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 2009-01-02

00:00:02mcuelenaerehas usb mass storage ever been tested in BOOTLOADER?
00:00:10gevaertsYes. It works for the gigabeat S
00:00:10 Join itcheg [0] (i=62db4767@gateway/web/ajax/
00:01:08gevaertsHave a look (or steal) at handle_usb() in gigabeat-s.c
00:01:26mcuelenaeregevaerts: shouldn't l.415 be moved outside that #if?
00:01:40mcuelenaere(I use the static buffer instead of the audio buffer)
00:02:29 Quit solexx (Read error: 104 (Connection reset by peer))
00:02:49gevaertsoops, yes, but the static buffer also needs to be bigger
00:02:56mcuelenaerehmm Windows seems to react a bit more positive now :)
00:03:14 Join advcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
00:05:24gevaertsmcuelenaere: I think something like
00:06:12mcuelenaereand what about tb.transfer_buffer ?
00:06:22mcuelenaereor is that another buffer?
00:06:29gevaertsThat's another one
00:07:32gevaertsIn the audio buffer case they both use a bit of the audio buffer, but in the static buffer case I think keeping them separate is a good idea. Check RAMDISK_SIZE to see if it's not too big for you (I think it's 8M now)
00:08:05mcuelenaereI get
00:08:33mcuelenaere#define RAMDISK_SIZE 2048
00:08:36mcuelenaereso it's 2MB
00:08:44mcuelenaereand AFAIK this target has 16MB
00:09:03*gevaerts misremembered it
00:09:11gevaerts1MB actually. That's in sectors
00:09:47*gevaerts wonders how linux would react to a non-power-of-two sectorsize :)
00:10:18 Quit {phoenix} (Read error: 104 (Connection reset by peer))
00:10:33mcuelenaere:) I get 'cat: /dev/sdc: No medium found' and 'cat: /dev/sdd: No medium found'
00:10:38mcuelenaereso there's still something wrong
00:10:55gevaertsDo you have all needed handling of things like SYS_USB_CONNECTED?
00:11:06mcuelenaerehmm no
00:11:30gevaertsThat's probably the problem then
00:13:54mcuelenaerebleh, I get a device reset
00:14:15 Quit CaptainKewl (Read error: 104 (Connection reset by peer))
00:14:26 Join CaptainKewl [0] (
00:15:36saratogajhMikeS: about your comment in fs#8523, what is the lcd_set_enable_hook callback for? doesn't the screen get updated anyway?
00:15:36 Quit itcheg (" ajax IRC Client")
00:16:16mcuelenaerewhenever I do usb_acknowledge(SYS_USB_CONNECTED_ACK); the device get reset for some reason
00:17:00 Quit Nibbl (Read error: 113 (No route to host))
00:17:26gevaertsdevice reset in the USB sense or more general?
00:18:39 Join itcheg [0] (i=62db4767@gateway/web/ajax/
00:19:00mcuelenaerefor some reason, the usb thread wants to system_reboot()
00:19:11gevaertsAh, ok.
00:19:39mcuelenaereI'm not sure whether it's the usb thread, but somebody wants to reboot :)
00:19:48gevaertsDo you have USE_ROCKBOX_USB defined, or only HAVE_USBSTACK?
00:20:14 Quit jhMikeS (Nick collision from services.)
00:20:19gevaertsIf you don't have it, this is working as specified :)
00:20:20 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
00:20:42mcuelenaereprobably only HAVE_USBSTACK :)
00:22:55mcuelenaerehmm it seems like it isn't transmitting the proper amount of data
00:24:02mcuelenaereand when I unplug, I get a panicf() mount: 0
00:24:59jhMikeSsaratoga: Does the patch prevent writes to lcd_framebuffer? If so, things will get out of sync and lcd_update won't suffice. I'm busy on another thing and haven't had a chance to really look at it.
00:25:21mcuelenaeregevaerts: this is part of the log:
00:25:30mcuelenaereit seems like it's stuck on something doing over and over
00:25:48mcuelenaereah no, now it stopped with scsi allow_medium_removal 1
00:27:12 Join anewuser [0] (i=anewuser@unaffiliated/anewuser)
00:28:31 Part Blackfrosch
00:28:40gevaertsmcuelenaere: The "Status code -84; transferred 3968/4096
00:28:47gevaerts" and similar don't look good
00:28:48mcuelenaereyep those
00:29:20 Quit robin0800 (Remote closed the connection)
00:29:44mcuelenaereis that transferring data to or from the device?
00:30:07gevaertsFrom. This is after a READ command
00:31:17mcuelenaerecould this be a problem with the last packet?
00:31:41mcuelenaerehmm no, it seems like there are 2 packets missing
00:32:19gevaertsIt's always two packets, except when the request was more than 16k
00:33:18*gevaerts didn't know that the linux usb-storage debug output was that helpful. He'd have started using that a long time ago otherwise
00:33:40mcuelenaereow, I just crashed my Linux VM :/
00:34:00*mcuelenaere curses Windows'es USB implementation
00:34:01saratogajhMikeS: it basically just disables gui_wps_refresh calls by returning without doing anything
00:34:08saratogai don't know what that does to the frame buffer
00:37:42 Join Aurix_Lexico [0] (
00:39:09 Join efyx [0] (
00:42:08 Join Blackfrosch [0] (
00:42:33 Part Blackfrosch
00:46:38 Quit itcheg (" ajax IRC Client")
00:47:53***Saving seen data "./dancer.seen"
00:49:30 Quit saratoga ("CGI:IRC (EOF)")
00:54:52 Join _Auron_ [0] (
00:55:43 Quit karashata ("G'bye everyone!")
01:00:47Unhelpfulnice, RockboxTesting has an item for Fuze, but no users. anybody have one to test something on?
01:01:40 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]")
01:03:22 Join kugel [0] (n=chatzill@unaffiliated/kugel)
01:03:40kugelUnhelpful: yes, I do :)
01:04:15Unhelpfulwell, i knew *you* did, i was going to see if i could perhaps confirm my hunch before you got back, and maybe have some *real* progress :/
01:05:37kugellucent and keanu have one too
01:06:45kugelUnhelpful: do you really think it's limited to my particular fuze?
01:08:17Unhelpfulit's obviously at least limited to certain devices, since it doesn't happen in sim, nor on any of the 16-bit color hardware i own. i don't know about your *personal* fuze... it might be that, or fuze in general. is fuze another PP target?
01:08:41kugelUnhelpful: no
01:09:21 Quit ZincAlloy ("CGI:IRC (EOF)")
01:10:16kugelUnhelpful: if it was my personal fuze, going 1 revision before your commit shouldn't help I suppose
01:10:34*Unhelpful goes digging in the wiki to see what the AMS devices' ancestry is, exactly...
01:11:04kugelUnhelpful: e200v2, c200v2, clip, m200v4 are ams too
01:11:14Unhelpfulkugel: even if it is only your device, it could be a glitch in the hardware that the code exposes, or a glitch in the code that the device exposes.
01:11:32kugelonly the first two (and the fuze) have a 16bit display though, and the c200v2 no working lcd dirver anyway
01:12:14kugelUnhelpful: I don't really think so
01:13:19kugelbut it's indeed weird that it doesn't show on my e200v1 (same display size (just portrait), same pixelformat
01:13:24Unhelpfulit's very odd for code that's that far from the hardware to be failing on one particular device, though. did you get / try the "fix" i had a hunch about?
01:14:09kugelno, will do now
01:14:31Unhelpfuli suspect dithering is to blame, somehow... though i don't see how that could fail for one target. anyway, try forcing it off, just add a format &= ~FORMAT_DITHER; above any of the tests against format
01:14:38Unhelpfulin bmp.c:read_bmp_fd
01:15:59 Quit einhirn (Read error: 104 (Connection reset by peer))
01:23:44kugelUnhelpful: doesn't help
01:25:11Unhelpfulkugel: good to know. i'll continue my work on bolting on a logger to the loader, that will gripe about various possible errors to a log file, then.
01:25:31kugelUnhelpful: uhm, weird things happen if I force FORMAT_RESIZE off
01:26:19Unhelpful...did i tell you to do that? ;)
01:26:32 Join krazykit [0] (
01:26:37kugelno, but I tried nevertheless
01:26:49kugeliiuc forcing that off shouldn't change anything?
01:27:10Unhelpfulno, it very clearly should not, in cases where it's not being requested.
01:27:27*kugel made a mistake
01:28:19kugelah, ok, no it doesn't change anything
01:29:33kugelUnhelpful: your defines in lcd.h look a bit weird to me
01:30:17kugelUnhelpful: why that? isn't sizeof(fb_data) sufficient?
01:31:47Unhelpfulsizeof is not understood by the preprocessor. there was some reason i had wanted to be able to compare the main and remote data type size in an #if, although i'm pretty sure i ended up not doing so.
01:33:02kugellooks like
01:33:51Unhelpfulyou could, to be very, very sure, try redefining DITHERXDY as 0... but i don't see any other way for dither to be true if you mask out FORMAT_DITHER at the top
01:34:22kugelUnhelpful: haven't you said you can't blame my lcd driver, since the issues how in the screenshot?
01:34:58kugelUnhelpful: forcing dithering of had an effect, but it not a good one
01:35:10kugeland it didn't fix the issue too
01:35:12Unhelpfulooh... how did it change thnigs?
01:35:14 Quit nplus (Remote closed the connection)
01:35:26kugelhard to describe
01:35:32kugelwanna see a ss?
01:36:10Unhelpfulplease, perhaps it would produce another hunch, maybe more useful than my first one.
01:41:57Unhelpfulyes, i see that it's exactly as broken as before. the banding is an expected change, since you disabled dithering... the glitch itself seems the same as before.
01:44:30 Join tmzt [0] (n=tmzt@
01:47:45kugelUnhelpful: well, I don't want to complain, but it might've been helpful if you committed that one in smaller parts
01:50:46 Quit perrikwp (" ajax IRC Client")
01:55:05Unhelpfulmany of the pieces depended upon others, unfortunately. but the ones that are likely to be behind your problem might not be hard for me to pick apart.
01:55:14 Quit matsl ("Leaving")
02:03:01 Quit mcuelenaere (Read error: 113 (No route to host))
02:14:22 Quit ender` (" Kiss a pair of boobs and the movie's rated R. Chop them off and it's PG-13. -- Jack Nicholson (?)")
02:17:03 Quit Aurix_Lexico ("Leaving.")
02:17:22 Join aurix_lexico [0] (
02:22:05 Quit tessarakt (Read error: 113 (No route to host))
02:25:45soapBaw, you called?
02:28:18*Unhelpful sees there aren't very many config differences at all between fuze and e200... but i'll compare preprocessed source just in case :/
02:28:45ameyerwhat exactly is the issue?
02:29:02*ameyer looks at kugel's screenshot
02:29:49*ameyer still doesn't get the problem
02:35:07*n1s hands ameyer new glasses
02:35:46 Join ameyer_ [0] (
02:36:25 Join Nibbler [0] (
02:37:19n1sJdGordon: yellow
02:40:10 Quit ameyer (Nick collision from services.)
02:40:16 Nick ameyer_ is now known as ameyer (
02:40:25JdGordonhmm... does changin a functions ret from void to bool need a plugin api bump?
02:40:36 Join itcheg [0] (i=62db4767@gateway/web/ajax/
02:42:23n1sat least the regular one maybe not the MIN
02:43:20JdGordonI'll p[lay it safe and bump both, and put the func where it should be
02:46:24Unhelpfulameyer: seriously, you see nothing wrong?
02:47:55***Saving seen data "./dancer.seen"
02:48:47ameyerit doesn't quite look right, but that screenshot is undithered.
02:49:07ameyerwait, the k and the b look... squished together
02:49:23Unhelpfulhere's the one before i had him disable dithering:
02:49:40ameyerkinda looks like KDE4's font mang...erm... handling
02:50:03Unhelpfuland you're still not noticing the diagonal rows of bright dots?
02:50:30ameyerahh, now I see them
02:50:48ameyerthey're kinda non-obvious without zooming way in
02:51:12 Quit faemir (Remote closed the connection)
02:52:30ameyer(on a 8.9" 1024x600 display at about 1m)
02:55:35 Quit tvelocity ("Αποχώρησε")
03:10:54 Join t0mas [0] (n=tomas@rockbox/developer/t0mas)
03:11:21t0masis there any active knowledge on rockbox and the Creative Zen?
03:11:38Lloreant0mas: Everything's in the forum thread.
03:11:43LloreanOr possibly the wiki
03:11:53t0masBecause I'm thinking about buying one (love the look of it :-)) and at the same time re-join active Rockbox development
03:12:01t0masyeah, I didn't spot a wiki page about it?
03:12:23LloreanWell it's possible there's not one.
03:12:31t0masoh wait
03:12:33t0masjust found one
03:12:44t0masseems like it is possible to restore a bricked player from software
03:12:59t0masso it should be fairly easy to get testers
03:14:07 Join saratoga [0] (n=62e14d04@gateway/web/cgi-irc/
03:14:42 Join Thundercloud [0] (
03:17:16scorche|sht0mas: the very definition of a bricked device is a player broken past the point of fixing via software..
03:17:50scorche|shand long time no see ;)
03:17:56saratogacan rockbox encode to flac?
03:18:23t0masscorche: indeed :)
03:18:28t0masbut you got the point..
03:18:41saratogat0mas: you did some of the MAD arm optization right?
03:18:47t0masI'll have to run now.. (lot of things keeping me busy the last year.. work.. girlfriend..)
03:18:56t0masno, I did the first graphics implementation
03:18:58t0masfor the WPS
03:19:06saratogaah must be thinking of someone else
03:19:11t0masback when it was still black and white ;-)
03:19:13Lloreansaratoga: tomal possibly?
03:19:18saratogayes tomal
03:19:23saratogatoo similar names
03:20:16t0masI'll drop by tomorrow or something.. have to get a Zen next week.. and then get back into Rockbox programming
03:20:29ameyersaratoga: as far as I know, there isn't a fixed-only FLAC encoder (yet?)
03:20:40t0masto bad I've done only J2EE programming the last 1.5 year :)
03:20:50n1ssaratoga: nope, rockbox can not encode flac
03:21:02n1sonly waw, mp3 and wv IIRC
03:21:20 Quit t0mas ("good night!")
03:21:26 Quit ibseo (Read error: 60 (Operation timed out))
03:21:29ameyeron a semi-related note, what does rockbox use for encoding to mp3? Shine?
03:21:37ameyern1s: apparently, AIFF as well
03:21:59n1sah, yes
03:22:25ameyerbut that's basically WAV in a different container
03:23:32ameyerwait, there is a fixed-point FLAC encoder. Unfortunately, it apparently still uses 64 bit ints, which apparently makes it unusable in rockbox.
03:25:29n1swouldn't any lossless encoder need to be fixed point?
03:26:05LloreanSeems like it.
03:26:14ameyerapparently some lossless encoders use floating point internally, to a certain extent. Probably related to FFT.
03:26:35 Quit Horscht ("electromagnetic radiation from satellite debris")
03:26:37saratogai think you can use fp internally for calculating things like filters or whatever in a lossless encoder
03:27:03ameyeralso, some lossless formats can do fp audio.
03:27:14n1sah, i don't know too much about encoders :)
03:27:35saratogai think a codec doing lossless fp audio still has to do int for most things
03:27:47saratogathe sample format shouldn't matter
03:28:12n1sameyer: 64 bit ints alone shouldn't make it unusable in rockbox, Tremor uses them for example
03:28:13saratogajust that your rounding errors are well defined
03:28:41 Join Krisinu [0] (n=Krisinu@
03:28:59ameyern1s: it was presented as some sort of obstacle in
03:29:00Krisinuhi guys
03:29:35 Join karashata [0] (
03:30:40 Quit tyfoo2 ("Carpe diem")
03:30:49Krisinuany advantage of installing manually vs auto?
03:31:03ameyernot really
03:31:40Krisinuso just a pride thing
03:31:55LloreanNot even a pride thing
03:32:09LloreanThe manual install is basically just "the old way before we got the new tools built"
03:32:12Krisinuif all else fails?
03:32:20Krisinufair enough
03:32:45LloreanIt uses a lot of the same code anyway. It's really only useful if you don't have a graphical environment.
03:33:30 Join Willwolfe [0] (n=cdea230e@gateway/web/cgi-irc/
03:34:04Krisinuyou guys have a reliable source for replacing ipod batteries?
03:34:21Krisinumy ipod mini only lasts 20 minutes now, ipod video about 1.5hrs :(
03:35:36n1sameyer: the libflac change log mentiones that it used floating point in "a couple places" and the reason against 64bit integer math is that it's slower than 32bit
03:35:41 Quit DerDome (Read error: 60 (Operation timed out))
03:35:43 Quit HBK (Read error: 104 (Connection reset by peer))
03:35:48Krisinu looks interesting
03:36:30Krisinuthey have solar chargers!
03:38:23WillwolfeI tried compiling r19633 w/ fs#9703 and fs#9706 as per yesterdays log but in my case patching with fs#9706 fails with a Hunk # failure?
03:40:53WillwolfeI was attempting a clip patch.
03:41:16WillwolfeThat is before compiling.
03:45:29Krisinureplacing batteries doesn't affect the rockbox installation in any way, correct?
03:46:22n1syou can change the battery capacity in rockbox to get a better runtime prediction but otherwise no
03:46:42ameyern1s: in my experience, "better" still == "horrible"
03:47:00n1si didn't say "good" :)
03:47:15*ameyer gets way more than 5h31m out of his "850 mAh" iPod mini battery
03:48:11Krisinuwhere did you get the 850?
03:48:38ameyeroff of I somehow doubt it's actually 850 mAh.
03:48:52Lloreanameyer: Well, it has to be calibrated for the player first. It probably hasn't been calibrated for Minis yet.
03:48:56Krisinui think i'll get the 750 kit from ipodjuice
03:49:36 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
03:49:37ameyeror was calibrated back in the day when rockbox had horrible battery life on the mini
03:49:44ameyererm, all ipods, even
03:50:32 Quit Nibbler (Read error: 148 (No route to host))
03:50:52 Quit itcheg (" ajax IRC Client")
03:52:37 Join itcheg [0] (i=62db4767@gateway/web/ajax/
03:53:58JdGordonis anyone else having problems with the cabbie theme (or any which used to work) no longer working and showing the plain inbuilt wps instaed?
03:55:30ameyerAFAICT, all provided WPSes worked on the c200 after the major statusbar commit.
03:57:31ameyerand on r19635, works on the mini2g as well
03:57:33saratogawhy does the e200 define CPU_FREQ as 75mhz in its config file
03:58:00n1ssaratoga: the cpu frequency defines are a mess
03:58:01JdGordonI wonder what Zagor did to have it not working? ;p
03:58:35 Quit n1s ()
03:59:18ameyerspeaking of not working, I need to report a couple bugs related to the statusbar commit
03:59:33saratogaameyer: theres no runtime estimate info at all for the mini, which is why it didn't work for you
03:59:37saratogai guess i can commit some now
03:59:45saratogaprobably the same as my e200
04:03:23 Quit saratoga ("CGI:IRC (EOF)")
04:09:51 Join blkhawk- [0] (
04:10:56Unhelpfulkugel: try this patch... right now, it's only logging input/output size, if delta or rgb values for an output pixel exceed the correct range, and if the scaler is called. it should be enough just to start the player, since wps bitmaps all get loaded @ startup.
04:11:02Unhelpful <- silly me, there's the patch
04:13:46kugellol, another pastebin size
04:14:15Unhelpfulthere will never be enough
04:15:09Unhelpfulif you don't start sliding_puzzle or pictureflow, or attemp to load albumart, none of the logfiles should contain "calling scaler", obviously
04:16:05Unhelpful... i see a missing \n in a printf. just find the line with OVERFLOW ERROR in it. and *that* should not appear in a log either, but i'm expecting that it might.
04:17:05 Nick fxb is now known as fxb__ (
04:17:19 Quit Bensawsome (Read error: 60 (Operation timed out))
04:17:33 Join saratoga [0] (n=62e14d04@gateway/web/cgi-irc/
04:17:35*kugel is getting weird warnings
04:17:59saratogamini and sansa runtime estimation is updated
04:18:07kugelCONFIG_RTC, CONFIG_CODEC, SWCODEC not defined :?
04:19:16kugelUnhelpful: seems all fine
04:19:16 Quit MethoS-- (Read error: 131 (Connection reset by peer))
04:19:45Unhelpfulhuh, none of the things i thought might go wrong are? :/
04:20:03kugelseems not
04:20:26 Quit fdinel (Read error: 104 (Connection reset by peer))
04:21:45Unhelpfulhrm. that's not good... i guess i'll try dumping the bitmap data before and after converting to screen format, and we'll see where that gets us :/
04:23:36WillwolfeTrying a compile with only fs#9703 applied.
04:24:22Unhelpfuloverflows in red would've been a perfect explanation for the funky green strip in the yellow part :/
04:25:11 Join Bensawsof [0] (
04:26:16soapsaratoga, the blue LEDs on the Sansa E200 pull 20 ma?
04:26:17 Nick Bensawsof is now known as Bensawsome (
04:27:23 Quit Thundercloud (Remote closed the connection)
04:27:25soapwow - I guess it should not be surprising, always known backlights were major power draws. I guess LED button lights aren't all that different from LED backlights.
04:27:25 Quit blkhawk (Read error: 110 (Connection timed out))
04:27:50 Nick blkhawk- is now known as blkhawk (
04:29:15saratogasoap: I actually thought I remembered it being a lot higher
04:30:29saratogabut yeah the LEDs are really, really bright even compared to the LCD
04:31:26 Quit Willwolfe ("CGI:IRC")
04:31:32 Join Willwolfe [0] (n=cdea230e@gateway/web/cgi-irc/
04:32:04 Quit Schmogel (Read error: 110 (Connection timed out))
04:33:24 Join HBK [0] (
04:34:28 Join lymeca [0] (
04:35:56 Quit Willwolfe (Client Quit)
04:37:33 Join Will [0] (
04:37:48 Nick Will is now known as Willwolfe (
04:41:41 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
04:43:45 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:47:56***Saving seen data "./dancer.seen"
04:53:05 Join Barahir_ [0] (
04:55:20 Join MOHDMACH [0] (
04:56:56 Part MOHDMACH
04:58:50 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.0.5/2008121623]")
05:05:36 Quit lymeca (Read error: 110 (Connection timed out))
05:06:28 Quit HBK ("ZNC -")
05:08:52 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
05:10:22 Quit Barahir (Read error: 110 (Connection timed out))
05:12:59 Quit itcheg (" ajax IRC Client")
05:15:50 Join itcheg [0] (i=62db4767@gateway/web/ajax/
05:17:20 Join MOHDMACH [0] (
05:18:09MOHDMACHIs there anyway to change the default status bar that is displayed in the main menu?
05:24:48karashatacurrently, no, the only options are on or off
05:25:42karashatathere may or may not be some way to display a custom statusbar (from part of the WPS, most likely) down the road, but as far as I know it's not currently in the works
05:26:00 Quit itcheg (" ajax IRC Client")
05:26:42ameyerkarashata: that's not entirely true. You can set the volume and charge remaining displays to numeric or graphic.
05:27:34ameyersettings > general settings > display > status-/scrollbar
05:27:36 Quit efyx (Remote closed the connection)
05:27:47karashataameyer, good point, I forgot about that...
05:27:53LloreanIsn't all of this in the manual?
05:28:22ameyerwell, it should be
05:28:22 Join DerDome [0] (
05:28:25*ameyer goes to look
05:29:08ameyeryeah, it is.
05:29:57 Join MOHDMACH_ [0] (
05:30:02MOHDMACH_So how do the developers edit the status bar when they do it?
05:30:28MOHDMACH_Yes I do know how to change that stuff.
05:30:38MOHDMACH_The graphical and numerical stuff.
05:31:21MOHDMACH_I was just thinking it is small and wanted to make it bigger or customize it.
05:31:43MOHDMACH_Sorry if I wasn't clear on that.
05:33:32 Quit MOHDMACH (Read error: 60 (Operation timed out))
05:37:53 Quit anewuser (Client Quit)
05:38:54ameyerI'm going to state the obvious and say "by editing the source". No, I'm not sure exactly where.
05:42:42MOHDMACH_I was thinking that but I guess I will persist with specifics when I gain more knowledge about programming.
05:49:08MOHDMACH_Is version 3.1 supposed to be able to support most of the themes on the "WPS gallery" from the website?
05:49:49LloreanThemes are made to support a build, not the other way around. If their authors (or others) don't update them, they fall behind pretty easily on occasion
05:53:28MOHDMACH_Thanks. That answers why some themes were not working with the 3.0 and 3.1.
05:55:25 Quit Willwolfe ("ChatZilla 0.9.84 [Firefox 3.0.1/2008070208]")
05:58:03Unhelpfulthat's why there's been a push semi-recently for modifiable/redistributable licenses for themes, so that the theme on the site can be fixed without having to ask the author to do it.
05:59:25LloreanIf the new site ever gets up, the idea is to both require a license that allows modification, and have a script that marks themes that have lost compatibility so people can know which need fixing.
05:59:44MOHDMACH_Yes. I know what you mean. It is hard to get in touch with them but also Rockbox staff cannot answer questions about themes they did not make.
06:01:18Unhelpfulright, but it's pretty often clear to the dev who changed WPS how to "fix" themes that use the changed tags. generally a not-very-smart perl script could do it, *if* we are allowed to replace the theme on the site with the result of processing via the script.
06:05:07 Join HBK [0] (
06:05:33LloreanUnhelpful: Sometimes it's impossible to replace tags, it often depends on what was changed.
06:05:52MOHDMACH_To me that sounds complex. I thought playing around with WPS tags usually is faster and easier unless the tags derived from perl or something. Sorry I do not know anything about perl scripting. I'm still trying to get used to WPS tags.
06:06:08LloreanFor example, margin tags to viewports would've been very, very complex to replace since you'd have to check the font and do some math to figure out Y coordinates that weren't even used before
06:06:43UnhelpfulLlorean: true. on the other hand, the progressbar change was pretty simple to update for
06:08:25MOHDMACH_Yeah I guess that is a big obstacle for me because I wanted to design a WPS that functions like the WPS on the original ipod firmware.
06:09:14Unhelpfulas i believe was stated before, certain aspects of that may be difficult or impossible to replicate with FPS as it exists now
06:09:21LloreanUnhelpful: yes, it's basically case-by-case.
06:10:01MOHDMACH_So it would be a little difficult or near impossible?
06:10:09 Join peter__b [0] (
06:10:20 Join pixelma_ [0] (
06:10:36 Quit peter-b (Read error: 60 (Operation timed out))
06:10:37 Quit amiconn (Nick collision from services.)
06:10:39 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
06:10:47MOHDMACH_What is FPS?
06:11:13 Quit pixelma (Read error: 110 (Connection timed out))
06:11:26karashataframes per second
06:11:29UnhelpfulMOHDMACH_: that all depends on which parts of the UI you might be willing to accept not being exactly the same. and i don't really know the iPod UI well enough to talk about specifics.
06:14:24*ameyer wonders if Unhelpful has an ipod
06:15:46Unhelpfulonly an S60, two E200s, and the horrid unsupported samsung thingy.
06:15:47 Quit jhMikeS (Read error: 104 (Connection reset by peer))
06:16:35 Join CaptainKwel [0] (
06:16:59MOHDMACH_If you are talking about frames per second then themes I have seen on the ipod so far don't need any adjusting. I am just trying to see if I can create a separate and temporary screen with a volume bar similar to the original firmware.
06:18:52 Quit advcomp2019 (
06:18:52 Quit solexx_ (
06:18:52 Quit write__erase (
06:18:52 Quit Xerion (
06:18:52 Quit SUSaiyan` (
06:18:52 Quit ch4os (
06:18:52 Quit bugonshoe (
06:18:52 Quit Kohlrabi (
06:18:52 Quit dionoea (
06:18:52 Quit tmzt (
06:18:52 Quit Dieterbe (
06:18:52 Quit rasher (
06:18:52 Quit _Auron_ (
06:18:52 Quit CaptainKewl (
06:18:52 Quit AndyI (
06:18:52 Quit gevaerts (
06:18:52 Quit goffa (
06:18:52 Quit JdGordon (
06:18:52 Quit crashd (
06:18:52 Quit crwl (
06:18:52 Quit keanu (
06:18:52 Quit courtc (
06:18:52 Quit larstobi (
06:18:52 Quit Krisinu (
06:18:52 Quit abo_ (
06:18:52 Quit Slasheri (
06:18:52 Quit jon-kha (
06:18:52 Quit FOAD (
06:18:52 Quit desowin_ (
06:18:52 Quit blahrus (
06:18:54 Quit parafin (
06:18:54 Quit freqmod_qu (
06:18:54 Quit Horscht (
06:18:54 Quit BlakeJohnson86 (
06:18:54 Quit scorche (
06:18:54 Quit amigan_ (
06:18:54 Quit lastebil (
06:18:54 Quit Hadaka (
06:18:54 Quit scorche|sh (
06:18:54 Quit DerDome (
06:18:54 Quit perrikwp (
06:18:54 Quit ameyer (
06:18:54 Quit Zambezi (
06:18:54 Quit shodanX (
06:18:54 Quit tim__b (
06:18:54 Quit Kopfgeldjaeger (
06:18:54 Quit preglow (
06:18:54 Quit DaCapn (
06:18:54 Quit Tuplanolla (
06:18:54 Quit balou (
06:18:54 Quit suom1 (
06:18:54 Quit liiwi (
06:18:54 Quit lostlogic (
06:18:54 Quit krazykit (
06:18:54 Quit sarixe (
06:18:54 Quit pabs (
06:18:54 Quit ment (
06:18:54 Quit HellDragon (
06:18:54 Quit flux (
06:18:54 Quit trisiak (
06:18:54 Quit loswillios (
06:18:54 Quit fred_2 (
06:18:54 Quit film42 (
06:18:54 Quit markun (
06:18:54 Quit Neovanglist (
06:18:54MOHDMACH_The only things I want to add is the swapping animation and that temporary screen with a volume bar and I was under the impression that it was doable for a beginner.
06:18:54UnhelpfulMOHDMACH_: i believe you could do that much, with conditional viewports. there's a condition for "while volume is changing"
06:19:25MOHDMACH_Yes I heard of that yesterday.
06:21:02BHSPitLappyWe don't need those other guys, anyway.
06:21:02MOHDMACH_Since there is no example I will I have to play around with it. But I am really a beginner.
06:21:02MOHDMACH_At making WPSes.
06:21:19MOHDMACH_So that part alone will probably take 10 days. :)
06:21:45 Quit reacocard (
06:21:45 Quit synergist (
06:21:45 Quit Unhelpful (
06:22:20 Quit amiconn_ (
06:22:20 Quit pixelma_ (
06:22:20 Quit peter__b (
06:22:20 Quit HBK (
06:22:20 Quit blkhawk (
06:22:20 Quit aurix_lexico (
06:22:20 Quit DataGhost (
06:22:20 Quit splinter (
06:22:20 Quit ze (
06:22:20 Quit Rob2222 (
06:22:20 Quit linuxstb (
06:22:20 Quit feisar (
06:22:20 Quit mud-rb (
06:22:20 Quit jfc (
06:22:20 Quit alexbobp (
06:22:20 Quit Nimdae (
06:22:20 Quit midkay (
06:24:11 Join __lifeless [0] (n=lifeless@
06:24:11NJoinamiconn_ [50] (n=jens@rockbox/developer/amiconn)
06:24:11 Join pixelma_ [0] (n=pixelma@rockbox/staff/pixelma)
06:24:11NJoinpeter__b [0] (
06:24:11NJoinHBK [0] (
06:24:11NJoinblkhawk [0] (
06:24:11NJoinaurix_lexico [0] (
06:24:11NJoinDataGhost [0] (i=dataghos@unaffiliated/dataghost)
06:24:11NJoinsplinter [0] (
06:24:11NJoinze [0] (i=ze@
06:24:11NJoinRob2222 [0] (
06:24:11NJoinlinuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
06:24:11NJoinfeisar [0] (
06:24:11NJoinmud-rb [0] (n=mud@
06:24:11NJoinjfc [0] (
06:24:11NJoinalexbobp [0] (n=alex@
06:24:11NJoinmidkay [0] (n=midkay@rockbox/developer/midkay)
06:24:11NJoinNimdae [0] (
06:24:17 Join _jhMikeS_ [0] (
06:24:17 Join thegeek [0] (
06:24:17NJoinDerDome [0] (
06:24:17NJoinperrikwp [0] (i=4aa794a0@gateway/web/ajax/
06:24:17NJoinHorscht [0] (n=Horscht@xbmc/user/horscht)
06:24:17NJoinKrisinu [0] (n=Krisinu@
06:24:17NJoinameyer [0] (
06:24:17NJointmzt [0] (n=tmzt@
06:24:17NJoinkrazykit [0] (
06:24:17NJoin_Auron_ [0] (
06:24:17NJoinadvcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
06:24:17NJoinsolexx_ [0] (
06:24:17NJoinsarixe [0] (
06:24:17NJoinBlakeJohnson86 [0] (
06:24:17NJoinAndyI [0] (i=AndyI@
06:24:17NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
06:24:17NJoingoffa [0] (n=goffa@
06:24:17NJoinKohlrabi [0] (
06:24:17NJoinpabs [0] (
06:24:17NJoinwrite__erase [0] (
06:24:17NJoinJdGordon [0] (n=jonno@rockbox/developer/JdGordon)
06:24:17NJoinabo_ [0] (
06:24:17NJoinment [0] (
06:24:17NJoinXerion [0] (
06:24:17 Join HellDragon [0] (n=jd@Wikipedia/HellDragon)
06:24:17NJoinscorche [0] (i=Blah@rockbox/administrator/scorche)
06:24:17NJoinZambezi [0] (
06:24:17NJoinparafin [0] (
06:24:17NJoinKopfgeldjaeger [0] (
06:24:17NJoinblahrus [0] (n=blahrus@
06:24:17NJoinfilm42 [0] (
06:24:17NJoinbugonshoe [0] (
06:24:17NJoinkeanu [0] (n=keanu@unaffiliated/keanu)
06:24:17NJoinsuom1 [0] (i=markus@2001:4c40:1:0:0:0:0:11)
06:24:17NJoinHadaka [0] (
06:24:17NJoinbalou [0] (
06:24:17NJoinFOAD [0] (
06:24:17NJoinloswillios [0] (n=jan@unaffiliated/loswillios)
06:24:17NJoinch4os [0] (n=ch4os@gentoo/user/ch4os)
06:24:17NJointrisiak [0] (
06:24:17NJoindionoea [0] (n=dionoea@videolan/developer/dionoea)
06:24:17NJoinmarkun [50] (n=markun@rockbox/developer/markun)
06:24:17NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
06:24:17NJoinDieterbe [0] (
06:24:17NJointim__b [0] (
06:24:17NJoinshodanX [0] (
06:24:17NJoinDaCapn [0] (
06:24:17NJoinpreglow [0] (
06:24:17NJoincourtc [0] (n=court@unaffiliated/courtc)
06:24:17NJoinSUSaiyan` [0] (
06:24:17NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
06:24:17NJoinliiwi [0] (
06:24:17NJoinrasher [50] (n=rasher@rockbox/developer/rasher)
06:24:17NJoinscorche|sh [50] (n=scorche@rockbox/administrator/scorche)
06:24:17NJoinfreqmod_qu [0] (i=quassel@2001:700:300:1430:213:d3ff:fee9:5ed0)
06:24:17NJoinamigan_ [0] (
06:24:17NJoincrashd [0] (
06:24:17NJoinjon-kha [0] (
06:24:17NJoinflux [0] (
06:24:17NJoincrwl [0] (
06:24:17NJoinlastebil [0] (
06:24:17NJoinfred_2 [0] (
06:24:17NJoindesowin_ [0] (n=desowin@
06:24:17NJoinlarstobi [0] (n=larstobi@
06:24:17NJoinSlasheri [0] (i=miipekk@rockbox/developer/Slasheri)
06:24:17NJoinNeovanglist [0] (i=Neovangl@
06:24:32 Join Beta2K_ [0] (
06:27:08 Quit _lifeless (Read error: 60 (Operation timed out))
06:27:20MOHDMACH_So this did not work on my ipod and all I did was add this line to the WPS text file: %Vl|?mv2.5<bluevol.bmp| solid background.bmp>|
06:27:54MOHDMACH_%Vl|?mv2.5<bluevol.bmp| solid background.bmp>|
06:27:54 Quit write__erase (
06:27:54 Quit thegeek (
06:27:54 Quit solexx_ (
06:27:54 Quit SUSaiyan` (
06:27:54 Quit ch4os (
06:27:54 Quit bugonshoe (
06:27:54 Quit Kohlrabi (
06:27:54 Quit advcomp2019 (
06:27:54 Quit dionoea (
06:27:54 Quit tmzt (
06:27:54 Quit Dieterbe (
06:27:54 Quit Xerion (
06:27:54 Quit rasher (
06:27:54 Quit crashd (
06:27:54 Quit crwl (
06:27:54 Quit keanu (
06:27:54 Quit JdGordon (
06:27:54 Quit gevaerts (
06:27:54 Quit _Auron_ (
06:27:54 Quit _jhMikeS_ (
06:27:54 Quit AndyI (
06:27:54 Quit goffa (
06:27:54 Quit courtc (
06:27:54 Quit larstobi (
06:27:54 Quit Slasheri (
06:27:54 Quit jon-kha (
06:27:54 Quit FOAD (
06:27:54 Quit Krisinu (
06:27:54 Quit desowin_ (
06:27:54 Quit abo_ (
06:27:54 Quit blahrus (
06:27:54 Quit parafin (
06:27:54 Quit freqmod_qu (
06:27:55 Quit amigan_ (
06:27:55 Quit Horscht (
06:27:55 Quit lastebil (
06:27:55 Quit Hadaka (
06:27:55 Quit scorche|sh (
06:27:55 Quit scorche (
06:27:55 Quit BlakeJohnson86 (
06:27:55 Quit shodanX (
06:27:55 Quit tim__b (
06:27:55 Quit Kopfgeldjaeger (
06:27:55 Quit preglow (
06:27:55 Quit DaCapn (
06:27:55 Quit Zambezi (
06:27:55 Quit ameyer (
06:27:55 Quit perrikwp (
06:27:55 Quit Tuplanolla (
06:27:55 Quit DerDome (
06:27:55 Quit balou (
06:27:55 Quit suom1 (
06:27:55 Quit liiwi (
06:27:55 Quit lostlogic (
06:27:55 Quit ment (
06:27:55 Quit flux (
06:27:55 Quit trisiak (
06:27:55 Quit loswillios (
06:27:55 Quit fred_2 (
06:27:55 Quit film42 (
06:27:55 Quit sarixe (
06:27:55 Quit krazykit (
06:27:55 Quit pabs (
06:27:55 Quit markun (
06:27:55 Quit Neovanglist (
06:27:55 Quit HellDragon (
06:27:55 Quit Nimdae (
06:27:55 Quit splinter (
06:27:55 Quit peter__b (
06:27:55 Quit mud-rb (
06:27:55 Quit linuxstb (
06:27:55 Quit aurix_lexico (
06:27:55 Quit HBK (
06:27:55 Quit pixelma_ (
06:27:55 Quit Rob2222 (
06:27:55 Quit amiconn_ (
06:27:55 Quit midkay (
06:27:55 Quit jfc (
06:27:55 Quit ze (
06:27:55 Quit __lifeless (
06:27:55 Quit alexbobp (
06:27:55 Quit feisar (
06:27:55 Quit blkhawk (
06:27:55 Quit DataGhost (
06:28:50MOHDMACH_bluevol.bmp is just a solid blue bar that is the same size as the progress bar on the solid background image.
06:30:22BHSPitLappyMOHDMACH_, you might just want to wait until freenode stops having a fit
06:30:46NJoin_jhMikeS_ [0] (
06:30:46NJointhegeek [0] (
06:30:46NJoinDerDome [0] (
06:30:46NJoinperrikwp [0] (i=4aa794a0@gateway/web/ajax/
06:30:46NJoinHorscht [0] (n=Horscht@xbmc/user/horscht)
06:30:46NJoinKrisinu [0] (n=Krisinu@
06:30:46NJoinameyer [0] (
06:30:46NJointmzt [0] (n=tmzt@
06:30:46NJoinkrazykit [0] (
06:30:46NJoin_Auron_ [0] (
06:30:46NJoinadvcomp2019 [0] (n=advcomp2@unaffiliated/advcomp2019)
06:30:46NJoinsolexx_ [0] (
06:30:46NJoinsarixe [0] (
06:30:46NJoinBlakeJohnson86 [0] (
06:30:46NJoinAndyI [0] (i=AndyI@
06:30:46NJoingevaerts [0] (n=fg@rockbox/developer/gevaerts)
06:30:46NJoingoffa [0] (n=goffa@
06:30:46NJoinKohlrabi [0] (
06:30:46NJoinpabs [0] (
06:30:46NJoinwrite__erase [0] (
06:30:46NJoinJdGordon [0] (n=jonno@rockbox/developer/JdGordon)
06:30:46NJoinabo_ [0] (
06:30:46NJoinment [0] (
06:30:46NJoinXerion [0] (
06:30:46NJoinHellDragon [0] (n=jd@Wikipedia/HellDragon)
06:30:46NJoinscorche [0] (i=Blah@rockbox/administrator/scorche)
06:30:46NJoinZambezi [0] (
06:30:46NJoinparafin [0] (
06:30:46NJoinKopfgeldjaeger [0] (
06:30:46NJoinblahrus [0] (n=blahrus@
06:30:46NJoinfilm42 [0] (
06:30:46NJoinbugonshoe [0] (
06:30:46NJoinkeanu [0] (n=keanu@unaffiliated/keanu)
06:30:46NJoinsuom1 [0] (i=markus@2001:4c40:1:0:0:0:0:11)
06:30:46NJoinHadaka [0] (
06:30:46NJoinbalou [0] (
06:30:46NJoinFOAD [0] (
06:30:46NJoinloswillios [0] (n=jan@unaffiliated/loswillios)
06:30:46NJoinch4os [0] (n=ch4os@gentoo/user/ch4os)
06:30:46NJointrisiak [0] (
06:30:46NJoindionoea [0] (n=dionoea@videolan/developer/dionoea)
06:30:46NJoinmarkun [50] (n=markun@rockbox/developer/markun)
06:30:46NJoinTuplanolla [0] (n=jani@unaffiliated/tuplanolla)
06:30:46NJoinDieterbe [0] (
06:30:46NJoinlostlogic [50] (n=lostlogi@rockbox/developer/lostlogic)
06:30:46NJoinliiwi [0] (
06:30:46NJoinrasher [50] (n=rasher@rockbox/developer/rasher)
06:30:46NJoinscorche|sh [50] (n=scorche@rockbox/administrator/scorche)
06:30:46NJoinfreqmod_qu [0] (i=quassel@2001:700:300:1430:213:d3ff:fee9:5ed0)
06:30:46NJoinamigan_ [0] (
06:30:46NJoincrashd [0] (
06:30:46NJoinjon-kha [0] (
06:30:46NJoinflux [0] (
06:30:46NJoincrwl [0] (
06:30:46NJoinlastebil [0] (
06:30:46NJoinfred_2 [0] (
06:30:46NJoindesowin_ [0] (n=desowin@
06:30:46NJoinlarstobi [0] (n=larstobi@
06:30:46NJoinSlasheri [0] (i=miipekk@rockbox/developer/Slasheri)
06:30:46NJoinNeovanglist [0] (i=Neovangl@
06:30:46NJoinSUSaiyan` [0] (
06:30:46NJoincourtc [0] (n=court@unaffiliated/courtc)
06:30:46NJoinpreglow [0] (
06:30:46NJoinDaCapn [0] (
06:30:46NJoinshodanX [0] (
06:30:46NJointim__b [0] (
06:31:03MOHDMACH_Ok then no problem.
06:31:06NJoinreacocard [0] (i=reacocar@
06:31:06NJoinsynergist [0] (
06:31:06 Join Unhelpful [0] (n=Militant@rockbox/developer/Unhelpful)
06:31:27NJoin__lifeless [0] (n=lifeless@
06:31:27NJoinamiconn_ [50] (n=jens@rockbox/developer/amiconn)
06:31:27NJoinpixelma_ [0] (n=pixelma@rockbox/staff/pixelma)
06:31:27NJoinpeter__b [0] (
06:31:27NJoinHBK [0] (
06:31:27NJoinblkhawk [0] (
06:31:27NJoinaurix_lexico [0] (
06:31:27NJoinDataGhost [0] (i=dataghos@unaffiliated/dataghost)
06:31:27NJoinsplinter [0] (
06:31:27NJoinze [0] (i=ze@
06:31:27NJoinRob2222 [0] (
06:31:27NJoinlinuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
06:31:27NJoinfeisar [0] (
06:31:27NJoinmud-rb [0] (n=mud@
06:31:27NJoinjfc [0] (
06:31:27NJoinalexbobp [0] (n=alex@
06:31:27NJoinmidkay [0] (n=midkay@rockbox/developer/midkay)
06:31:27NJoinNimdae [0] (
06:33:04 Quit karashata ("G'bye everyone!")
06:34:25 Quit thegeek_ (Read error: 110 (Connection timed out))
06:39:56MOHDMACH_Also If this helps I would like to mention that I am trying to add this feature to the "icatcher" theme and have no patches installed on my firmware. The version is 3.1 and it was unmodified as far as patches go.
06:44:42MOHDMACH_Hmm? Tried it again and it did not work
06:47:57***Saving seen data "./dancer.seen"
06:54:11 Quit Beta2K (Read error: 110 (Connection timed out))
06:55:18 Quit HBK ("ZNC -")
06:56:34 Join Willwolfe [0] (
07:04:13MOHDMACH_So I guess I will repeat the problem once again. Sorry for if I was not clear.
07:04:27MOHDMACH_So this did not work on my ipod and all I did was add this line to the WPS text file:
07:04:38MOHDMACH_%Vl|?mv2.5<bluevol.bmp| solid background.bmp>|
07:04:49MOHDMACH_bluevol.bmp is just a solid blue bar that is the same size as the progress bar on the solid background image.
07:04:54JdGordonthats completly wrong...
07:05:06 Quit aurix_lexico ("Leaving.")
07:05:14JdGordonget rid of Vl| and it should work
07:05:18JdGordonand the last |
07:05:48MOHDMACH_How would I get the viewport then?
07:06:48JdGordonsetup the viewport on a seperate line
07:07:53JdGordonkugel: if oyu happen to read logs.... it seems that viewport_set_defaults() is called WAY to often in lists.. I'm not sure if that was caused by 9603 or not, but if your bored feel free to investigate :)
07:09:08 Quit Willwolfe ("ChatZilla 0.9.84 [Firefox 3.0.1/2008070208]")
07:09:35MOHDMACH_Is that bad?
07:10:15 Join HBK [0] (
07:10:34JdGordonis what bad?
07:11:02MOHDMACH_The stuff you said about the viewport set defaults.
07:11:24JdGordonno... its just a bit of a waste
07:12:15 Quit sarixe ("Connection reset by the motherfucking peer")
07:17:58JdGordonkugel: sorry... no its my fault...
07:18:42MOHDMACH_Are you saying sorry to me?
07:18:56MOHDMACH_Or is there a user named kugel?
07:20:18krazykitthere's a user named kugel
07:20:45MOHDMACH_Alright my mistake?
07:21:18MOHDMACH_Ignore the question mark.
07:21:52Bawthe apple talk support that is in SVN now is not 100% functional correct?
07:22:16MOHDMACH_So this is how I modified the line:
07:22:22Bawif anyone knows that is, i know there was some recent changes to it
07:22:44MOHDMACH_%?mv2.5<bluevol.bmp|solid background.bmp>
07:22:55JdGordonBaw: what apple talk support?
07:23:05MOHDMACH_and this is the viewport I came up with:
07:23:06LloreanBaw: What "apple talk" support?
07:23:17JdGordonMOHDMACH_: no, you have to load the bmp's first...
07:23:44Bawsorry, the apple talk protocol for accesories
07:23:51Bawfor iPods
07:24:20JdGordonhow is it not working?
07:24:34Bawsoaps custom patch which was like a huge forum post for months was merged to svn on the 25th
07:25:05scorche|shappletalk is ancient and used for networking computers...i assume you mean apple accessory protocol..
07:25:07Bawi have a 5.1 surround sound system that has an iPod dock attached and it doesnt connect to it like it normally would with apple fw
07:25:17MOHDMACH_You mean like this:
07:25:19MOHDMACH_<bluevol.bmp|solid background.bmp>%?mv2.5
07:25:34JdGordonMOHDMACH_: read the CustomWPS wiki page...
07:25:36 Join _lifeless [0] (n=lifeless@
07:25:40 Quit __lifeless (Read error: 104 (Connection reset by peer))
07:25:49Bawoh my mistake i heard it referred to like that i think
07:26:30JdGordonBaw: we never claimed to have every accessory working.. so if yours isnt in the wiki page its likely it just isnt supported
07:26:46JdGordonand by supported I mean we dont handle it..
07:26:49LloreanBaw: It's the 'accessory protocol' patch. The name "apple talk" isn't really associated with it at all.
07:27:15LloreanIf you can provide more information in a bug report it may be possible to improve support for your device, but you'd need to test what information it gets from the accessory/dock
07:27:41Bawsure how do i help?
07:27:46Bawi looked at the debug info
07:27:52Bawand it all kinda is random
07:27:57Bawfor the view i/o ports
07:28:10Bawthe IAP packet and accessory # keep changing
07:29:22Bawits a 5.5 80gb iPod video and im docking to a panasonic 5.1 surround system in the apple fw it has like a panasonic logo and says connected and the surround system display says connected. With rockbox it just flashes connecting on my surround sound and nothing happens no sound or anything
07:30:45LloreanIt sounds like you've got one of the accessories that actually uploads something to the iPod. That's not supported at all, as far as I know.
07:31:01Bawthe only reason i ask about this is because i saw some huge long forum post with some custom patch that was being made and it got moved into svn on the 25th, i installed the daily build on the 31st to check it out
07:32:53Bawbut should that really matter? its still sending the same info back right?
07:33:09Bawthe custom info being recieved is just ignored
07:33:27saratogaobviously it does matter
07:33:36LloreanI think it's actually possible for accessories to upload code, isn't it?
07:33:56 Join MOHDMACH [0] (
07:34:15Bawupload and execute code from an accessory? not sure but that sounds dangerous
07:34:21saratogai'd be surprised if apple allowed that
07:34:52 Join MOHDMACH__ [0] (
07:35:18LloreanI don't see why they wouldn't necessarily. You'd still have to license everything.
07:35:28LloreanBut it may just be bitmaps.
07:37:15saratogait'd be really ackward to arrange i think, since they probably can't share the PP docs, and I doubt most accessory developers would have the skills anyway
07:37:39saratogathey probably provide something more like a WPS screen
07:38:27lucentUnhelpful: I've got an 8gb fuze, need me to whack something?
07:38:32 Quit Bensawsome (Read error: 110 (Connection timed out))
07:38:54Lloreansaratoga: Well they definitely provide a means to upload a bitmap. I seem to have misremembered what I read.
07:39:05Bawso basically im screwed? can i do anything to help?
07:39:16Lloreanbut there's a "simple remote mode" (which we support) and an "advanced remote mode" (which I suspect we don't, or at least not fully yet)
07:39:30Unhelpfullucent: thanks, but i've been lazy since the last test patch that kugel tried already. if you want, though, you can see if you get this corruption that he saw:
07:39:50LloreanSimple basically just lets it pass certain button presses. Advance allows the hardware to interrogate the ipod for all kinds of information, perform playlist manipulation, etc.
07:41:35lucentUnhelpful: I'm going to go ahead and (re: your comment about the Wiki) add myself and Kugel to the page
07:42:45lucenthe should have received the Clip I sent, or will receive it soon
07:43:13BawLlorean is there any kind of information i can provide that would help out the enhancing of this protocol support? or a custom build with extra data collection for this or something
07:44:03LloreanBaw: It's more a case of work needing to be done, I think, by someone who owns such an accessory, than simply information gathering.
07:45:11Unhelpfullucent: thanks. if you do see the corruption, it seemed to come in with my large batch scaler update at r19592 - so r19591 should work
07:45:18Bawwhere do i start? o.O
07:45:37LloreanBaw: Do you know C?
07:46:19LloreanThen you should probably look at the existing accessory support, and possibly write a plugin that allows you to attempt to communicate with your dock in various ways.
07:47:17lucentUnhelpful: it may be delayed before I try this, due to an accident with an 'rm -fr' command as root and an extra space character
07:47:21*lucent :(
07:47:59lucentI'll remember to ask you again once I'm running with a devel environment again :)
07:48:07Bawthis would work for some other docks though right, so there is some support it just doesn't work with my complicated dock
07:48:24Unhelpfullucent: oh, no... i hate it when that happens...
07:48:24LloreanBaw: There's several different modes for using the protocol.
07:48:35Bawah ok
07:48:36lucentBaw: the docks use a serial protocol that is undocumented
07:48:41LloreanBaw: As I said, your dock probably uses the advanced mode, which may not be well (or even at all) supported yet.
07:49:23Bawwell it has to be documented somewhere lucent lol, its just very very secret
07:49:34lucentbaw: undocumented.
07:49:44 Quit MOHDMACH_ (Read error: 110 (Connection timed out))
07:49:47 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
07:50:21lucentif you want to document the protocol, it would be helpful
07:50:49 Quit MOHDMACH (Read error: 110 (Connection timed out))
07:50:51lucentfigure out how and then write a document
07:51:19saratogai'd probably look into get a serial port snooper, or making one, or rigging another ipod to act as a recorder, and then see what the OF and Rockbox do
07:51:29saratogabut thats just how i tend to approach problems like this
07:52:11scorche|shBaw: no amount of googling will magically make the work you think we didnt know about previous efforts?
07:52:37Bawlmao no but there is information on it
07:52:43Bawa lot
07:53:00Bawthat looks fairly well documented
07:53:13scorche|shcould you implement this based on this information?
07:53:41scorche|shthen please dont suggest that we could when you lack a full understanding of these matters
07:53:43MOHDMACH__JdGordon: is there a difference between an "identifier" and an "image id"
07:53:58Bawwhen did i suggest that you could?
07:54:12lucentladies, please shut the fuck up
07:54:26scorche|shlucent: no need for the language...
07:55:41lucentwhat can Baw do to help rockbox, anything?
07:56:27Bawnot at the moment it looks like
07:56:28Lloreanlucent: As has already been said, pretty much anything that either increases the existing knowledge or adds to the existing functionality.
07:56:32Bawi understand the issue a lot better now
07:57:05LloreanEither analyzing what his dock does, or implementing commands that are documented but not coded yet.
07:57:31lucentBaw, thanks for stopping by the developer channel. If you want, check into what saratoga mentioned about a serial port snooper, it really is a good place to begin (and has nothing to do with Rockbox)
07:58:02krazykitlucent, this is both a development AND support channel
07:58:24*lucent :P
07:58:27scorche|shand snooping the serial port for rockbox implmentation is perfectly ontopic
08:00:04 Quit BHSPitLappy (Remote closed the connection)
08:00:47 Quit Rob2222 (Read error: 104 (Connection reset by peer))
08:01:20lucentUnhelpful: thanks again. I've committed the TWiki revision with myself as a fuze owner and have added kugel as a Clip owner
08:05:01saratogahow much more does the HD16xx need before it could be a supported target?
08:06:02saratogapage says it has sound, display, buttons and disk
08:06:38*JdGordon is very confued by 9743 :/
08:06:55JdGordonsaratoga: probably just lack of devs.. or it would be already
08:06:56Lloreansaratoga: Manual, rbutil integration, dual boot or USB?
08:07:47 Join Rob2222 [0] (
08:08:37 Quit Rob2222 (Read error: 104 (Connection reset by peer))
08:11:16 Quit saratoga ("CGI:IRC (EOF)")
08:13:04tmztBaw: the logo might be sent as part of the protocol, I remember reading something about that on mp3car for an interface being developed
08:14:54Bawyeah it is, but there is more to the connecting hand/shake that isnt working then just the photo
08:14:55 Join Rob2222 [0] (
08:15:36Bawthe dock is waiting for some kind of a response back from iPod, its hard to tell without watching the send/recv communication
08:16:32Unhelpfuli wonder how much of the necessary snooping could be done from within rockbox? it may be enough to start with logging unsupported accessory commands, and then implement those commands, to the extent that thi ipodlinux docs allow, and see where that gets us
08:16:48Unhelpfulbut it still depends on somebody with the device and willing and able to put in the work.
08:17:09 Join gartral [0] (n=Gareth@
08:17:10JdGordon... not if the accessory is expecting answers for its commands
08:17:13Bawwell u could see in rockbox all the strings sent but that really doesn't do much good i would think if you don't know what to send back
08:17:19JdGordonwhich is proably would if its sending anything important
08:17:44Bawthe dock connector is basically a serial connector right?
08:20:25Unhelpfulat least *some* of the commands seem to have meaningful docs. there may not be any devices where we can figure out enough without watching retailOS talk to them, though :/
08:20:43gartrali found a bug with the fix to fs9742: after quiting the Version screen, the status bar disappears until either A) you go through to settings>general setting>display>status-/scrollbar> and reset it, or B) the backlight turns off
08:21:39gartralBTW its a sansa e250 v1, running r19639
08:23:07 Join webguest73 [0] (n=44282cb2@gateway/web/cgi-irc/
08:23:08gartralalso, just wondering, why is "status-/scrollbar" instead of "status/scrollbar"?
08:23:20webguest73Is there every going to be Sansa m200 firmware?
08:23:31JdGordongartral: works fine here.. takes maybe .5s to show up...
08:23:43scorche|shif people work on it...perhaps
08:23:51webguest73Is anyone working on nit?
08:24:04webguest73I really like my ogg collection...
08:24:21scorche|shcurrent progress should be in the forum thread for that device in the new ports forum
08:24:27gartralit wont for me, and im running your build, not my own, as my new system for building's proc died the day i turned it on
08:24:48JdGordonwhich target?
08:25:08gartralsansa e250, as i said above
08:26:21gartralok, so, somtimes it comes back, sometimes it soesnt
08:26:31gartraldoesnt* even
08:27:07 Join J-23 [0] (
08:30:40JdGordonhmm... it seems to be only coming out of plugins so im not too woried
08:31:43 Join Rob2223 [0] (
08:32:03 Quit Rob2222 (Read error: 104 (Connection reset by peer))
08:33:04ameyerJdGordon: I have another one. Lamp.
08:33:54JdGordonseems to be all plugins
08:34:12JdGordonnot sure why, and sick of these fixes so its somethng for the weekend or monday
08:34:29JdGordonoh, lamp is fubar
08:34:54JdGordonand batt bench on resume
08:38:11ameyeryeah, that too. I *think* that's the last two.
08:42:16 Quit Rob2223 (Read error: 104 (Connection reset by peer))
08:42:18 Join Rob2222 [0] (
08:47:59***Saving seen data "./dancer.seen"
08:48:15 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
08:49:09 Quit Rob2222 (Read error: 104 (Connection reset by peer))
08:49:25gartral just wondering, why is "status-/scrollbar" instead of "status/scrollbar"?
08:49:34gartralis it*
08:50:42 Join Rob2222 [0] (
08:50:47gartralok, i understand its a "bar glyph" but it looks inconsistent...
08:51:31JdGordonthat text is probably 4 years old.. if not older...
08:53:25MOHDMACH__In my last two attempts I have tried the following two codes at the end of the .WPS file and it still did not produce the desired effect:
08:53:51MOHDMACH__and this one:
08:56:21UnhelpfulVl specify a viewport without actually displaying it. what you want to put in the condition are Vd tags
08:57:47JdGordonand the font is 0 or 1... look at examples... almost every wps avilable will be using those tags
08:58:27MOHDMACH__The viewport must be displayed when adjusting the volume otherwise the background will be displayed.
08:59:00UnhelpfulMOHDMACH__: but, what you need to put inside the %?mv tag are %Vd tags, not %Vl tages.
08:59:43 Quit Rob2222 (Excess Flood)
08:59:58MOHDMACH__I will try that.
09:00:19Unhelpfulwhen you use a %Vd tag, you need to first specify the matching %Vl tag
09:00:21gartralMOHDMACH__: what device is that WPS for?
09:00:36MOHDMACH__Ipod 5g
09:02:06MOHDMACH__When you were referring to the font I don't understand what it means.
09:02:40UnhelpfulMOHDMACH__: you do not use the filename of a font in a %V or %Vl tag
09:03:02Unhelpfuljust a 0 or 1, to indicate whether the viewport uses the "system" font or the one selected by the user.
09:03:36amiconnUnhelpful: I think that the gfx glitches on the fuze are somehow storage driver related
09:03:40MOHDMACH__So 1 means selected by user.
09:03:43 Quit J-23 ("Flying cow pressed ^D on my keyboard.")
09:04:05 Join J-23 [0] (
09:04:16Unhelpfulamiconn: any idea what i "did" to trigger them?
09:04:43amiconnYou changed the timing of the bmp loader.
09:05:12amiconnI think this points to some timing related bug in the storage driver.
09:05:19Unhelpfulah, i see. fixing that will be a bear. :/
09:05:19*webguest73 is meh
09:05:27*webguest73 waves goodbye.
09:05:29webguest73Bye people.
09:05:31 Quit webguest73 ("CGI:IRC")
09:05:38 Join Nibbler [0] (
09:06:12amiconnI could be something else, but I fail to think of a reason why code like the bmp loader could fail on just some targets, while other targets with the same pixel format and same endianess are working fine
09:07:01amiconnHmm, does it only happen for certain scaling factors perhaps?
09:07:37Unhelpfulwps and background bitmaps both before and after that revision are not scaled.
09:07:42gartralare we talking about why pic flow sometimes draws lines through part of a cover when it renders?
09:09:15ameyeras far as I can tell, no.
09:10:05Unhelpfulgartral: no, not at all.
09:10:28Unhelpfulalso, lines? and are you using the newest revision?
09:10:32 Join moos [0] (
09:14:00MOHDMACH__That attempt also did not work so if anyone wants to have a look at .wps file -
09:14:56*gartral disassembles his dap to get the little peice of dirt thats settled in the dead center of his screen
09:17:25Lloreangartral: That sort of thing is pretty much off-topic here and better simply left unsaid, please.
09:19:10gartralit also was ment for a diff schat, lol, sorry
09:20:16 Join Rob2222 [0] (
09:24:31ameyerJdGordon: for whatever reason, the sluggishness isn't nearly as bad on r19640. You may not need to tweak it.
09:24:31 Quit Rob2222 (Read error: 104 (Connection reset by peer))
09:30:49gartralhmmm... rapidy skipping back or foreward breaks AA display, and causes the WPS to report the wrong album
09:31:53JdGordonso dont do that :p
09:32:43gartralok, definitio of "rapid": skipping 3-4 songs without listening to the intros
09:33:11amiconnJdGordon: When returning from the credits plugin, it takes a long time until the status bar reappears
09:33:33JdGordonamiconn: yeah, returning from any plugin causes that.. im not sure why
09:34:12 Join MOHDMACH [0] (
09:34:50Unhelpfulpossibly the reappearance is happening when the status changes?
09:35:21JdGordonno.. after exiting plugins it should force a statusbar redraw...
09:36:06gartralit just... lags in doing so
09:36:32gartralgotta remember to hit alt tab.... really gotta remember that
09:37:27 Join MOHDMACH_ [0] (
09:39:05gartralok, anyone know if the chip8 plugin works with superchip roms?
09:40:11 Nick JdGordon is now known as JdGordon|afk (n=jonno@rockbox/developer/JdGordon)
09:42:51Unhelpfulgartral: i think i recall reading that the answer is yes, but you should probably check the manual or wiki
09:43:14gartraleasier solution, load one of each and see if they work
09:44:03gartralodd... both report "undiffined chip instrution"
09:44:46MOHDMACH_Thanks for the help.
09:44:54 Part MOHDMACH_
09:44:55gartralbye bye
09:46:35 Quit kachna (Read error: 60 (Operation timed out))
09:48:13gartralchip8 dosnt respond to key inputs... at all
09:49:57 Quit MOHDMACH__ (Read error: 110 (Connection timed out))
09:53:03 Quit MOHDMACH (Read error: 110 (Connection timed out))
09:53:10gartralohh, nvm, i broke the keymap in the file
09:55:07amiconnJdGordon|afk: Why does solitaire not have the status bar in its menu anymore?
09:55:10*Unhelpful can not imagine how he expected those %Vd tags with just an identifier to do *anything* without a matching %Vl
10:01:04 Join Horschti [0] (
10:01:08*ameyer mutters something about r19622 not being ready for prime time
10:01:18 Quit Horscht (Nick collision from services.)
10:03:47Lloreanameyer: It's not like it's a release... It's a development build.
10:04:02JdGordon|afkamiconn: the bar has been disabled for all plugins temporarily.. i forgot to mention it in the commit message... they can renable it individually
10:04:07*JdGordon|afk really gone now
10:04:12LloreanIf you're using the dev build you're a tester, so you shouldn't expect everything to always be "ready for prime time"
10:13:59*ameyer is just impressed by the extent of the breakage
10:16:53 Quit _lifeless (Remote closed the connection)
10:17:09 Join _lifeless [0] (n=lifeless@
10:22:32amiconnLlorean, JdGordon|afk: Sure latest SVN isn't a release. But in such a case I would expect (a) a mention of the breakage, and (b) expect the complementing fix asap.
10:23:18Lloreanamiconn: I'd expect a fix ASAP, but that doesn't necessarily mean immediately.
10:23:31splinterhi, i installed for the first time. looks fine. when i plug it in, i get the 'do not disconnect' window and it wont go away.
10:23:38LloreanThere's not really a location for a dev to "mention" a breakage, anyway
10:23:57amiconnAhem, in the commit message, perhaps?
10:24:17LloreanIt was expected to work when it was committed, I thought.
10:24:33amiconn[10:04:02] <JdGordon|afk> amiconn: the bar has been disabled for all plugins temporarily.. i forgot to mention it in the commit message... they can renable it individually
10:24:41LloreanEven with the non-release state, I wouldn't expect intentional commits of flawed code.
10:25:39ameyerand I think the dynamic playlist boosting bug fixed in r19304 (FS #9319/FS #9576) is back. Probably should be going to bed, I'll try to track it down later.
10:27:44splinterit's like it's trying to sync with something?
10:28:12Lloreansplinter: That screen is handled by the original firmware. If your computer says it's ejected, it's safe to disconnect.
10:28:38splinterit's the rockbox font
10:28:52splinterdoes it just always say that when you plug it in?
10:29:07Lloreansplinter: It's not the Rockbox font.
10:29:27splinterit's not the ipod font either
10:29:39LloreanYes, it is.
10:29:59LloreanIt's just one of their other fonts.
10:30:07splinterdoes it just always say that when you plug it in?
10:30:19LloreanAny time it tries to go into USB mode, yes.
10:30:27ameyerunless you installed a semi-mangled unofficial build, it's the ipod font, as rockbox doesn't handle usb on portalplayer targets.
10:30:44Lloreanameyer: Rockbox's USB screen doesn't say that anyway.
10:31:16splinternah just 3.1
10:31:18ameyerI find that I can unmount my ipod without causing the ipod to leave the "do not disconnect" screen
10:32:24splinterwhen i plug it in, it reboots to that screen, is that what it is supposed to do? unless i press the button that keeps it in 'charge only' mode?
10:33:49ameyerSupposed to? Yes. Does it always? No.
10:35:03amiconnLlorean: That's what I'd expect too. Hence I have to say that was a typical JdGordon commit, sorry :\
10:37:41 Quit film42 ()
10:47:45 Join ibseo [0] (
10:48:00***Saving seen data "./dancer.seen"
10:50:05Unhelpfulamiconn: a fairly large timing change in that commit, i think, would be the change in the max bmp chunk size from 8px to LCD_WIDTH...
10:50:50Unhelpfulpreviously, it was "stopping" after reading 8px worth to convert that to the native bitmap
10:51:39amiconnHmm? Iirc the bmp reader used to read a complete line at a time
10:52:27*amiconn would ike to know whether the gfx glitches on the fuze are reproducable, or whether they are different on each run
10:52:40amiconnIf they're fixed, a timing issue is rather unlikely
10:53:04amiconnBut then why don't other targets see this? Perhaps it's a matter of screen width?
10:53:27 Join robin0800 [0] (
10:53:40Unhelpfulthose are from two different builds - i had him force dither off, since i thought that might be part of the problem
10:54:40*amiconn should probably try on his ipod color (same screen width), although he outright hates cabbiev2
10:54:51Unhelpfulthe defects look pretty much the same, to me
10:55:53 Quit ze (Read error: 110 (Connection timed out))
10:56:58amiconnHmm, my unfinished lcd-16bit work stops the problem from showing up
10:57:21Unhelpfulbut svn is broken on the target for you?
10:57:43amiconnDidn't try yet - just downloading
10:59:06amiconnPlain SVN works for me
10:59:48soapBaw, it was not MY patch. Give credit where credit is due. I just made a build with it to get people to test it.
10:59:52Unhelpfuli'll have kugel try changing BM_MAX_WIDTH when he's available next
11:01:08Unhelpfuloh, i meant to ask... are FORMAT_NATIVE rgb565 bitmaps always in host order?
11:01:27amiconnWhat I'm working on in lcd-16bit.c is support for opaque background when a background bitmap is loaded, selectable via an extra drawmode bit. Right now it's broken, i.e. uncommittable
11:01:41 Join ajb [0] (
11:02:01amiconnipod color uses RGB565SWAPPED though
11:02:06Unhelpfuli noticed that the bmp loader doesn't use a packing macro, just some shifts and ors
11:05:10 Join ze [0] (i=ze@
11:05:50 Join kachna [0] (
11:06:07Unhelpfulhrm... and i just realized that on RGB565SWAPPED targets, pictureflow.c passes pixel values through swap16 before doing channel-wise math on them.
11:07:20 Nick Barahir_ is now known as Barahir (
11:18:31amiconnUnhelpful: Yeah, as it needs to be
11:19:58Unhelpful... i'm a little confused, then? it would appear that pictureflow assumes in-memory bitmaps and the framebuffer to be swapped on these targets, while the loader that produces these bitmaps does not.
11:20:45Unhelpfulhrm. the "old" bmp.c does not appear to work this way. wonder if that's my bug :/
11:21:17amiconnI'm not sure what you mean, then
11:21:20Unhelpfulby which i mean a completely different bug from the one kugel sees, which i may have introduced :/
11:21:53amiconnRGB565SWAPPED means the low and high bytes are swapped in memory (that is optimized for how lcd_update has to transfer the data)
11:22:48Unhelpful...i'm no longer sure what i mean, either, i could swear i was looking at it and it read (r << 11) | (g<<5) | (b << 6)
11:22:56Unhelpful... rm, just | b obviously.
11:23:41amiconnYou said there's a swap going on in pictureflow.c, which then can't be the bmp loader. So I assume there's some calculation going on that uses an in-memory bitmap as the source
11:24:14amiconnAnd such a calculation, needs to pre-swap and post-swap if it needs to operate on the individual colour channels
11:24:38Unhelpfulsorry, i was looking at the preprocessed bmp.i on fuze, which isn't using swapped... bmp.c uses LCD_RGBPACK_LCD
11:26:03Unhelpfulthe source of the confusion was that i thought that the bmp loader was *not* swapping.
11:31:48ajbUndefinned instruction @ 40000184 (0) doesn't sound good
11:33:45ajbLooks like there should be more info dumped for a unhanddled exception
11:34:07 Quit abo_ (Read error: 145 (Connection timed out))
11:34:16amiconnJdGordon|afk: rockbox_flash.rock is broken...
11:35:33amiconnIt returns immediately
11:37:26 Join n1s [0] (n=nils@rockbox/developer/n1s)
11:38:09*ajb wonders why a VA_ARGS macro is wrapped around a format string
11:41:38 Join {phoenix} [0] (
11:42:49amiconnJdGordon|afk: And your fix fir credits.rock has at least one significant bug. *Never* compare ticks using < or >, always use TIME_BEFORE() or TIME_AFTER()
11:44:22ajbWhat offset if rockbox.elf loaded into memory at on the iPod Video (5.5g)?
11:45:07amiconnrockbox.elf isn't loaded at all, it's not even uínstalled on target
11:45:31amiconnWhat is loaded is rockbox.ipod, which is essentially rockbox.bin with a header
11:45:57ajbwhat will objdump want to dissasemble it?
11:46:03amiconnrockbox.bin in turn is built from rockbox.elf. You can check addresses in the linker script or the .map file
11:48:13ajbhmmm, according to the map the address is bogus, so it must of jumped off into the tther for some reason
11:48:39*ajb wishes for a backtrace
11:48:58 Quit _lifeless (Read error: 131 (Connection reset by peer))
11:49:00 Join _lifeless [0] (n=lifeless@
11:49:35amiconnThe address is not bogus. 0x4000xxxx is in IRAM
11:49:56Unhelpfulamiconn: regarding the read sizes, old bmp.c was reading whole lines. the original resize-on-load was reading whole lines into a static buffer, and failing if they exceeded a maximum size.
11:50:13 Join culture [0] (
11:50:50Unhelpfulwhen i reworked it to read partial lines, so that the static buffer could go away, i arbitrarily chose 8 pixels, as the smallest size that could be read in whole bytes regardless of the input bmp file depth
11:51:43 Join nplus [0] (n=nplus@141.25.Globcom.Net)
11:51:47Unhelpfulthe "problem" revision doesn't change very many things inside bmp.c, but it changes BM_MAX_WIDTH to LCD_HEIGHT rounded up to the nearest multiple of 8, so the read size is *much* larger on most targets than in the initial scaled loader commit.
11:52:20ajbamiconn: but will it be rockbox code or a bad jump?
11:52:58amiconnRockbox code, almost certainly
11:54:20Unhelpfulwhich code do we load into IRAM?
11:55:36n1sUnhelpful: some of the most time critical code in core/codecs generally
11:56:16 Quit ameyer (Remote closed the connection)
11:57:21Unhelpfuli suspected something like that might be the case... is it possible his address isn't in the map because it's dynamically loaded code, ie from a codec?
11:58:32 Join ender` [0] (
11:59:01n1sstatic functions don't show up in the map for example, codecs have their own maps so it could be there
12:03:41amiconnActually it might be bogus. I don't have a G5.5 build handy, but on the Color, .ibss comes first in IRAM, and this address would be in .ibss
12:06:24 Join gromit` [0] (
12:07:39 Quit _lifeless (Remote closed the connection)
12:09:55 Join _lifeless [0] (n=lifeless@
12:10:25 Quit _lifeless (Client Quit)
12:21:15 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
12:21:20 Quit {phoenix} (Read error: 110 (Connection timed out))
12:22:09Unhelpfulkugel: whenever you're around again, try changing BM_MAX_WIDTH in bmp.h. it was set to 8 before that revision, any multiple of 8 should be valid without breaking anything.
12:28:13 Quit soap (Remote closed the connection)
12:32:12 Join {phoenix} [0] (
12:33:24 Join bytie [0] (
12:33:32 Quit bytie (Remote closed the connection)
12:33:54 Quit robin0800 (Read error: 60 (Operation timed out))
12:34:49 Join merbanan [0] (n=banan@
12:34:56 Join robin0800 [0] (
12:40:50gartralwait, soo, as of right now, what does not work?
12:42:12n1syou may want to be a little more specific :)
12:48:04***Saving seen data "./dancer.seen"
13:02:15ajbHow do I identify which codec is loaded at that address
13:03:15n1sajb: all codecs are loaded to the same adress but as amiconn said that adress seems to be in ibss (not code)
13:04:29*ajb continues trying to repeat the crash
13:05:58ajbwell that was unreasuring
13:06:14ajbI just switched to a new song
13:06:36amiconnThis address is not in codec/plugin iram, but in core iram
13:08:27n1sin a local build it seems voice_tread ibss section begins at 0x40000180
13:08:40 Join flydutch [0] (
13:14:38 Quit ibseo (Read error: 113 (No route to host))
13:15:49 Join kugel [0] (n=chatzill@unaffiliated/kugel)
13:16:07ajbCan I get more info out of the crash now I can reproduce on my hardware?
13:20:05n1snot sure, what do you do to reproduce it?
13:20:30ajbSelect a particular track (which I think is a m4a)
13:20:40ajbI assume it crashes as it attempts to play
13:20:53n1sdoes it crash a simulator?
13:21:06kugelJdGordon|afk: I was wondering about droppen the support for passing NULL, so lists/menu wouldn't have to bother with reinits at all
13:21:23ajbnls: no
13:22:23n1sajb: so does it play ok in a sim?
13:22:27 Join Domonoky [0] (n=domonoky@rockbox/developer/domonoky)
13:22:30 Part Domonoky
13:23:14ajbnls: that's correct, it works fine in the sim.
13:23:57n1shmm, that makes it a bit tricky, can you file a bugreport and attach the file?
13:24:27ajbn1s: Attaching the file is tricky as it's a copyright track
13:25:30 Join AndyIL [0] (i=AndyI@
13:26:16n1sah, yes
13:27:59 Join faemir [0] (
13:29:37 Join MethoS- [0] (
13:29:55kugelUnhelpful: will do
13:32:55ajbFS #9745 raised (crash on m4a)
13:33:21*ajb tests on latest svn
13:34:50write__eraseHi! what's the difference between −−add-bootloader and −−add-bootloader-bin ? thx
13:35:23n1sajb: does it happen with any m4a or just a specific one?
13:35:26 Join t0mas [0] (n=tomas@rockbox/developer/t0mas)
13:36:07*ajb tests a few more
13:37:52 Join stoffel_ [0] (
13:38:01 Quit AndyI (Read error: 110 (Connection timed out))
13:41:13 Join tyfoo [0] (
13:42:52kugelUnhelpful: and
13:45:22 Join miepchen^schlaf [0] (
13:47:21 Quit MethoS- (Remote closed the connection)
13:47:48 Join MethoS- [0] (
13:47:57ajbn1s: hmmm, well some other m4a's all skipped and failed to play but didn't crash
13:48:37ajbn1s: The playlist shows "(ERR) song name"
13:49:30pixelmais there something special about these m4as (specific format, tags, length)?
13:50:10ajbNot as far as I can tell. Do we have a reference m4a encoded file for comparison/testing?
13:50:35ajbhmm, different crash but still a crash from an m4a
13:51:55 Quit MethoS- (Remote closed the connection)
13:52:15 Join MethoS- [0] (
13:54:08 Quit MethoS- (Remote closed the connection)
13:54:44 Join MethoS- [0] (
13:56:19t0masBagder! long time no see :-)
14:00:57 Join bertrik [0] (
14:01:46 Join Thundercloud [0] (
14:11:34 Join ajonat [0] (n=ajonat@
14:12:11*ajb grabs
14:13:46moost0mas: welcome back, long time no see indeed :)
14:14:06ajbWell that m4a seems to hang the player
14:14:33t0masmoos: sorry! (A)
14:15:15gartralin pictureflow, it shows an entry for "Untagged", the only untagged files are those i recorded...
14:15:42moosstill have your iriver h1xx? :)
14:16:29t0masI do.. but don't use it anymore
14:16:45t0mashave been using an ipod nano since ehm.. 2 years or something
14:17:29moostime to change yet then ;)
14:17:35t0masand I'm looking into getting a new player now.. with rockbox support.. problem is that I sort of need an ipod for car connectivity etc.. and there are no ipods currently for sale that run Rockbox
14:17:47 Quit stoffel_ (Read error: 113 (No route to host))
14:17:48ajbpixelma: n1s: Ok, it seems all m4a's either crash or hang which I assume most people are not seeing?
14:18:18 Join ze0 [0] (i=ze@
14:18:32 Quit ze (Read error: 131 (Connection reset by peer))
14:19:35 Nick ze0 is now known as ze (i=ze@
14:19:39ajbohh new crash
14:19:54n1sajb: i would guess we would get quite a lot of bugreports if m4a's didn't work for anyone, try installing a new build, scan the disk for errors, reset settings etc
14:19:56ajbUndefined instruction at 40017F30 (0)
14:20:08gartralt0mas: i found some on ebay...
14:20:27gartraloops, didnt copy the turl
14:21:08t0masI did too.. but I want to replace my nano because the battery is degrading, which is normally a problem with older ipods :-)
14:21:39gartralso buy a new battery
14:22:49ajbn1s: I turned off Cross Fade and now the crash is somewhere else in memory
14:23:17 Join Febs [0] (
14:23:20t0masgartral: that costs almost as much as a new ipod.. and the nano is a 4gb scratched old ipod..
14:23:23*ajb uploads some test files
14:24:02gartralwhat gen is your current one?
14:24:36t0maseeeehm.. first
14:24:52gartral1st what
14:25:18gartralnano, or mini?
14:25:26t0masit's this one:
14:26:00gartral$20 US
14:27:10t0mas"Unlike the other iPod models, the iPod nano is one of the most challenging to open and repair"
14:27:19t0masand I really want a new one ;-)
14:27:23t0masjust for the gadget factor..
14:27:57gartralyou can add their toolkit for an extra 10
14:29:38gartralor, if you'd rather not do the work, i could do it :P
14:31:01t0mashm.. you do make it attractive ;-)
14:31:14n1st0mas: it's not terribly hard to open just putting it back together so that it looks like it wasn't opened is a bit tricky, also the battery is soldered in iirc
14:32:12t0masok.. other approach..
14:32:34t0masI've read about some progress getting Rockbox running as an application on the ipod touch..
14:33:09t0masdoes somebody know who did that?
14:33:18gartralyou could also have them do it, and that also gets you a re-replacement warrenty
14:33:19t0masmaybe I can try to help (A)
14:34:31gartralthats somwhere between 12-30 bucks, depending on shipping speed
14:35:41gartralthere running a special, free shipping on all orders either for repair, or it its over 10 bucks
14:37:06ajbn1s: I've uploaded a test file to the bug
14:38:06t0masanother option is the Creative Zen.. and then spend the price difference on changing my car-connector from ipod to usb.. and get back into rockbox development by helping to port Rockbox to the Zen..
14:38:10 Join stoffel_ [0] (
14:38:36 Join casainho [0] (
14:39:00casainhohello :-)
14:40:04*Febs cannot get iriver_flash to work on his H120.
14:41:31n1sajb does it work with reset settings?
14:44:17FebsI'm using bootloader 7pre3. I have r13749 in ROM. (Yeah, it's been a while since I upgraded.) I can boot r19642 from disk. But I cannot flash either bootloader.iriver or rombox.iriver with iriver_flash.
14:46:00gartralok, im trying these M4As on my sansa e250
14:47:56gartralok, other than making it laaaaaag after a button press, it does play
14:48:08***Saving seen data "./dancer.seen"
14:48:12casainhocan someone help me? : I am having a problem on my code, on kernel_init(), I am getting all the time data abort interrupts, I saw that "cores" struct get everytime an invalid memory adress... does anyone knows what problem can be?
14:48:16ajbn1s: I did a fresh install, default settings. Crashes at the new address (but I suspect it's just switching off CrossFade that affects that)
14:48:34ajbn1s: Has anyone built a QEMU simulator for rockbox?
14:49:15n1sajb: ok, no i don't think anyone has made an emulator that can fully emulate any of our targets
14:49:47n1sthe attached test file works fine on my gigabeast both with crossfade and without
14:50:00casainholinuxstb: can you help me?
14:50:22ajbn1s: Do the gigabeast and ipod share the same codec code?
14:50:27casainhomarkun: can you help me?
14:50:35gartralohh... i haddnt noticed, but these M4As dont crossfade, with, or without CFing active
14:51:29n1sajb: depends on how you mean, most of the apps code is shared but they use different SoC's so the binaries will differ quite a bit
14:51:48ajbn1s: is it all compiled C or hand written asm?
14:51:57n1sa bit of both
14:52:07n1smostly compiled c though
14:52:24casainhoamiconn: can you help me please?
14:52:28ajbWell if I could decode that address will it tell me somewhere...
14:52:33*ajb checks map again
14:52:51casainhokugel: are you there?
14:53:57 Quit n1s ()
14:54:14 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
14:54:15kugelcasainho: sorry, I can't help you, I'm not into kernel stuff at all (yet)
14:54:38 Join n1s [0] (n=nils@rockbox/developer/n1s)
14:55:03 Quit n1s (Client Quit)
14:55:41 Join itcheg [0] (i=41d59de2@gateway/web/ajax/
14:56:40casainhokugel: ok, but this question: can you please look at this picture? :
14:57:26 Join tvelocity [0] (
14:57:34 Join n1s [0] (n=nils@rockbox/developer/n1s)
14:57:45casainhokugel: the address of cores.running is wrong!!! and I saw that righ at my bootloader main(), it already have that wrong value...
14:58:11kugelare you sure your is correct?
14:58:41casainhokugel: yes, it's correct, I have seen my arm-elf-objdump -h bootloader.elf
14:59:00kugelno idea then, really
14:59:09casainhoand there I see clearly the VMA and LMA adresses, also I can see that on
15:00:02kugelyou should concentrate on getting the main() address right I suppose
15:00:33casainhokugel: being the cores.running global, what defines it's address? the compiler?
15:01:26kugelI don't know dude...
15:01:56kugelI have neither enough knowledge nor enough time to help you in this stage
15:02:06casainhokugel: okok :-) thanks :-)
15:02:21kugelbut if your main address is wrong, all others address are wrong too I'd think
15:02:41 Quit stoffel_ ("leaving")
15:02:59casainhokugel: main() have adress? I didn't know....
15:04:18casainhokugel: I am getting that global variable and other of that struct with wrong memory address values...
15:05:57 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
15:11:53 Quit CaptainKwel (Read error: 104 (Connection reset by peer))
15:12:05 Join CaptainKwel [0] (
15:27:57casainhobertrik: are you there?
15:28:56 Part gartral
15:35:56 Quit Febs (Read error: 54 (Connection reset by peer))
15:36:42 Join MethoS-- [0] (
15:40:27 Quit peter__b (Read error: 104 (Connection reset by peer))
15:44:20 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
15:46:58 Quit ajonat ()
15:50:17 Quit MethoS- (Read error: 110 (Connection timed out))
15:51:16 Quit kachna (Read error: 60 (Operation timed out))
15:55:21 Quit {phoenix} (Read error: 145 (Connection timed out))
15:58:00 Join ibseo [0] (
16:01:37 Join Veruz [0] (
16:02:59VeruzI have a problem with my iHP 120 player
16:03:14Veruzwhen I start it I get an ATA error (-32)
16:03:20VeruzI use rockbox
16:03:48Veruzand I cannot access my hardisk by USB anymore
16:04:13VeruzDoes anyone know where this problem comes from?
16:04:36Veruzis my mp3 player dead?
16:06:20 Quit t0mas ("switching pcs")
16:06:24gevaertsCan you still boot the original firmware?
16:06:59Veruzyes but with a lot of problem
16:07:19VeruzI dont have any files on my disk anymore
16:07:23VeruzI mean media files
16:07:40Veruzthen I can load the default settings but it doesn´t change anything
16:08:15gevaertsOK. That's definitely a hardware issue then. I think either the disk is bad, or a connector has come loose
16:08:49VeruzI can go on the menu settings of my original firmware
16:08:58Veruzbut there is nothing I can do there
16:10:05Veruzwhen I plug in the player on my computer by USB I can hear the sound of new hardware detected but cannot see it in my window explorer
16:10:39Veruzit says that it´s connected on my player too
16:11:03rasherIs the iAudio M3 a release target or not? It's listed on the front page, but not on the 3.1 release page...
16:11:33gevaertsrasher: it isn't. No manual
16:11:44VeruzI tried to reset the player using the reset button but it looks the same
16:12:34 Join LambdaCalculus3_ [0] (
16:12:38rasherSo it's... supported, but not a release target? Ew.
16:13:05Veruzis it possible to remove rockbox firmware directly on the mp3 player, without using a computer with USB?
16:13:20Veruzand load the original default firmware with its settings
16:13:23gevaertsThat won't help you.
16:13:56gevaertsrasher: indeed. We need to think about this a bit more
16:14:02Veruzany way of repairing the hardisk or connector?
16:14:35gevaertsYou can open it and re-seat the connector. If that doesn't help, you can replace the disk
16:14:36VeruzI looked at the rockbox forum but I couldn´t find an ATA error -32 there
16:15:31VeruzIn case of replacing the disk, isn´t that cheaper to buy a new mp3 player?
16:16:44gevaertsDepends. A new disk is certainly cheaper than another H120
16:16:50n1sdepends on which player you want and which disk of course
16:17:08 Quit ibseo ("quit")
16:19:46Veruzhow does a connector look like?
16:20:19 Quit LambdaCalculus37 (Read error: 113 (No route to host))
16:20:58 Nick LambdaCalculus3_ is now known as LambdaCalculus37 (
16:23:53VeruzWhere can I buy a new disk for an iHP-120 player?
16:28:47 Quit casainho (Read error: 113 (No route to host))
16:28:55 Quit moos ("Rockbox rules the DAP world")
16:32:59 Join redwood [0] (
16:35:09 Join troy__ [0] (n=toppy@
16:37:01 Join evilnick [0] (i=0c140464@gateway/web/ajax/
16:38:03 Quit Veruz ("ChatZilla 0.9.84 [Firefox 3.0.5/2008120122]")
16:48:09***Saving seen data "./dancer.seen"
16:48:28 Join ibseo [0] (
16:49:09 Join Schmogel [0] (
16:52:23 Quit troy_ (Read error: 110 (Connection timed out))
16:56:08 Join fdinel [0] (
17:01:58 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
17:02:16 Join {phoenix} [0] (
17:04:07 Join Seed [0] (
17:10:35 Quit jgreco (Read error: 110 (Connection timed out))
17:24:34 Join toffe82 [0] (
17:25:01 Quit Bensawsome (Connection timed out)
17:25:28 Quit {phoenix} (Read error: 60 (Operation timed out))
17:28:02 Join kachna [0] (
17:31:00 Join Bensawsome [0] (n=Bensawso@unaffiliated/bensawsome)
17:31:53 Quit flydutch ("/* empty */")
17:40:06 Join flydutch [0] (
17:45:02 Join soap [50] (n=soap@rockbox/staff/soap)
17:45:45 Join {phoenix} [0] (
17:52:50 Quit robin0800 (Remote closed the connection)
17:53:06 Part toffe82
17:53:08 Join casainho [0] (n=chatzill@
17:53:17casainhohello :-)
17:56:23 Join aurix_lexico [0] (
17:56:39 Nick fxb__ is now known as fxb (
17:58:06 Join robin0800 [0] (
18:06:55 Quit robin0800 (" - Chat comfortably. Anywhere.")
18:07:47 Join Anges [0] (
18:11:07 Join robin0800 [0] (
18:11:11casainhohello :-) is there anyone who can help me with kernel_init()?
18:11:38casainholinuxstb: hello :-) - can you help me a bit?
18:16:15casainhomarkun: can you help me a bit?
18:17:07 Quit scorche|sh ("Lost terminal")
18:18:44linuxstbcasainho: I am just about to go out, so can't help. But you should just ask your question and hope that someone who knows the answer will reply, rather than limit your questions to certain people.
18:19:23casainholinuxstb: thanks. I will do it then. Also I am thinking in writing on mailing list.. or forum shoulkd be better?
18:19:42Bagdercasainho: dev list is probably read by more devs than the forum is
18:19:45 Join spiorf [0] (
18:20:06casainhoBagder: please, do you have 3 minutes?
18:20:23casainhoBagder: ok, I will use then the mailing list
18:20:37Bagder(kids in the bath atm, but that will end any minute...)
18:20:57casainhoso, my probelm: I am having wrong addresses memory being acessed, so, I get data about interrupt!
18:21:21 Join MethoS_ [0] (
18:22:02BagderI don't understand
18:22:02casainhoon kernel_init⁽), after debug, I saw that cores.running are gewting a wrong adress memory!!! :-(
18:22:39 Part redwood
18:23:25casainhoI will put an image shoing debug window...
18:23:30 Join karashata [0] (n=karashat@
18:23:57 Join scorche|sh [50] (n=scorche@rockbox/administrator/scorche)
18:24:08 Join redwood [0] (
18:24:46casainhoon that image, the address value of cores.running are wrong, they don't exist on mhy system then I get data abort interrupt, because of ARM trying to acess at a not available adress
18:25:39Bagderright, so you need to figure out how it becomes wrong
18:26:16casainhoBagder: BUT, using debug, I saw the same value, right on the beguinning of main()!!! :-( :-(
18:26:47Bagderso it sounds something changes it along the way
18:26:50 Nick Horschti is now known as Horscht (n=Horscht@xbmc/user/horscht)
18:27:29casainhocores is a global structure... shouldn't it having the right adress since beginning?
18:28:47 Quit MethoS-- (Read error: 145 (Connection timed out))
18:29:32 Nick fxb is now known as fxb__ (
18:29:34casainhobut, is not just cores.running, other inside cores also get wrong address values
18:31:19Bagderso it sounds like something overwrites the memory of the cores struct
18:32:07bertrikthe 0xe5 in cores[0].running looks like an arm instruction
18:32:27casainho −− here is the image of debug right on the beginning on main()
18:33:38 Quit martian67 (Connection timed out)
18:34:39casainhoBagder: okok... I will have to look at it, so, maybe it's on C startup...
18:35:03casainhobertrik: hmmm, now I remember to see that value on disassembly window...
18:36:34casainhodo you have one idea of what could I do wrong? why code is in place of bss or data?
18:37:52 Join martian67 [0] (i=lol3izer@about/linux/regular/martian67)
18:38:22casainhookok - I will look better at that, and searching for that values at cores to see if they appear on dissasemble file :-) −− many thanks :-)
18:38:26casainhobye bye :-)
18:38:30 Quit casainho ("ChatZilla 0.9.84 [Firefox 3.0.5/2008121622]")
18:38:37bertrikIIRC, the C startup code should clear the bss section and initialise the data section with data from just after the text section in the application image
18:44:42 Quit Nibbler (Remote closed the connection)
18:48:10***Saving seen data "./dancer.seen"
18:52:35 Join Nibbler [0] (
18:53:17 Join fredddy [0] (
18:58:33 Quit ibseo (Read error: 110 (Connection timed out))
18:59:04 Quit spiorf (Remote closed the connection)
19:01:05 Quit itcheg (" ajax IRC Client")
19:03:45 Quit LambdaCalculus37 ("Ka-chunka")
19:07:00 Join tessarakt [0] (
19:10:32 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
19:23:26 Join toffe82 [0] (n=chatzill@
19:26:02 Join Willwolfe [0] (
19:30:36 Join mib_fg71pv7s [0] (i=c6bbc8fe@gateway/web/ajax/
19:31:12 Nick mib_fg71pv7s is now known as MarcGuay (i=c6bbc8fe@gateway/web/ajax/
19:36:20 Quit MarcGuay (" ajax IRC Client")
19:36:59 Join MarcGuay [0] (n=c6bbc8fe@gateway/web/cgi-irc/
19:39:34 Join mib_x0z7nzxy [0] (i=c6bbc8fe@gateway/web/ajax/
19:40:50mib_x0z7nzxyBagder, Zagor: FYI, the manual page still mentions 3.0.
19:40:59 Quit Willwolfe ("ChatZilla 0.9.84 [Firefox 3.0.1/2008070208]")
19:42:26 Quit MarcGuay ("CGI:IRC (Ping timeout)")
19:42:44 Quit mib_x0z7nzxy (Client Quit)
19:45:16 Join excitatory [0] (
19:47:19excitatoryquestion.. i've been installing rockbox on several of my friend's ipods over the last few months.. and well, I absolutely love the automated installer, however, for the past few months it has continued to fail on installing the themes claiming that the Theme Site is still being reworked.
19:47:46excitatorySo my question is if there is any time line as to when this will be completed?
19:48:24Bagderexcitatory: unfortunately, no
19:48:28domonokyexcitatory: we dont have any timelines, and yes the theme site is still under rework...
19:49:43excitatoryalright.. no big deal, just curious really.
19:51:00excitatorylast night i bought a 5th gen 80 gig ipod to upgrade my 1st gen nano.. looking forward to rockbox on a larger display
19:51:26excitatoryand video features
19:51:33excitatorynever got to try that out on the nano
19:54:19splinteri just put rockbox on my 5th gen 30gig
19:54:24splinternever used it before
19:54:44splinterpretty sweet so far :)
19:55:35excitatorysplinter, i suggest using a theme.. the default uses hard to read fonts (imho)
19:56:47splinteri'll have to start from scratch on that, where to get them, how to install them
19:57:08splinterright now trying to get some 20gigs of music back on so i can go out to the shed...:)
19:57:54excitatoryfor your device:
19:59:03splinterah sweet
20:09:24 Part toffe82
20:14:21 Join virtuoso015 [0] (n=vinay@
20:17:38 Quit reacocard ("BRB")
20:27:14 Join gregorovius [0] (
20:28:24 Part virtuoso015
20:29:06 Join reacocard [0] (i=reacocar@
20:30:11 Join Highlander [0] (
20:39:14 Quit MethoS_ (Remote closed the connection)
20:48:11***Saving seen data "./dancer.seen"
20:59:42 Join BHSPitLappy [0] (n=BHSPitLa@unaffiliated/bhspitmonkey)
20:59:54*bertrik wonders about the FOURTHERTZ tick task
21:06:27 Quit CaptainKwel (Read error: 60 (Operation timed out))
21:06:43 Join troy_ [0] (n=toppy@
21:13:43 Join Zagor [242] (n=bjst@rockbox/developer/Zagor)
21:23:09 Nick keanu is now known as [keanu] (n=keanu@unaffiliated/keanu)
21:23:21 Quit troy__ (Read error: 110 (Connection timed out))
21:26:34 Quit fredddy (Remote closed the connection)
21:33:38 Join itcheg [0] (i=62db4767@gateway/web/ajax/
21:37:06bertrikAre there any pictures available somewhere of the insides of a c200v2?
21:42:07 Quit evilnick (" ajax IRC Client")
21:42:13 Join _Waheed [0] (n=1033D017@
21:44:06 Join Willwolfe [0] (
21:50:15 Quit _Waheed (Remote closed the connection)
21:52:22 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
22:03:47bertrikDidn't find c200v2 pictures, but did find some chip numbers in a forum thread, I'll update the c200v2 wiki
22:06:10 Quit robin0800 (" - Chat comfortably. Anywhere.")
22:10:55 Join CaptainKwel [0] (
22:11:45 Join robin0800 [0] (
22:18:34kugelZagor: re watermark patch: I'm not sure how I can help you more. I wrote down what I noticed
22:19:01kugelwith the latest patch it's indeed much better, but the UI is still more laggy compared to SVN
22:19:10kugelespecially with ogg
22:19:56Zagorcan you see a significant difference in backlight fade speed between mp3 and ogg whe using the patch?
22:19:57kugelyou said you expect the codec thread to run more often, but doesn't constantly having *R mean it runs all the time?
22:20:14Zagorhow significant? double time?
22:21:38Zagorthe os stacks screen only shows snapshots
22:22:35kugela bit more than 50% slower
22:22:47Zagorand it doesn't show *R constantly for me. it toggles between S and *R.
22:23:47kugelI have never S
22:24:10kugelwith mp3 I sometimes see +R between *R, with ogg only *R
22:25:31Zagorkugel: I'm looking at SVN on my c200 now and see constant *R...
22:25:44Zagorplaying mp3
22:26:43Zagorplaying ogg makes it alternate between *R and +R
22:27:04kugelZagor: hm, this is interesting: when the ogg started freshly, it shows S sometimes. Roughly after 25s (backlight was off inbetween) it gets slow and shows only *R for the rest of the song
22:27:24 Join mib_mlrkam2x [0] (i=d8ef49f3@gateway/web/ajax/
22:27:30 Nick mib_mlrkam2x is now known as MarcGuay (i=d8ef49f3@gateway/web/ajax/
22:27:36kugeland the first fade down isn't even that slow, but the second one definitely is
22:28:13MarcGuayBagder, Zagor (take 2, please let me know if message has been received): is still talking about 3.0.
22:29:46ZagorMarcGuay: fixing
22:30:30kugelZagor: it gets slower after buffering is done (i.e. boosting over)
22:30:56Zagorkugel: what does the buffering thread screen say about boost ratio?
22:31:01kugelso if you don't want for the audio buffer filled you don't notice the laggy UI etc
22:32:00kugelwell, it's boosted until the buffering ended, then returns to normal (which is ~50MHz here for ogg + some dsp effects)
22:35:19 Quit robin0800 (Remote closed the connection)
22:35:23Zagoron my c200 ogg has a boost rate of ~20% on SVN
22:35:53 Join robin0800 [0] (
22:36:10kugelZagor: it's at 28% now after some minutes, but I have crossfeed enabled
22:36:29 Quit excitatory (Remote closed the connection)
22:36:46Zagoryou have crossfeed? well that would have been useful to know earlier... trying that now.
22:37:22 Quit itcheg (" ajax IRC Client")
22:38:20kugelthe ui is still laggy without crossfeed
22:38:44kugelthe backlight fading is like normal though
22:39:49Zagorit's starting to sound like the patch simply uses less boost than svn
22:40:27kugelI suppose the UI laggyness is more obvious if you use a scrollwheel
22:40:55MarcGuaygevaerts: Hi there. I'm looking at a solution for FS #9704 and was wondering what you thought of changing the options 'release30' and 'unreleased' to more standard HAS_RELEASE/UNRELEASED type vars.
22:41:03ZagorI tried it with an e200 yesterday but couldn't feel it. maybe I'm just less sensitive.
22:41:24kugelZagor: but ~8% more boost for crossfeed sounds like what I know from svn
22:41:25Zagorhowever if the cause of this is that we use less boost, then that is actually a benefit...
22:42:08Zagorcrossfeed isn't changed by the patch
22:43:02kugelwell, scrolling through lists is noticeably laggy here (it seems to stop scrolling for a splitt second every 0.5s or so)
22:43:33kugelI wouldn't call that beneficial
22:43:46kugelremember that it'll be much worse if ipods
22:44:02 Quit Willwolfe ("ChatZilla 0.9.84 [Firefox 3.0.1/2008070208]")
22:44:15kugelwith and without crossfeed, that is
22:44:51Zagoreverything that makes playback use less boost is a win. UI lagginess is another issue and must be dealt with separately.
22:45:24kugeland I have slightly more than 20% with ogg, if you have the same on the e200, you don't really boost less
22:45:49kugelwith SVN
22:47:21kugelI'd do some test_codec session before claiming less boost too early
22:47:23ZagorI'm getting <20% with the patch. hovers around 18.5%
22:48:09 Join fenugrec [0] (
22:48:15***Saving seen data "./dancer.seen"
22:48:51kugeldo you think 1% less boost is worse major UI slowness?
22:48:59kugel(major for me)
22:49:00ZagorI'm just saying UI lagginess is not a benchmark for performance.
22:49:31*kugel 's typing sucks today, wow
22:50:55kugelZagor: is it correct that the pcm buffer isn't refilled fully?
22:51:03kugelit seems to be only below the half
22:51:09kugelall the time
22:51:09 Part redwood
22:51:24MarcGuayAnyone with an Archos Player around? Is this sentence "Use the MINUS and PLUS keys to step through several pages of information." required at the beginning of the Main Menu -> System section of the manual?
22:51:28Zagorthe patch changes the debug screen. the fill bar now shows different values.
22:52:02 Join moos [0] (i=Mustapha@rockbox/staff/moos)
22:52:04kugelbut the first one is still the pcm buffer isn't it?
22:52:05Zagorthe old display didn't say anything useful about the pcm buffer
22:52:10Zagoryes it is
22:52:12kugeland that bar is never full
22:52:40kugeldoes that mean the pcm buffer is never full either?
22:52:43Zagorbut that has always been the case
22:52:51Zagoryes it does mean that
22:53:07kugelyea, but before it was always 75%, now <40-50%
22:53:19Zagorno. as I said, I changed the display
22:53:57kugelso that it shows correct values now, I suppose
22:54:09kugeland if it's not exceeding 50%, then we're wasting RAM on that
22:54:20Zagor"correct" is a matter of definition. the pcm buffer is not just a buffer with pcm data.
22:54:21kugeland boost more often (but shorter) than needed
22:55:12kugel(iirc boosting is wasteful on at least coldfire, so longer but fewer boosts should be favorable)
22:55:53Zagorcorrect. but smaller pcm buffer (= larger file buffer) is even more beneficial
22:56:06Zagoreven if it means fewer short boosts
22:56:46kugeldepends on how much boosting (without taking higher clocks into account) alone costs
22:58:19kugelamiconn knows more about that topic I think
22:58:44Zagorit's not a fixed value. the worst case is rather bad, but doesn't happen very often iirc
23:00:19bertrikhmm, lamp doesn't seem to work anymore on the clip
23:00:21Zagorat any rate, boost delay has its' problems but does not affect power. pcm buffer size has a direct effect on runtime
23:00:46 Quit karashata ("G'bye everyone!")
23:01:23Zagorbertrik: someone blamed jdgordons statusbar patch for that. I don't know if that's the case.
23:01:33amiconnZagor: Boosting/unboosting too often does affect power consumption especially on coldfire
23:01:43Zagoramiconn: how?
23:01:58amiconnThis is because the pll relock needs to be polled for, and during that time the cpu can do nothing else
23:02:00bertrikZagor, yeah it seems to exit immediately after starting
23:02:24amiconnPLL relock time is 10ms worst-case, and 2..3ms typical
23:02:58kugelamiconn: how's it on pp in SVN?
23:02:58amiconnSo there shouldn't be more than 1..2 frequency changes per second in order to keep the loss due to pll relock <1%
23:03:26kugeliirc it's not free either (but it can be made to be free)
23:03:40amiconnPP pll relock is 500µs (fixed safety margin on PP5020) resp. 250..500µs (PP5022)
23:04:20Zagoramiconn: huh? if it's 2-3ms typical, doesn't that mean 3-5 changes is 1%?
23:04:21kugelis that enough to get noticeable lags while scrolling?
23:04:28amiconnAdditionally, clock changing needs core synchronisation on PP, but I don't know why
23:05:12amiconnZagor: Yes, but worst-case is 10ms, so 1 change per second would be 1%.
23:05:22Zagorkugel: if you can sense 500s delays, you should be in another line of work :-)
23:06:25kugelZagor: it's not about how long the delay is, but rather that there's one. And having more of those (which always may lock the scrollwheel) can be noticeable
23:06:54Zagorthe length is definitely a factor
23:06:56kugelthe scrollwheel is interrupt based, that means it's read much more often than each 10ms like in a tick-task, if you turn it fast enough
23:08:07Zagorhuh? 10ms is 100 hz. I hope you're not complaining about lagginess when scrolling >100 lines/sec?
23:08:44Zagoror is there several interrupts per "click"?
23:09:35kugelit's noticeable because actually stops scrolling for a split second, intead of fluently keep scrolling at a constant rate
23:09:46kugelseveral, yes
23:09:56kugel48 per line afaik
23:10:11Zagor48 interrupts per line???
23:11:30kugelhm, maybe not 48, but it's way more than just 1
23:11:46 Join tyfoo2 [0] (
23:11:48amiconnSansa wheel has 48 status changes per revolution. iPod wheel has 96
23:12:05Zagoramiconn: per revolution != per line
23:12:34amiconnSure. Iirc it's 6 interupts per line on the ipod (without acceleration)
23:12:36kugelyea I got that wrong, but it's still more than just a few
23:12:41Bagderrevolution? is that a full "circle" ?
23:13:02kugelthe fuze wheel has about 5 per line
23:13:55amiconnAnd btw, it's *very* easy to achieve several interrupts per tick on the ipod. This is already achieved at moderate speed, and the reason why polling in a tick task isn't enough
23:14:47fenugrecwhy isn't it possible to service input control interrupts while waiting for the PLL to stabilise ?
23:14:55*amiconn remembers the behaviour of the old ipod G1..G3 wheel driver back when he got his G2
23:15:34kugelamiconn: might be less status changes per rotation, but I find the current fuze's wheel driver "quite acceptable"
23:16:54*Bagder still considers BFG (boosting for gui) is a fine idea
23:17:02 Quit tyfoo (Read error: 145 (Connection timed out))
23:17:16kugelthe worst thing is the acceleration, which is basically not sufficiently possible
23:17:56ZagorBagder: indeed
23:18:30*kugel would like to see a ipod video user testing the watermark patch
23:18:44Bagderin our quest to lower cpu clock for playback, I don't see how we can drive the gui properly at the same time without boosting
23:18:47Zagorkugel: me too!
23:19:31Bagderthat, and stop updating wps when backlight/screen is off are two of my want-to-sees
23:19:47ZagorBagder: btw which BFG method do you mean? :) saratoga and I have suggested two rather different schemes.
23:19:51kugelBagder: yea, and I'm basically with saratoga, who propsed a gui boost level (i.e. not fully boosted)
23:20:06kugelbut there's some hurdles to master, like the boosting issue on cf
23:20:24BagderI'm not convinced introducing more levels is worth it
23:20:30amiconnBagder: Not updating the wps without backlight should *only* be done on pitch-black type colour LCDs
23:20:37Zagorwe'll have to run this both ways, like everything else. #ifdef HAVE_ZERO_COST_BOOST
23:20:46kugeland I'm not convinced that the gui needs full boost
23:20:55Bagderamiconn: whatever, I'd prefer my greyscales to stop it too if it runs playback longer
23:20:58amiconnAnd we don't need more than 2 clock levels, except on targets with variable (officially) core voltage
23:21:12Zagorkugel: it most certainly doesn't. that's why we should only boost sections of code that need it.
23:21:25amiconnBagder: I would hate that. I often check wps without backlight
23:21:31Zagoranything that introduces sleep-on-boost is a no-no in my book
23:21:40amiconn*If* we add that, it should be an option on such LCDs
23:21:58Bagderamiconn: still only minor details, and frankly I'm not too concerned with such details
23:22:24BagderI'd prefer a first version that simply boost while the backlight is on or similar
23:22:28amiconnBtw, I doubt that this will have a measurable effect on grey/mon and small colour LCDs
23:22:42amiconnBagder: Again, very bad idea.
23:22:58amiconnThen it will be always boosted if backlight is set to always-on
23:22:59Bagderagain, subjective opinions
23:23:11Bagderagain details that I don't bother mentioning
23:23:33BagderI'm not writing full and complete suggestions here as you might guess
23:23:43Bagder"or similar" would imply that
23:24:25BagderI mean that it could boost rougly the same as backlight works with timeout 2 secs or so
23:24:29Bagderperhaps 0.5 secs
23:25:07amiconnAgain, I don't see the need for gui boost on any target I own
23:25:12 Join itcheg [0] (i=62db4767@gateway/web/ajax/
23:25:20ZagorI think that's a fallback to use for targets that can't do 0-cost boost
23:25:28amiconniPod Video is probably the only target that would benefit from it
23:26:06Bagderbut won't we be able to lower the clock even more at least on some targets ?
23:27:03gevaertsMarcGuay: my reasoning for those (release30, release31, unreleased) was that we may skip a release for a particular target at some point. That wouldn't make that target unreleased, it would just keep at the previous release.
23:27:52MarcGuaygevaerts: I see. Looks like it might be a pain to maintain, though... I'll leave it be for now...
23:28:05amiconnBagder: I don't know. Which target would you suggest doing this on?
23:28:09Bagder... and if Zagor's work for example makes less boosting and that (in itself) affects the gui in a bad way, it too is a sign that the codec-boosting is not enough to drive the gui
23:28:29amiconnLowering the unboosted clock might create undesired side effects
23:28:33gevaertsMarcGuay: I'm not opposed to dropping it. I have no idea at all how likely that scenario actually is
23:28:54Bagderamiconn: I'm basing this on what I recall from work by Buschel but I can't say that I have the details clear to me, so I'm not sure
23:29:50gevaertsMarcGuay: it does add maintenance work for a hypothetical scenario. That is actually against normal rockbox rules for code, so...
23:30:08MarcGuaygevaerts: If you're interested, see getting_started/installation.tex line 223 for a place where it seems a bit strange...
23:30:29 Quit itcheg (" ajax IRC Client")
23:30:31MarcGuayDo we update that on every release?
23:31:12*gevaerts slaps forehead
23:32:08 Join forehead [0] (n=54bd4b18@gateway/web/cgi-irc/
23:32:15MarcGuayWe hard-coded a lot of "3.0" when it came out... Presumably thinking it would take 2 years for 3.1. :)
23:32:26gevaertsThe way it is now, it's meant to be updated on every release, yes (although we could do a few in advance). I did update it for the 3.1 branch, but I forgot to merge that for trunk
23:35:13gevaertsMarcGuay: I'll merge the relevant commit to trunk now, so at least the immediate problem is solved. We can see what to do later then
23:36:01 Join Buschel [0] (
23:36:31Mode"#rockbox +o forehead " by ChanServ (ChanServ@services.)
23:36:49Mode"#rockbox -o forehead " by ChanServ (ChanServ@services.)
23:37:16Buschelhi folks, just saw you were talking about gui boosting aka lowering normal clock? ;-)
23:37:20 Quit {phoenix} (Remote closed the connection)
23:37:46Bagderit's that time of the year! ;-)
23:37:49 Quit forehead ("CGI:IRC")
23:38:17*Zagor spots zero cost boosting on ams
23:38:22Buschelbtw: happy new year! first time on irc in 2009 :-)
23:38:43bugonshoeis there a way to use a rockboxed ipod while it's plugged in usb?
23:38:50bugonshoeso it's just charging while you use it
23:39:37 Join draft [0] (
23:39:41MarcGuaybugonshoe: Quick Start section should cover that. Depends which model, though.
23:40:17bugonshoethanks marc
23:40:20domonokyZagor: jup, enough dividers available on ams so we dont change the pll at all, and we even have a spare pll :-)
23:40:26 Quit merbanan (Read error: 110 (Connection timed out))
23:40:49drafti'm having problems with my iPod Classic 160GB. :| it asks me to contact
23:40:55Buschelcould anyone test FS #9746 and FS #9708 yet?
23:41:04draftand it has red cross between circle.
23:41:20bugonshoei thought those weren't cracked
23:41:30MarcGuaydraft: You'll have better luck asking about that somewhere else.
23:41:38kugelZagor: indeed, and I head zero cost boost is doable on PP too
23:41:49domonokydraft: why are you telling this to us ? its offtopic, rockbox doesnt run on the classic..
23:41:51BuschelFS #9708 works, but the DMA transfer does not seem to save that much HDD "up-time". at least the battery runtime is not affected
23:42:10amiconnkugel: It *might* be doable. Depends on whether we can get rid of the core sync on clock change
23:42:22BuschelI will apply and test FS #9746 tomorrw. that one also sound promising
23:42:23*amiconn hopes _jhMikeS_ knows why this is necessary.
23:42:27amiconnHe added that
23:42:57amiconnkugel: Oh, and it limits the choice of frequencies
23:43:34kugelis that really an issue?
23:44:06amiconnIt might be, depending on target
23:45:50 Quit bmbl ("Woah!")
23:45:58gevaertsMarcGuay: I've merged my version bump from the 3.1 branch. Feel free to move to a simpler released/unreleased system though if you think it's better
23:46:50amiconnOn PP5020 we can use 80MHz divided by integer numbers as the normal clock (that is without increasing the PLL clock). On PP5022 we can also use half-integer dividers (the PLL is running at 160MHz when CPU clock is 80MHz on these)
23:47:10bertrikams as3525 can also do voltage lowering from 1.2V to 1.1V for CPU frequencies <= 200 MHz (Note on page 12 of the data sheet)
23:47:59Zagorbertrik: that's nice!
23:48:04MarcGuaygevaerts: Alrighty. Will look later, off to take care of Real Life.
23:48:42 Quit Highlander ("Leaving")
23:48:49amiconnZagor: In fact it's not really nice. Variable core introduces some extra trickiness
23:48:56Zagoramiconn: can't we simply change CLOCK_SOURCE from/to PLL?
23:48:59amiconn*Vraiable core voltage
23:49:34amiconnZagor: That can be done as well. It provides 24 MHz in addition to the integer/ half integer dividers
23:49:43draftMarcGuay: do you have idea where should i ask?
23:49:51MarcGuaydraft: No.
23:50:41Zagoramiconn: what is the problem with changing core voltage?
23:51:08MarcGuaygevaerts: Let's assume that a player had a 3.0 release but not a 3.1 release. The way it is now, they would both point to the same /download/ page anyway...?
23:51:32amiconnWith variable core voltage, power consumption becomes nonlinear. This means our simple boost/unboost scheme becomes sub-optimal, and we'd need gradual clock adjustment
23:51:49amiconnAlso core voltage adjustment means clock change is no longer zero cost
23:52:15gevaertsMarcGuay: yes
23:52:33*gevaerts is getting convinced that his scheme complicates things without really solving anything
23:52:45 Join solexx [0] (
23:52:50 Join rockboxi29 [0] (n=4e3567fc@gateway/web/cgi-irc/
23:53:05amiconnIn a simple 2-level scenario like on the AMS sansas, one extra speed might be sufficient - 200MHz, the possible maximum without increasing voltage
23:53:45 Quit rockboxi29 (Client Quit)
23:54:57bertrikamiconn, sounds like a good plan to arrange it that way
23:54:57amiconnThen switching between 2 levels (normal and 200MHz) won't be different from continuous adjustment. If we need more than 200MHz it becomes sub-optimal
23:55:11Zagoramiconn: that's illogical
23:55:26Zagorit will still be more efficient than today
23:55:49Zagorjust because we can't make something theoretically perfect does not mean we should not do it
23:55:55amiconnYes, sure. What I mean is that it won't be worse than doing real continuous adjustment
23:56:17amiconnFortunately not many usage cases should require more than 200MHz
23:56:20 Quit robin0800 (Remote closed the connection)
23:56:36 Quit solexx_ (Read error: 60 (Operation timed out))
23:56:39amiconnBut on the beast we do need real continuos adjustment
23:57:15amiconnBecause it has fine-grained core voltage adjustment
23:57:51Zagoroh you defined "contious adjustment" as "switching voltage with frequency"?
23:58:09bugonshoeMarcGuay, i figured it out. you hold menu as you plug in the usb
23:58:21 Join Roxiboxi [0] (
23:58:37amiconnNo, I mean continuous clock adjustment depending on load (as opposed to simply switching between 2 speeds)

Previous day | Next day