia_ ("Leaving")
18:07:45amiconnSarge is stable?
18:08:00humulusamiconn: yes
18:08:01amiconnNo wonder it doesn't recognise anything fairly recent
18:08:46amiconnbusinesscard CD image (20-50 MB) -> AMD64
18:09:03humuluswell we want stable
18:09:26humuluscuz my boss says so ;P
18:09:37amiconnah, it's the boss :/
18:10:09preglowugh, stable :/
18:10:13amiconnMy experience is that 'testing' is really stable, and 'stable' is just old, and doesn't support anything recent
18:10:47amiconnThe only annoyance is the occasional dependency breakage
18:11:00*amiconn still cannot install grip :(
18:11:38humulusamiconn: we need red hat 9.0 with a 2.4 kernel, which does not recognize the disk controller and the nic either
18:11:49amiconnugh, 2.4
18:18:09 Quit Strath (Read error: 104 (Connection reset by peer))
18:18:17 Join Strath [0] (
18:23:41 Quit SereR0KR ("XChat Aqua")
18:25:20 Join linuxstb_ [0] (
18:25:28amiconnIs there anyone who has a H1x0 with a H300 remote? (both LCD and non-LCD wanted)
18:25:53 Quit linuxstb (Read error: 104 (Connection reset by peer))
18:26:25amiconnpetur: I've checked the disassembly a bit. Didn't find a reset so far, but I found that the 1362 is connected to GPIO5 with one of its interrupt lines
18:26:44amiconnRetailos sets up an isr as part of the init
18:27:20peturyou obviously know assembly better than I do ;)
18:27:39amiconnI wonder what's going wrong with your driver. It looks like you're doing almost the same as retailos
18:27:57amiconnWhat I didn't find yet is the CS3 setup.
18:28:07amiconnSearching in ida is dead slow
18:28:11peturin the disassembly?
18:28:44peturI found it some time ago and took over the register values so that shouldn't be the problem
18:29:39amiconnAh, it's at the very beginning of the rom...
18:29:55peturbut yeah, searching is very slow
18:29:55amiconnThe register values are generous
18:30:23amiconnCSMR3 = 0x1 slould be enough (smallest possible window, 64K)
18:30:42amiconnWe're only using 4 words after all...
18:30:42peturI just took everything over to be sure
18:31:00amiconnThe waitstates are also generous
18:31:06 Join _Lucretia_ [0] (
18:31:17peturonce it works we can tweak
18:31:24amiconnyes of course
18:31:30peturfirst make it work, than optimize :)
18:31:44peturthen even
18:31:46crashdpetur: what you working on?
18:31:58peturnothing right now ;)
18:32:01crashd: )
18:32:16peturisp1362 (usbotg) for h300
18:32:57 Join hardeep [0] (i=hardeeps@SDF.LONESTAR.ORG)
18:32:58peturam processing a concert i recorded yesterday - I can't live without agc anymore
18:33:15 Join filoktetes [0] (
18:34:25amiconnpetur: The ISR setup routine is located at 31010B50
18:34:26peturi start recording with +/- safe levels, put everything in a small bag and drop it somewhere. the agc reduces gain in case my levels weren't so safe
18:34:54amiconnThe debug message texts are sometimes helpful
18:35:29petursome even have typos in them :)
18:35:52amiconnyup, saw that
18:36:11 Join SereR0KR [0] (
18:37:11 Quit goffa (Read error: 110 (Connection timed out))
18:37:41peturregarding the 21 char limit for stored filenames: shouldn't we allow the user to edit only the filename (no path, no extension) - that way it's easier to pass a nice limit to the keyboard routine.
18:38:05peturdisadvantage: can't save to another folder
18:38:19amiconnHmm, right
18:38:29peturright now it's impossible to enforce the 21 char limit :(
18:47:04petursaving to another folder causes issues anyway: the name gets stored but not loaded because we only look for it in the default location
18:47:34amiconnYes, but that's actually a featire
18:47:49peturit will confuse the average user when he makes an error and saves the file somewhere else
18:48:09amiconnFor other similar files (.fnt, .wps), if they're stored in their respective directory, the name gets stored in the config, so they're permanent
18:49:04amiconnHmm, but then you can't save .fnt ans wps files from within rockbox
18:49:14amiconnSo .fmr files are a little different
18:49:27peturI think fmr files are theeonly ones where it matters
18:49:33 Join OnkelJonas [0] (
18:49:43petur.cfg files aren't stored in the config
18:49:46 Quit lodesi (Read error: 110 (Connection timed out))
18:49:57amiconnpetur: No, as they set the config
18:50:34 Join OnkelJonas2 [0] (
18:50:39 Join lodesi [0] (
18:50:40peturwhat I meant was: it doesn't really matter where you save them, nothing is depending on a certain location
18:51:03peturif the fmr file is saved elsewhere, it won't be loaded automatically next time
18:51:15amiconnI prefer my partial .cfg files in the root
18:51:39amiconnBg3r: ??
18:53:06peturI have no problem with the default location, but I rather have the default location a tree next to RB - not under .rockbox - so whiping the tree doesn't wipe your settings,... (granted,you do that only once, curse and remember for ever)
18:53:54amiconnI never wipe .rockbox completely
18:54:04peturI did once :/
18:54:08 Join Massa [0] (n=Massa@
18:54:15amiconnFrom time to time I go thorugh the dirs and remove clutter
18:57:20 Part OnkelJonas2
18:58:09peturright, I still don't know how to solve this. checking the file length after the user entered it and send him back to the keyboard if it's too long?
18:59:06 Quit OnkelJonas (Read error: 60 (Operation timed out))
19:00:11MassaHi everybody!
19:00:21crashdHi Dr. Massa!
19:00:55MassaIs somebody here with low level knowledge about iRiver H300, especially about reading pcf50606 addresses?
19:01:45Massacrashd: Hi, but why are you calling me a doctor?
19:02:01crashdhehe, the way you said it reminded me of Dr. Nick in the simpsons
19:02:47MassaHuh? How are you able to read how I say things ;) ?
19:03:28zeMassa: it was just the 'everybody!' part
19:04:28MassaSorry, I don't know much about the Simpsons - especially not in English :D
19:05:17zeeverybody in the US is so obsessed with it that they'd elect homer simpson as president if they could
19:05:24zeg.w.bush is the closest they could get though
19:05:41MassaROTFL :D :D :D
19:12:10 Quit twisted` (Remote closed the connection)
19:16:00 Join Shadowarrior13 [0] (
19:25:30***Saving seen data "./dancer.seen"
19:29:21 Join linuxstb__ [0] (
19:44:17 Quit linuxstb_ (Read error: 110 (Connection timed out))
19:51:06 Quit lodesi (Read error: 110 (Connection timed out))
19:52:57 Quit new5guser ("CGI:IRC (Ping timeout)")
19:54:59 Join [TCK] [0] (
19:55:07 Join lodesi [0] (
19:59:41 Quit vmx (Read error: 110 (Connection timed out))
19:59:47 Join vmx_ [0] (
20:02:18 Join Tobu [0] (
20:03:34 Quit mirak (Read error: 104 (Connection reset by peer))
20:06:16 Join DrMoos [0] (
20:08:02 Quit Moos (Read error: 104 (Connection reset by peer))
20:08:23 Join bagawk_ [0] (
20:10:02 Nick DrMoos is now known as Moos (
20:12:44 Join Bagder_ [0] (
20:13:43 Join Lear [0] (
20:14:03 Join bagawk__ [0] (
20:15:05 Quit Bagder (Read error: 110 (Connection timed out))
20:15:21 Quit bagawk (Nick collision from services.)
20:15:26preglowsuddenly the label on my h120 disk has vanished
20:15:31 Nick bagawk__ is now known as bagawk (
20:15:34 Quit piroko ("telinit 0")
20:16:41preglowany bugs with deleting in rockbox? it's a strange coincidence that i just deleted my recordings/ dir, then this happens
20:17:19 Join mirak [0] (
20:19:47Tobupreglow: This happened to me also, I'll see if there's a bug
20:19:57 Quit [TCK] (Read error: 110 (Connection timed out))
20:22:24 Quit lodesi (Read error: 110 (Connection timed out))
20:24:51 Join lodesi [0] (
20:26:37preglownot only that
20:26:41preglowthe directory wasn't correctly deleted
20:26:54 Quit bagawk_ (Read error: 110 (Connection timed out))
20:27:20TobuHere is the bug I just added, but I don't have many useful details:
20:46:03 Join Massa_ [0] (n=Massa@
20:49:21 Join macdonalder [0] (
20:53:19 Quit filoktetes ("Leaving")
20:53:49*preglow summons coldfire audio expert
20:53:49 Part amiconn
20:54:22Massa_preglow, it seems no ghost is coming :D
20:57:00 Quit Inc (Remote closed the connection)
20:59:21 Join tvelocity [0] (
21:01:53*petur certainly isn't a coldfire audio expert
21:03:45*Massa_ search for an expert in USB charging ;)
21:03:55preglowpetur: it seems that just writing to DATAINCONTROL sometimes hangs audio
21:04:30preglowi don't know how either, recording peak meter just stops, and the unit just stops responding, even though backlight still turns on at keypresses and so on
21:05:29peturno clue, got to go but i'll be back in 30 mins or so
21:05:44*lostlogic rewrites about 40% of crossfade
21:05:55lostlogicit compiles, but I don't feel like testing it yet :-D
21:06:07preglowsweet lord! antonius hellman just rewrote that bloody imdct36 mess of mine
21:06:37 Join damaki__ [0] (
21:08:58 Join goffa [0] (n=goffa@
21:10:34 Quit _Lucretia_ (Connection timed out)
21:11:10 Quit linuxstb__ (Read error: 104 (Connection reset by peer))
21:11:16 Join linuxstb__ [0] (
21:11:55 Join _Lucretia_ [0] (
21:12:10Massa_does anybody know, if some target currently enables USB charging when connected (USB disk mode)?
21:12:10 Quit _Lucretia_ (Connection reset by peer)
21:12:41 Join _Lucretia_ [0] (
21:15:52 Join lodesi_ [0] (
21:17:00 Quit lodesi (Read error: 110 (Connection timed out))
21:17:18midkayMassa, the FM recorder..
21:17:24midkayand v2..
21:18:33Massa_midkay: do they automatically and always charge when connected with USB? Or is there a specific register/memory address which has to be set for it?
21:19:18Massa_midkay: and what happens when these devices will be connected to an USB hub without power?
21:21:51midkayMassa, i think the charging is hardware. i don't know too much about it..
21:22:04 Quit damaki_ (Read error: 110 (Connection timed out))
21:23:23 Quit macdonalder ("bye bye ;D")
21:23:23 Quit ProgramZeta (Read error: 104 (Connection reset by peer))
21:23:35 Join macdonalder [0] (
21:23:36 Join ProgramZeta [0] (
21:25:31***Saving seen data "./dancer.seen"
21:28:03Massa_midkay: Hmm - I currently try to create an option for enabling and disabling USB charging - currently only for the iRiver H300; I don't know for which other target it is also possible...
21:28:18midkayMassa, why?
21:28:25 Quit quobl_ (Remote closed the connection)
21:28:25 Join damaki_ [0] (
21:29:13Massa_there is an existing patch (made by pyro, who seems no longer be around) which always enables USB charging for the H300
21:30:34Massa_I had a discussion with Linus why this patch doesn't found it's way into CVS and he's the oppinion, that it may cause problem with hubs and USB devices which do not have enough power
21:30:52Massa_And so an option to switch it on or off should be made - voila ;)
21:36:43TobuThe X5 does it within the firmware, and has a setting for it too - either PC (charge) or hub (no charge)
21:37:05Tobubut the setting can't be changed from rb :(
21:37:39Massa_Yeah - someone has to find out how to enable/disable it :D
21:38:35Massa_I'm nearly sure the X5 does currently not charge when connected, or am I wrong?
21:39:04TobuI think it does, but it may require rockbox to be off
21:39:36 Quit lodesi_ (Read error: 110 (Connection timed out))
21:39:44Massa_does the X5 make USB connections in rockbox?
21:39:50 Quit Hansmaulwurf ("( :: NoNameScript 3.81 :: )")
21:40:20 Join lodesi_ [0] (
21:40:25Massa_Or are these handled by the original firmware?
21:42:24*Tobu was checking
21:42:42Tobuit does, but it's the original firmware who does the real work
21:43:33Toburockbox just show a status screen
21:43:46 Quit damaki__ (Read error: 110 (Connection timed out))
21:44:38Massa_What do you mean? Is that a special "charging" screen? Or does it charge while normally connected (and accessible as disk device)?
21:45:12TobuIt's a screen with the USB logo
21:45:25 Join infamis [0] (
21:45:37TobuIt isn't charging currently, but I think I must change the setting in the original firmware
21:45:53infamismiika: tagtree.c:437: error: `LANG_TAGNAVI_ALL_TRACKS'
21:46:06*macdonalder listens to Taking Back Sunday - Twenty-twenty surgery
21:46:39TobuIt wd be nice if rockbox cd refuse usb connections though, because it sometimes detects spurious ones when I'm charging :(
21:46:41Slasheriinfamis: hmm?
21:46:58Tobu(with both firmwares)
21:46:58 Quit Rob2222 (Read error: 104 (Connection reset by peer))
21:46:59infamismight the lang file been updated?
21:47:18 Join Rob2222 [0] (
21:47:28Slasheriinfamis: make sure you have the most recent sources
21:47:34Slasheriincluding updated tools/genlang
21:47:54Massa_Tobu: I don't know much about the X5 port - for the H300 it currently does either charge or connect as disk, not both together (charge during disk connected mode)
21:48:58preglowhas anyone at all made recording hang?
21:49:19TobuDoes it keep the low-level firmware too, or is the X5 a special case?
21:49:30infamisSlasheri: just did a cvs update; nothing changed/added
21:49:53infamislemme see somethin
21:50:20Massa_Tobu: for H300 it boots either rb or the original firmware; rb does not use anything from the original firmware
21:50:41infamisyeah tagtree is the only occurrence of that LANG_ def...where is it really?
21:50:49Massa_Tobu: and I don't think it's different for X5 :-|
21:50:51infamisin that genlang?
21:51:47Massa_Tobu: or will you say RB does call functions from original firmware? I can't believe that...
21:53:06 Join Kyomi [0] (
21:53:25Slasheriinfamis: try to remove the tools dir and checkout from cvs again
21:53:27KyomiWould any of you know what would cause a usb connection to reset in the middle of transferring a file?
21:53:46Slasheriinfamis: it should be in apps/lang/english.lang
21:53:57KyomiI've tried it with rockbox and original firmware
21:53:57TobuI think it doesn't, finally.
21:54:33TobuBut both stay installed at the same time, and it's the original firmware which boots when stg is plugged
21:55:02infamisSlasheri: ok, lemme see
21:56:00preglowanyone present now that can test spdif?
21:57:19KyomiWhat is that?
21:57:37Massa_preglow, if you add an spdif input/output to my H300 I'll test it :-p
21:58:13KyomiI think my battery is dying
21:58:42KyomiIt can barely keep a charge for like 2-3 hours
21:59:07KyomiAnyone know where I could get a new one? Preferrably one that would last longer
22:00:52 Join TCK [0] (
22:00:57preglowthe iriver firmware doesn't handle the spdif monitoring properly either
22:01:43infamisSlasheri: I removed the tools & lang directories & did a cvs update but the new files aren't getting added...what am I missing: cvs -z3 update
22:01:57macdonalderopps... sry about the announce ^^ didn't know a particular switch was on :s
22:02:07infamiswhile in my c:\cygwin\usr\src\rockbox-devel dir
22:04:56infamison another note search-in-files never works in explorer even though I disabled the indexing service
22:09:09preglowtesters, please :>
22:09:47preglowanother bug
22:10:12 Join akaidiota [0] (
22:10:39lostlogichardeep: <3!
22:16:21infamisSlaheri: ignore me...a veryclean fixed the 'problem'
22:16:37infamisSlasheri: ^
22:17:36infamis...but on another note....anyone know how to get debug working with the x5? undef ref to debug_init on ld rockbox.elf
22:20:43 Join webguest60 [0] (
22:21:30 Quit webguest60 (Client Quit)
22:22:57 Quit infamis ("slow afternoon I guess")
22:25:56 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
22:30:17amiconnpreglow: So what's the benefit of the imdct36 rewrite?
22:30:35amiconnI presume it's faster, but how much? and for which platforms?
22:30:51 Quit lodesi_ (Read error: 110 (Connection timed out))
22:32:14preglowamiconn: don't know anymore than what it says on the patch page
22:35:03Learamiconn: imdct36 is for coldfire, since it is a re-write of the asm code.
22:40:44amiconn...and yet another +400 bytes on archos :(
22:43:04lostlogicmmmm, infinite looping goodness
22:43:16lostlogichow would it change archos if it's an imdct!?
22:43:25KyomiAnyone know what would cause the usb connection between the pc and my h320 to suddenly... reset itself while transfering files?
22:44:00preglowamiconn: ehh?
22:44:14 Join ender1 [0] (i=ychat@
22:44:42amiconnpreglow: Today's changes... most of it is probably the tagcache change
22:45:01amiconnLet's see if it has at least some effect...
22:45:11preglowyes, changes do affect binary size, no surprise there
22:45:40amiconnAffect, yes.
22:45:51amiconnBut there are 2 possible directions: up and down
22:45:58preglowboth will happen
22:46:13preglowdevs can't go around checking every single target and tweak and tweak every commit
22:46:16preglownothing would get done
22:46:48amiconnWe pretty much need to tweak for archos
22:47:11*amiconn sighs
22:47:17 Quit ender` (Nick collision from services.)
22:47:23preglowyes, we do
22:47:23 Nick ender1 is now known as ender` (i=ychat@
22:48:11amiconnThat doesn't necessarily mean to drop features. It means that we should code efficiently. This will help all targets, not only archos
22:48:37preglowi don't think anyone tries to code inefficiently
22:48:40amiconnOf course we should also look for additional methods. I hope to get -Os builds working
22:48:47preglowif you see a way to do something more efficiently, fix it
22:48:59preglowdidn't you do a -Os related commit a while ago?
22:49:20amiconnYes, but that was only a change to get it compiling with 3.4 and -Os
22:49:21hardeeplostlogic: here now
22:49:44amiconnThe build failed before with 'pcrel too far'
22:50:11lostlogichardeep: didn't need anything, just saying that I appreciate you fixing some of the playback related bugs I've introduced :)
22:50:17amiconnIt's a first step... I hope that there aren't too many gcc bugs to work around
22:50:42preglowamiconn: i'm sure you'll see a surprise or two...
22:51:22hardeeplostlogic: ah, np. =) can't reproduce this resume issue that the people on the boards are reporting though
22:51:24amiconnInterestingly, our recommendation of gcc 3.3.x for SH1 comes from the fact that 3.3.x produces smaller binaries than 3.4.x when using -O (or -O2)
22:51:41amiconnBut with -Os, 3.4.x is better than 3.3.x ...
22:52:13lostlogichardeep: yeah, saw the posts, I haven't looked at it yet, rewrote crossfade the same way I redid the voice mixing in pcmbuf.c, but apparently made a boo-boo, because it just data aborted on me.
22:52:21amiconnWe would save 11KB (on v1 recorder) with 3.4.6 and -Os compared to 3.3.6 and -O
22:52:35amiconnThat's 5% of the binary size...
22:52:44ender`what about 3.3.6 with -Os ?
22:52:53amiconnender`: [22:51:42] <amiconn> But with -Os, 3.4.x is better than 3.3.x ...
22:53:22amiconnNot much though, about 1KB
22:53:56amiconnWhat's also interesting is that the more compact code generated with -Os is less compressible than the code generated with -O
22:54:05lostlogicbagh, arm only loading by longs is ... odd.
22:54:18preglowyou mean alignment?
22:54:38ender`amiconn: what about the speed (cpu usage) difference?
22:54:44amiconn(using our ucl compressor), so the compressed .ajz for v1 recorder ends up a bit bigger even though the uncompressed binary is smaller
22:55:16amiconnender`: Didn't test yet. Last time I tried -Os and -O2 binaries, they crashed at boot...
22:55:32amiconnWell, -O2 with 3.3.x worked with one small fix
22:55:38preglowamiconn: but yeah, there's no reason not to use -Os, no?
22:55:53amiconnpreglow: Currently there is one: binaries crashing...
22:55:59preglowassuming that's fixed, yes
22:56:27amiconnAssuming I can fix that, there's no real reason against using -Os
22:56:31preglowbtw, did you do any more eq related cpu usage testing?
22:56:41lostlogicpreglow: I'm indexing a short pointer, so it seems to elft shift my index by 1 and then use it to load a long from the array
22:57:08amiconnpreglow: No. What tests would you think should be done?
22:57:11preglowarm can load both long, short and byte
22:57:25lostlogicmaybe I'm misreading it then, yeah, I am.
22:57:34preglowamiconn: i haven't got anything in mind, just wondered if you had done anything
22:57:42preglowamiconn: just a but surprised the eq is so slow on arm
22:58:09amiconnI just wanted to know how much impact the eq has, and whether it will run realtime with mp3 when maxing out the 5022
22:58:12preglowjust seems like emac is bunches faster than the native arm multiplier
22:58:33amiconnIt does, and the result is furthermore that the eq consumes the equivalent of 15MHz
22:58:47amiconn(with all 5 bands enabled)
22:59:11lostlogicpreglow: if you have a moment, this is the problem code, line 6 is the data abort.
22:59:14preglownot much more i can do for it either, without compromising sound quality
22:59:43amiconnYOu said something about 64bit multiplies? Any chance to reduce the number of 64bit multiplies?
22:59:43lostlogicpreglow: that's probably why the wolfson codec has a decent 5 band equalizer built in? (at least the specs on it seem decent)
23:00:09Ribs'lo everyone
23:00:17preglowlostlogic: well, r6 is just incremented by one, it'll be invalide sooner or later unless it's aligned
23:00:26Ribshey lostlogic −− are there any plans to use the second CPU in the iPod?
23:00:28preglowlostlogic: yeah, it seems pretty decent, just a wee bit limited
23:00:30amiconnlostlogic: The 8731 and derivates don't have any hardware tone controls :(
23:00:34preglowRibs: yes, but that's a lot of work
23:00:44Ribsnot for a while then :)
23:00:58preglowamiconn: well, not really
23:01:09lostlogicpreglow: ah, I understand the problem in my code then.
23:01:26preglowamiconn: you need five multiply and accumulate instructions per sample
23:01:46preglowamiconn: if we can't use the smlal instruction, then we need to use the mla one, which again requires everything to fit in 32 bits
23:02:41preglowamiconn: and a couple of guard bits would be nice too, to avoid overflowing samples without having to do carry checks all the time, which leaves us with less than sixteen bits of audio precision
23:03:14preglowi hate the bloody arm extended multiplier unit
23:03:32preglowit uses early termination, so the time it uses is dependent on input data
23:03:35amiconnWow... Slasheri's tagcache commit speedup cuts commit time on my recorder down to less than half
23:03:50amiconn3:30 before, 1:30 now
23:03:56preglowthat is indeed very decent
23:06:33amiconnAnd that's already the second speedup. First it went 5:00->3:30
23:07:27 Join BHSPitLappy2 [0] (i=Steve-O@
23:08:02 Join chi [0] (
23:08:04preglowbtw, has anyone started to make a host based tagdb updater?
23:08:46 Join ep0ch [0] (n=ep0ch@
23:08:59KyomiHas anyone had the problem of their usb connection to their DAP suddenly die and start again in the middle of file transfer?
23:09:44lostlogicahahaha, when I commit this pcmbuf rework, it will make the actual code faster, but the debug screen slower (8 or 10 more instructions per audio thread debug update, but a few fewer instructions in the DMA interrupt :)
23:10:23ep0chpreglow: speaking of crossfeed, on ihp the attenuation controls in the UI should be inverted? i.e. joystick up is currently making the dBs go down and vice versa
23:10:33amiconnlostlogic: What do you think how short before the dma we could mix voice with a secondary pcm buffer?
23:10:58preglowep0ch: i know, but i can't be bothered
23:11:02Learlostlogic: btw, when seeking (at least mp3 for sure), it seems you've re-introduced the "display the file time before the seek for a brief moment" thing. :)
23:11:05preglowep0ch: also eq precut is like that
23:11:35ep0chahh never touched eq precut, i may have a look and post a patch
23:11:41amiconnAlso, do you think a secondary pcm buffer would make the code less complex? Iiuc you said that voice is currently handled very different depending on whether music is playing in parallel or not. That could be changed with a secondary pcm buffer, couldn't it?
23:11:57lostlogicLear: yeah, I've noticed.
23:12:29lostlogicamiconn: the voice is handled as music when music is not playing, so a secondary pcm buffer would only add complexity
23:12:37lostlogicamiconn: safe mix distance is ~1/4s
23:12:48chibefore i ever heard of rockbox, i accidently formated my nano. and then i thought may as well try rockbox. so im folowing the nano directions for the install but it requiers some stuff from the nanos boot partition wich was formated heh. is there a way to get a bootpartition.bin? or am i out of luck?
23:13:24amiconnHmm... I would think that this different handling wouldn't be necessary anymore with a secondary buffer
23:13:43 Quit Lear ("Chatzilla 0.9.72 [Firefox]")
23:14:10*preglow kicks coldfire data sheets
23:14:41kapheineI could die a happy man if I never had to look at the 5282 or 5271 datasheet ever again :P
23:14:42lostlogicamiconn: would still have to choose between mixing and playing
23:14:58lostlogicwould need to have a separate DMA callback for when just voice is playing vs. when voice and music
23:14:59ep0chany up-to-date at&t english voice files available?
23:15:11amiconnYes, but the voice codec would always decode into the secondary buffer
23:15:12 Quit TCK (Read error: 104 (Connection reset by peer))
23:15:27lostlogicyes, the decoding side would be simplified somewhat.
23:15:29preglowpetur: i pretty much have had to experiment my way forward in finding out what needs to be reset and so forth
23:15:52amiconnShould make preemption rather simple
23:16:05 Join lostnihilist [0] (
23:16:09preglowpetur: if i after recording when and played a file, recording would be broken after it
23:16:11chii wonder if someone has a bootpartition.bin from a nano i could download from them?
23:16:24preglowseems i just had to reset some stuff a bit early, and etc etc etc
23:16:40lostlogicamiconn: yes, and it would also help eliminate the bug that's been reported of audio stop during voice freezing.
23:16:54lostlogicamiconn: I'm definitely not against the idea of having a separate ~200k voice decode area.
23:16:55 Quit BHSPitLappy (Read error: 110 (Connection timed out))
23:17:07lostlogicspecially if we can get rid of malloc in codecs and reclaim that 500k
23:17:19KyomiAnyone have any suggestions for me?
23:17:36amiconnlostlogic: 200k? Is voice already resampled when buffering?
23:17:48lostlogicamiconn: yes, DSP happens before pcm buffer
23:17:54KyomiI have problem of my usb connection to my h320 suddenly dying and starting again in the middle of a file transfer...
23:17:59bluebrotherKyomi: I have the USB sometimes stalling
23:18:19peturKyomi: usb is handled by hardware
23:18:22bluebrotherbut this is simply because the hdd of my h120 is too slow and the buffer on the PC got full.
23:18:29Massa_Kyomi: does your problem only exist with the H320? Or do other USB devices show the same problem? Does this problem only exist on one PC or on all PCs you can get your hands on?
23:18:44preglowlostlogic: you have any issues with lots of small files in playback?
23:18:48bluebrotherthe dap hdd light is always on, so there is activity −− just wait a bit.
23:18:49 Join damaki__ [0] (
23:18:52Massa_Kyomi: I have no such problems with my H340 device...
23:18:56lostlogicpreglow: not tried playing lots of small files.
23:19:04preglowit seems to cut them rather badly here
23:19:22lostlogicdefine cut?
23:19:32preglowas in the ends are removed
23:19:44preglowand the progress bar has serious behavioural problems
23:19:49lostlogicgee fun.
23:19:56 Join Poka64 [0] (
23:19:57preglowthat's what i said too
23:20:09preglowthen i said "no wait, that's not fun at all!"
23:20:24 Quit Massa_ ("IceChat - it may not be hot but damn it's cool")
23:20:26amiconnlostlogic: Hmm, maybe voice should be resampled before pcm... but does the pcm buffer handle mono?
23:20:30lostlogicI have had ends of songs get cut (pcmbuffer length worth), which is part of why I started looking at the crossfade code. Not sure if it relates though.
23:20:48Poka64will there ever be a version for 1st generation Ipod mini ?
23:20:56lostlogicamiconn: no, the pcm buffer is data fed _directly_ to the hardware, so is stereo, 44.1khz
23:21:02preglowPoka64: it should work now
23:21:17bluebrotherPoka64: look at the daily builds page.
23:21:20amiconnHmm, voice buffered as stereo seems like a waste...
23:21:24Poka64preglow, are you sure, tried a couple of weeks ago
23:21:32preglowPoka64: things are happening fast around here
23:21:34Poka64everything got messed up
23:21:40lostlogicamiconn: sure, but how are you going to do the just-in-time conversion −− would require another thread in the audio system, I think :(
23:22:05preglowwaste, yes, can we survive it for now? yes
23:22:12preglowlets just make it work properly
23:22:17preglowwe can fix it after 3.0
23:22:32amiconnHmm. If we're mixing, the low-latency mixer could easily handle mono-to-stereo
23:22:41amiconnThe question is what to do when not mixing
23:23:03lostlogicmm, I still don't know where you're going to do your low-latency mixing (threadwise) I don't have a place for it in mind.
23:23:29amiconnWhere is it currently done?
23:23:36lostlogicon the decode thread
23:24:08amiconnAh, yes, so introducing a secondary buffer would add the need for another thread...
23:24:12 Quit bluebrother ("Leaving")
23:24:39amiconnThe pcm playback code itself runs in isr context?
23:25:01 Quit Ribs ("Leaving")
23:25:10lostlogicjust the dma callback, everything else is done on the decode thread.
23:25:35***Saving seen data "./dancer.seen"
23:26:22lostlogicdo these two lines seem idiotic to you guys?
23:26:23lostlogic int sample = ((int)*input_buf++ * factor) >> 8;
23:26:23lostlogic sample += ((short *)crossfade_chunk->addr)[crossfade_sample];
23:26:50preglowamiconn: 320kbps is now realtime at 45mhz
23:26:55preglowwith a nice margin
23:27:05lostlogicfactor is an unsigned int defined as ((size_t - size_t)<<8)/size_t
23:27:27preglowantonius is doing great work
23:27:34preglowsomeone buy him an ipod
23:28:25amiconnSlasheri: tagcache gets confused if you delete the tagcache*.tcd files at runtime and then switch to the tagcache view
23:29:06preglowep0ch: you still here?
23:29:08 Join Arrogant [0] (
23:29:15amiconnDoesn't crash, but displays funny strings instead of the first 4 standard views, and when trying to enter these, displays a vkeyboard
23:29:22ep0chep0ch: yup
23:29:25preglowep0ch: how about you testing spdif recording? :)
23:29:29ep0chi'm ep0ch!
23:29:48preglowif it doesn't work for you people now, i don't know what to do
23:30:50amiconnDid you try different frequencies?
23:30:51 Join filoktetes [0] (
23:31:11amiconnI would, if I only had the needed cable...
23:31:35preglowamiconn: i did
23:31:46preglowbleh, wrong paste
23:32:08preglowamiconn: i've tried 44.1khz and 48khz, the only two i have access to
23:32:24peturI would, if I had the needed player :D
23:32:26amiconnYeah, would be the same here
23:33:00amiconnI can only use my dvd player as s/pdif source, playing an audio cd as 44.1kHz source, and a dvd as 48kHz source
23:33:11preglowi can use my pc and my cd player
23:33:25preglowpc is locked to 48khz, cd 44.1
23:33:41 Quit damaki_ (Read error: 110 (Connection timed out))
23:34:31preglowother sample rates sound shit while monitoring, but i can think of no way around that, and the iriver firmware sounds just as bad
23:34:37amiconnSlasheri: The 'Couldn't load codec' bug is still there....
23:34:38 Join matsl [0] (
23:35:34amiconnHow is monitoring done?
23:35:51amiconnJust passing through pcm?
23:36:46 Join ep0ch| [0] (n=ep0ch@
23:37:17 Quit BHSPitLappy2 (Connection timed out)
23:37:25ep0ch|preglow: i'll be able to do some testing for you tomorrow evening
23:37:33preglowyou can route pretty nicely without letting anything go through the cpu core
23:37:40preglowi just redirect ebu in 1 to iis2
23:38:18amiconnOkay, but that's still just passing through the signal
23:38:18preglowdoing it through pcm would of course be the better solution, allowing proper resampling, but it requires more knowledge about the current system than i have
23:38:32amiconnMaybe we need to resample in order to not make it sound shit...
23:39:04preglowit doesn't sound _that_ bad
23:39:16preglowbut you get some high frequency buzzing
23:39:21amiconn[23:34:33] <preglow> other sample rates sound shit while monitoring,
23:39:32preglowdepends what you think of shit :)
23:39:35 Join TCK [0] (
23:39:36amiconnI can't judge myself...
23:40:10*amiconn should go looking for such an optical cable (or adapter)
23:40:24amiconnI have a standard toslink cable
23:42:22 Part ep0ch| ("Kopete 0.11.1 :")
23:46:02 Quit ep0ch (Connection timed out)
23:47:15preglowthe tos/jack adapter is small and cheap
23:47:28preglowyou just tack it onto the end of the tos cable plug
23:48:12lostlogicalright, am I insane? casting a short to an int sign extends the short, right?
23:48:43preglowyea, if it's singed
23:48:46preglowheh, signed
23:48:52lostlogicyeah, it's getting singed :(
23:49:07 Quit chi (Remote closed the connection)
23:49:49amiconnlostlogic: sry, what's the problem
23:51:03lostlogicwhen I should be crossfading, I'm instead getting only the new audio, very distorted
23:52:12pillsome tracks end prematurely using yesterday's build
23:52:56pilland sometimes resume does not work, rockbox just stats un dir listing mode
23:52:59 Nick Bagder_ is now known as Bagder (
23:53:01lostlogicpill: yes, as I said a while ago, I've experienced that adn it's part of why I started looking into crossfade (crossafde is initialized on almost all track changes)
23:53:13lostlogicpill: both known bugs
23:53:15pilleven when turned off?
23:53:24lostlogicpill: yes
23:53:47pillbut it's getting better, i can tell
23:53:52lostlogicamiconn: this is where I _believe_ the problem must be happening, as the audio from the new track does play.
23:54:20amiconnlostlogic: Should the problem with the erased resume position be solved in current cvs?
23:54:34lostlogicamiconn: no
23:54:41lostlogicamiconn: haven't gotten to i t yet
23:56:28peturwon't sample * factor overflow?
23:56:58lostlogicpetur: that's why sample is a 32 bit int, which holds temporary results from a short * factor
23:57:32 Join mareviq [0] (
23:58:16lostlogicpetur: thanks for looking... I mean I'm sure I'm missing something similar to that :(

Previous day | Next day