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 2008-03-25

00:00:20 Quit BigBambi (Remote closed the connection)
00:01:06 Quit DefineByte ("Bye all")
00:02:25linuxstbpixelma: Your WPS works fine in the M3 sim, but doesn't work on the remote for the M5...
00:02:58linuxstbSo there seems to be a bug somewhere.
00:03:10*linuxstb compiles an X5 sim
00:03:41pixelmabtw. I also notice that the M3 sim seems to have problems changing the font (I used a nimbus-12 with a different WPS and now loaded cabbiev2 as theme and the font didn't change)
00:04:09pixelmaso maybe it's not font loading but theme loading
00:04:50pixelmaah yes, I can see now that it's working in the M3 sim too
00:05:36 Quit Soap_ (Connection timed out)
00:05:39 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
00:06:07 Quit ChAoS ("CGI:IRC")
00:06:12 Quit Jon-Kha (Remote closed the connection)
00:06:21BigBambiChAoS: The linux version works perfectly here
00:06:32 Join ChAoS [0] (n=44e4e61c@gateway/web/cgi-irc/
00:06:38jhMikeShello...I've got two main "what ifs" I appreciate having checked out before committing this threading weirdness: iPod 5G, large sector (overall buffering speed was fixed without spinlock but glitched at first playback so that change had to wait), and a 64-bit sim (odd sensitivity to exclusion practices).
00:08:53ItalianPianistgood night guys
00:08:56 Part ItalianPianist
00:09:12linuxstbpixelma: (unsurprisingly), same problem in the X5 sim
00:09:51pixelmathat's what I would have expected too (just started to compile though myself)
00:10:44 Quit SoapH ("CGI:IRC (Ping timeout)")
00:10:58ChAoSBigBambi: Trying a reboot since my laptop was suspended from use at work and trying again... any other suggestions?
00:11:23pixelmalinuxstb: I tried commenting out the text viewport and then the status bar viewport shows, so maybe it's not general viewport handling (?)
00:11:41BigBambiNot so far as rbutil goes - I just tried the linux version and it worked perfectly. You can try the manual method
00:12:02*amiconn suspects the problem in the line array handling
00:12:03 Quit cool_walking_ (Read error: 104 (Connection reset by peer))
00:12:19jhMikeSlinuxstb: I noticed cabbiev2 just brings up the default WPS on x5 and H10 (both color @160x128) and works on my other players. Full clean, rebuild and reconfig was done.
00:12:26amiconnNow there can be several lines with the same line number (within different viewports)
00:13:01linuxstbjhMikeS: Ouch... I've just got the same thing in the X5 sim...
00:13:02ChAoSBigBambi: Only reason I am using the windows one is because the OSX one wont run on my G4.. know anything about that?
00:13:15linuxstbamiconn: Which line array are you talking about?
00:13:28BigBambiChAoS: When did you download the Mac one?
00:13:36ChAoSfew days back.
00:13:37linuxstbjhMikeS: Investigating...
00:13:40amiconnThe one that you multiplied in your first commit, and later changed back to a single one
00:13:41 Quit Soap__ (Read error: 110 (Connection timed out))
00:13:53amiconnI'm just guessing without actually looking at the code...
00:14:22BigBambiChAoS: There was a problem with the OSX one initially being compiled Intel only, but I thought that had been fixed
00:14:43linuxstbamiconn: This bug seems to just happen on remote LCDs... There is one line array per screen, so I would expect the bug to happen with the main LCD as well if there was that problem.
00:14:48ChAoShmm.. maybe I will downlaod again and try over on my mini if this doesnt work.. thanks for your suggestions!!
00:14:48BigBambiChAoS: Perhaps try the sansapatcher + unzipping the build method
00:15:19 Join ceclin [0] (n=clincher@
00:15:44BigBambiChAoS: rbutil is usually fine but it still in development so may have the odd issue, but if you still have problems, try the manual method outlined in the manual
00:16:29 Join SoapH [0] (n=621131c8@gateway/web/cgi-irc/
00:16:45linuxstbjhMikeS: Your H10 also has a 160x128x16 LCD?
00:16:54linuxstbjhMikeS: Sorry, you answered that...
00:17:02ChAoSBigBambi: will do.. thanks again for the pointers.
00:17:03linuxstbI found the bug, committing now (missing |)
00:17:18BigBambiChAoS: No probs, I hope you get it working
00:17:34 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
00:18:31linuxstbjhMikeS: It was actually my later commit to make the %V parsing stricter that broke it - at the time I changed the WPS, the parser allowed that...
00:19:20 Join DaCapn [0] (
00:20:35 Join EspeonEefi [0] (i=espeonee@STRATTON-FOUR-SEVENTY-EIGHT.MIT.EDU)
00:20:54jhMikeSlinuxstb: you didn't update the 160x128, color versions?
00:21:58ChAoSAnyreason I couldnt run the rbutil 1.0.3 on osx since the 1.0.4 appears to be intel proc only?
00:22:00 Quit SoapH ("CGI:IRC (Ping timeout)")
00:22:44jhMikeSlinuxstb: just asking because the e200/gigabeat had no problem
00:22:45 Quit ender` (" The trouble with having an open mind, of course, is that people will insist on coming along and trying to put things in it.")
00:23:17linuxstbjhMikeS: Yes. The H10/X5 problem was a typo that the parser used to ignore, but I made the parser stricter.
00:23:38linuxstbs/typo/missing terminating | character/
00:25:26pixelmamaybe I should just go on committing all my greyscale cabbiev2 changes except the Iaudio remote
00:25:38 Quit fehmicans (Read error: 104 (Connection reset by peer))
00:27:10BigBambiChAoS: You can try. I seem to remember that Complete Install under 1.0.3 had a bug, so you will need to use the second tab to do the vootloader and build installs individually
00:27:11jhMikeSanother thing that happened but seems to have disappeared on H120 was that cabbiev2 would use the fixed font. Not sure if that was deliberately fixed or just silently showed and went away.
00:27:35 Part toffe82
00:30:25ChAoSBigBambi: Thanks
00:31:38BigBambiChAoS: You are welcome
00:31:49pixelmaI observed a strange behaviour on the Mini yesterday - going from cabbiev2 to the old text default (by loading the saved.cfg). Icons and backdrop vanished as should be but the WPS layout (tag info) was placed like in cabbie but using the reloaded font and clipped to the area the loaded WPS should have had (I think)
00:32:25linuxstbI think amiconn had the same problem - does your .cfg contain the line "wps: " (with nothing else) ?
00:32:26pixelmaa reboot "fixed" it
00:32:59pixelmaok, then he reported that already (sorry didn't read all of the logs...) :D
00:33:14pixelmait was his Mini
00:33:27linuxstbYes, I think he mentioned it was his mini...
00:33:28jhMikeSreading _all_ the logs would be way too time consuming
00:33:52linuxstbSomeone should post a daily executive summary...
00:34:02*linuxstb is not someone
00:34:05 Quit shotofadds (Read error: 104 (Connection reset by peer))
00:34:39Lloreanlinuxstb: Time for Rockbox to hire a secretary? :-P
00:35:40jhMikeSif anyone cares to do a quick check 80gig or 64-bit sim:"> (actually does a little more than the name says, ignore the crud about /uisimulator/sdl/kernel.c that may come up)
00:36:16ChAoSBigBambi: Ouch... cand find file rockbox.e200.mi4 or something then autooff.. suggestions?
00:36:34LloreanjhMikeS: I don't know if this is a priority issue or what. I've noticed on my Sansa that during first buffer, it's completely unresponsive to keypresses.
00:37:08ChAoSBigBambi:rockbox.mi4 and rockbox.e200 both not found.
00:42:06amiconnLlorean: That seems to depend on the theme. I observed it while testing bitmapstripped cabbiev2 on H300
00:42:07jhMikeSLlorean: yes, it is that.
00:42:20 Join SoapH [0] (n=621131c8@gateway/web/cgi-irc/
00:42:41amiconnMy guess is that the buffering code somehow "forgets" to yield at all under certain circumstances
00:43:01*amiconn shrugs at reading the word 'preempt' in that patch
00:43:03LloreanI hadn't noticed it on my Gigabeat, but then I rarely press anything on first buffer.
00:43:44 Quit ChAoS (Nick collision from services.)
00:43:51amiconnIt never seems to happen when resuming after booting completed, but for the test I enabled resume on startup
00:43:52 Join ChAoS [0] (n=44e4e61c@gateway/web/cgi-irc/
00:43:57 Quit ChAoS (Nick collision from services.)
00:44:57 Quit hannesd (Read error: 113 (No route to host))
00:45:12Lloreanamiconn: My e200 had been booted for some time when I ran into it. At least long enough for me to change themes.
00:45:26amiconnah, hmm
00:45:33LloreanI'll double check
00:45:37amiconnAnyway, there seems to be a serious problem
00:45:53amiconnButtons aren't even pre-registered, at least on H300
00:46:38amiconnHmm, that'd mean interrupts are somehow disabled...
00:47:20 Quit SoapH ("CGI:IRC (Ping timeout)")
00:47:24jhMikeSthe patch has no UI stickiness on any of my players and also allows codec priority to be brought up gradually such that it's brought up smoothly
00:47:42 Join linuxstb__ [0] (
00:48:27 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
00:49:40 Quit linuxstb_ (Read error: 104 (Connection reset by peer))
00:49:48 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
00:50:33 Quit linuxstb (Nick collision from services.)
00:50:35 Nick linuxstb__ is now known as linuxstb (
00:54:13*JdGordon loves the size of the recent wps commits :p
00:55:22 Join fasmaie [0] (
00:55:23jhMikeSas in file count?
00:55:26*pixelma pffs at JdGordon
00:55:54JdGordonyes, filecount
00:56:03JdGordontaking 455min to scroll down the screen
00:56:24 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
00:57:28 Join ChAoS- [0] (n=44e4e61c@gateway/web/cgi-irc/
00:58:38*jhMikeS thinks JdGordon exists in an accelerated frame of reference if it takes 455 minutes to scroll :p
00:59:10JdGordonthat was either meant to be 4, 5, 45 or 55... depends where you want the typo to be :p
00:59:24 Quit saratoga (Client Quit)
00:59:27linuxstbpixelma: I've tracked the bug down to a problem with the wps parser, but I'm not sure exactly what yet...
00:59:30 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
00:59:53jhMikeSsqrt(1 - v^2/c^2)?
00:59:58*amiconn wants more data registers on coldfire :/
01:00:02 Join cool_walking_ [0] (
01:00:42pixelmalinuxstb: thanks for looking into it, I'm currently preparing a commit without the Iadio remote (something more to scroll for Jd :P )
01:00:45 Quit mf0102 ("Verlassend")
01:00:58***Saving seen data "./dancer.seen"
01:04:58 Quit saratoga ("CGI:IRC (Ping timeout)")
01:05:13 Quit XavierGr (Nick collision from services.)
01:05:24 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
01:07:47 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
01:07:55 Join Xerion_ [0] (
01:07:59saratogathe webclient has been all screwy for me lately
01:08:56 Quit Xerion (Read error: 104 (Connection reset by peer))
01:08:57 Nick Xerion_ is now known as Xerion (
01:09:20 Join HEx [0] (n=HEx@
01:09:29ChAoS-BigBambi: Thanks!! have my first rockboxed device... starting to customize my theme already!!
01:09:39 Quit tvelocity_ (Remote closed the connection)
01:11:47 Quit loudergood (Read error: 110 (Connection timed out))
01:12:34*linuxstb slaps forehead
01:13:08 Join corevette [0] (
01:14:04 Join forehead [0] (n=54bd755c@gateway/web/cgi-irc/
01:14:08 Quit DaCapn (Remote closed the connection)
01:14:21 Join DaCapn [0] (
01:15:36pixelmalinuxstb: does that mean I should hold back the commit and wait for the remote WPS problem being fixed or should I just go ahead and commit without it (and get my commit count up ;) )
01:16:00linuxstbI'm just compiling an ipod sim to test (no remote LCD), but yes, I think I've found the bug
01:17:05pixelmanice... but I was about pressing enter...
01:17:11 Join crzyboyster [0] (n=6210f195@gateway/web/cgi-irc/
01:17:11*linuxstb prefers cabbie on the smaller LCDs - no enormous icons...
01:17:53*pixelma just went ahead
01:18:17pixelmamaybe there is something missing anyways
01:18:27 Join toffe82 [0] (
01:20:35crzyboysterI'm having trouble with a wps that uses viewports and then putting in a progressbar and other icons (play, pause, etc) with not showing up. Here is my code:
01:21:01linuxstbpixelma: OK, my fix is committed.
01:21:09saratogaBagder: did you ever add my machine as a build server?
01:22:31linuxstbcrzyboyster: The icons are being displayed in the viewport you defined for the time, which is only 30x10 pixels in size.
01:22:31BHSPitMonkeylinuxstb, (little off the subject): do you happen to use facebook?
01:22:50linuxstbBHSPitMonkey: No, but I think I may have registered
01:23:06 Join RoC_MasterMind [0] (
01:23:11BHSPitMonkeylinuxstb, ah
01:23:23BHSPitMonkeyI'll bug Bagder about it then, I know he does
01:23:41crzyboysterlinuxstb: Thanks! Let me poke around some more...
01:23:45 Quit crzyboyster ("CGI:IRC")
01:23:57*forehead is annoyed by getting slapped so often :|
01:24:02 Quit RoC_MasterMind (Client Quit)
01:24:05 Quit forehead ("CGI:IRC 0.5.9 (2006/06/06)")
01:24:21*linuxstb would have thought forehead was used to it by now
01:24:33*BHSPitMonkey is tired of the cheap /me jokes
01:24:34cool_walking_I assumed you specifically chose that name so people would slap you
01:24:57BHSPitMonkeycool_walking_, who are you talking to?
01:25:42 Join ukl [0] (
01:26:19uklHiya. Can anybody already confirm that the bigpockets sansa c250 is _not_ v2?
01:27:00 Quit moos ("adios")
01:27:02linuxstbI doubt anyone has received one since you were last here..... I've ordered one though.
01:27:37uklI don't know when the offer began...ok, thanks for your help. I'll order tomorrow...
01:27:45uklsee you
01:27:52 Quit ukl (Client Quit)
01:28:00linuxstbOnly 288 left...
01:28:34*jhMikeS gueses by now there's no trouble with that posted patch in those two obscure corners :)
01:31:35 Quit amiconn (Nick collision from services.)
01:31:41 Join amiconn [50] (n=jens@rockbox/developer/amiconn)
01:32:23pixelmalinuxstb: build server stuck again?
01:32:56linuxstbI think it just started...
01:32:58pixelmaah no, just taking some time and complaining here helped (once more)
01:33:06 Quit Ravendug (Read error: 110 (Connection timed out))
01:33:15 Join Ravendug [0] (
01:33:23linuxstbAlthough it combined two commits...
01:33:35amiconnThat's okay
01:34:19amiconnThe build system combines commits which are arriving within the time window, which is normally 60 seconds, but if a build is already running, the window is extended
01:34:37pixelmalinuxstb: I'll blame any red/yellow on you ;)
01:34:46 Join kushal_12_27_200 [0] (n=kushal@
01:34:51amiconnThere's also some recovery time after one build during which it won't start a new one. I don't know why though
01:35:23 Join crzyboyster [0] (n=6210f195@gateway/web/cgi-irc/
01:36:28crzyboysterI have another question with my viewports wps coding. The progressbar is not being drawn with this code:
01:37:35linuxstbcrzyboyster: Again, you're drawing it outside the viewport
01:38:02linuxstbYour viewport is 14 pixels high, and you're drawing it at "toppos" 102
01:38:50crzyboysterWow. I should really analyze the viewports from an elementary level.
01:39:13crzyboysterAnd how are people making the bitmap strips. Like what program to use?
01:39:35linuxstbI use the command-line "montage" program, which is part of imagemagick
01:40:55 Join piga [0] (
01:42:45 Join jas6180 [0] (
01:43:35 Quit mud-rb (Read error: 104 (Connection reset by peer))
01:44:24 Quit quaal (Read error: 113 (No route to host))
01:45:02crzyboysterNow I'm truly stumped with this viewports progressbar thing. It still won't show up! Here is my code:
01:45:12 Join quaal [0] (
01:45:22crzyboysterSorry about this, I'm still getting the hang of this viewports code
01:46:01Nico_Plinuxstb: I've started to look at your WPS changes a bit... I like parse_list a lot :)
01:46:03pixelmayour viewport is still only 20 pixels high, if I read correctly
01:46:22Nico_PI remember wanting to add scanf to the core for similar reasons
01:46:35linuxstbNico_P: It's not as useful as I thought it would be though... Mainly because of the optional parameters everywhere (e.g. albumart0
01:47:09linuxstbAlthough maybe it could be extended to support them...
01:48:02Nico_Pyes, it would be good. then it could also be used for the progressbar
01:48:06*amiconn wonders why the 240x320 cabbiev2 backdrop is saved as 32-bit BMP
01:48:29crzyboysterMy progressbar is only 14 pixels high, though. Updated code at
01:48:43crzyboyster(Still won't work)
01:49:07linuxstbcrzyboyster: Change 102 to 0 in your %pb line
01:49:17linuxstbThat's the y position of the bar...
01:49:31 Quit perrikwp (" ajax IRC Client")
01:49:41linuxstbRemember that all co-ordinates are relative to the viewport the item is displayed in.
01:49:42pixelmacrzyboyster: you only changed the y position of the viewport not the height btw.
01:50:17pixelmaor even made it less tall (only 14 pixels)
01:50:21 Quit saratoga ("CGI:IRC (EOF)")
01:50:24*jhMikeS goes to sneak one in between all the WPS activity
01:50:37crzyboysterYES! (sorry, I got a bit excited) I have a lot of wps to update to viewports so I better get the hang of this quickly.
01:51:41Nico_Plinuxstb: how maintainable does the WPS code look to you? (I think you're the first one to make sizeable changes to it since I committed the parser)
01:52:17linuxstbNico_P: It's nice, but somehow still seems very big for what it does... Although I can't think of ways to reduce it.
01:52:52linuxstbDid you see my statusbar patch I posted earlier today (in IRC)
01:53:52linuxstbStarting here -
01:54:27 Join kushal [0] (n=kushal@
01:54:35 Quit kushal (Remote closed the connection)
01:54:38 Join perrikwp [0] (i=4aa794a0@gateway/web/ajax/
01:54:42Nico_Pthe screenshot looks nice :)
01:55:01Nico_Pwhere do you store the tokens?
01:55:39linuxstbKeep reading... It's specified as part of the main .wps file, using viewports
01:55:49Nico_Pah right
01:55:49 Join jas6180_ [0] (
01:56:29linuxstbSo no more RAM usage, and very little extra code
01:56:36Nico_Pit's a nice approach indeed
01:56:52crzyboysterSo album art needs to be in a viewport, too?
01:57:18Nico_PI had started making changes to the current code so that a wps_data struct would actually have pointers to buffers instead of containing the buffers
01:57:27linuxstbcrzyboyster: It's cleaner - and prevents it being overwritten by text lines. If possible, you should design your WPS to put everything in a viewport, and for those viewports to not overlap.
01:58:00 Quit crzyboyster ("CGI:IRC (EOF)")
01:58:02 Quit kushal_12_27_200 (Connection timed out)
01:58:09kugelHey Nico_P, I wanted to ask you about the resizing stuff. You played around with integrating in the core, didn't you?
01:58:28Nico_Pkugel: I did a few weeks/months ago
01:58:36linuxstbNico_P: I haven't really thought about the recording and radio screens, and how they would fit in though, if they were customisable.
01:58:51kugelAnd how's it going? Or did you give it up?
01:59:19linuxstbBut I think it might make sense for the radio and recording screens to be able to define their own status bars, which are displayed when those features are active.
01:59:23Nico_Pkugel: I haven't had much time lately. I think it's still a desired feature for the core, but I haven't been following the patch lately
01:59:57kugelThere wasn't much to follow
01:59:59Nico_Plinuxstb: my idea was to make the wps_data struct generic, and have one for each customizable screen
02:00:15Nico_Peach pointing to a different set of buffers
02:00:50linuxstbAlthough maybe it would make sense for them to share images.
02:01:19Nico_Pit probably would, yes
02:02:27linuxstbAnd I don't think 52 images is an issue now we have subimages. e.g. cabbiev2 just needs 6 of those slots
02:02:39 Join jas6180__ [0] (
02:02:42Lloreanlinuxstb: I agree that they should be able to define their own. As to images, if you have a ThemeA.wps, and a ThemeA.rcwps and a ThemeA.fmwps, they should all look in the ThemeA folder, right?
02:02:52JdGordonif the image buffers are shared they should be dumped to disk in 1 file so reloading one screen isnt slwoed down too mcuh
02:03:28 Quit jas6180__ (Remote closed the connection)
02:03:36JdGordonLlorean: wps/{wps,rec,fm}/<folder name> ?
02:03:42linuxstbLlorean: Yes, although maybe it would be nice for a .wps to be able to specify its bitmap dir - e.g. the unicode and non-unicode versions of a theme
02:03:43 Quit fasmaie (Read error: 113 (No route to host))
02:04:19 Join jas6180__ [0] (
02:04:34pixelmalinuxstb: ah, I can see the icons on the remote now :) going to commit... (first wps using viewports in svn)
02:04:44Lloreanlinuxstb: Some WPS meta data might be nice.
02:04:46linuxstbNo turning back now...
02:05:10JdGordontime to get the new theme site up then!
02:05:22linuxstbLlorean: What kind of metadata?
02:05:34LloreanJdGordon: Well, the idea was to share images. So /wps/blah would contain the images used the the While FM Screen, and the While Recording Screen if they all had the filename as blah
02:05:54scorcheJdGordon: hush...gotta take care of RL first =<
02:06:29Lloreanlinuxstb: Well, as you suggested the Image folder. But also perhaps the Theme Name (there might be some reason later to know if all screens were using the same overall "theme") or other such things.
02:06:31 Quit jas6180__ (Remote closed the connection)
02:06:49 Join jas6180__ [0] (
02:07:03LloreanMaybe "expected font heights for used fonts" to give warning if someone changes it to a too-tall font.
02:07:36LloreanThere's surely more info that could be stored about a WPS, really, to make them more "friendly" from a user perspective when the user starts making mistakes.
02:07:53 Quit jas6180__ (Remote closed the connection)
02:08:02 Quit Mouser_X (Read error: 110 (Connection timed out))
02:08:14*amiconn thinks that a themable recording screen doesn't make sense
02:08:47Lloreanamiconn: I think it does on some targets.
02:09:04LloreanFor example, the c200 has a fairly short screen, and different people have different views on what has priority there.
02:09:05linuxstbI would just like to see the recording screen adapt more to the different screen sizes - i.e. not treat every LCD as if it's 112x64x1
02:09:08 Join jas6180__ [0] (
02:09:40amiconnThat's a different thing. Of course the recording screen needs rework to no longer use the sysfont
02:10:12amiconnBut the recording screen has numerous necessary widgets, and not really optional information to choose from
02:10:41linuxstbAnd I would guess that recording types don't want eye-candy - just a clear layout of all the available info.
02:11:03Lloreanlinuxstb: Look at the various evolutions of the recording screen in the Recording Enhancement Pack.
02:11:25LloreanThey wouldn't necessarily want eye candy, but there's some disagreement as to what's necessary, and where screen space should best be used.
02:11:30amiconnFM is somewhat different, although I wonder how many pieces of optional information are on that one
02:11:57amiconn...unless a target has RDS - but we don't have one yet
02:12:07linuxstbIt's also a style thing though - making it look similar to the WPS, even if the info is the same.
02:12:15Lloreanamiconn: I think FM should be themed just for the sake of having its appearance consistent with the rest of the UI. Not really much in the way of "optional" info, but there can still be appearance preferences.
02:12:35 Quit PaulJam_ (".")
02:13:06 Quit jas6180 (Connection timed out)
02:13:33LloreanBut the recording screen has different priorities on the information based on whether you're using it in a well lit environment where you care about all the information and don't mind squinting, or whether you're stealthing and want to be able to see certain very key things at a glance (if my understanding is clear) but don't care much about other non-key things.
02:13:56amiconnStation name (from .fmr), frequency, mono/stereo, and (one some targets) signal strength
02:14:22Nico_Plinuxstb: are you still planning on allowing something like %?bl<%xdB> (with the expansion)?
02:15:02linuxstbNico_P: I'm not planning on implementing it myself...
02:15:12Nico_Phehe :)
02:15:54linuxstbBut I'm not 100% convinced it's a good idea - it may just be confusing.
02:15:56Nico_Pit'd probably need a clear syntax to allow subsets, though, don't you think?
02:16:30linuxstbI never like having multiple ways to do the same thing...
02:17:04linuxstbBut I'm not against it if others think it's useful.
02:17:08LloreanI'd rather my suggestion. ;)
02:17:24linuxstbLlorean: What was that?
02:17:24LloreanThough it wouldn't work for playback state.
02:17:56Lloreanlinuxstb: For conditionals that represent values, do like the progress bar and display a relative portion of the image. Only really useful for volume and battery though.
02:18:12LloreanBut it does mean you don't have more than one way of doing the same thing. ;)
02:18:18linuxstbNot even for volume - with the special cases
02:18:55LloreanWell, volume could have <mute, progressive image, clipping>
02:18:56*pixelma didn't want to drop the info at the top of the WPS, stupid c+p mistake...
02:19:28LloreanI just like the idea of giving WPS authors as much level of granularity as they can stomach, without needing giant images.
02:21:15LloreanSince people are already complaining about running out of space in the buffer, that is.
02:21:36*Llorean would also like it applied to skinnable peakmeters, while he's dreaming.
02:21:56*linuxstb points Llorean to the source code... ;)
02:21:58Nico_PLlorean: I'm not sure it's really useful for anything other than the progressbar
02:22:21Nico_Pthe battery, for example, usually has an outline. you couldn't do that
02:22:25 Quit jas6180_ (Connection timed out)
02:23:19LloreanNico_P: The outline can be managed either by the backdrop, or by four other images (only useful if the battery had more than five steps to begin with)
02:23:38LloreanOr, if image stacking ever works well, by one other image.
02:23:40Genesisarchos recruts some linux C coder , if some french are unployed :)
02:24:06JdGordonhehe peakmeters done the same was as the progress bar would be awesome
02:24:14jhMikeSLlorean: RB isn't about what you'd like, it's about KISS </end extreme sarcasm>
02:24:36LloreanjhMikeS: :-P
02:24:57*JdGordon grumbles off to uni :(
02:25:52Nico_PLlorean: suggest a good tag syntax and I'll implement it ;)
02:25:55*jhMikeS 's e200 just gave him an ES discharge while plugged-in for that comment :
02:26:20LloreanNico_P: For which idea? I've been spouting off too much nonsense to be sure which one you're interested in. :-P
02:26:59Nico_P"continuous" enums
02:27:22linuxstbHorizontal and/or vertical?
02:27:24Nico_PI like the themeable peakmeter even more but that one doesn't really require additional syntax :)
02:27:31jhMikeSvs. quantum enums?
02:28:05Nico_Plinuxstb: the syntax would probably need to take this into account, yes
02:28:26LloreanNico_P: For the battery one, I'd just make the the percent battery one a single-image enum.
02:28:29linuxstband direction...
02:28:59LloreanBut I thought there was a percent volume WPS tag too.
02:29:20 Join midgey [0] (
02:29:23LloreanBecause I *think* we can assume that nobody's going to use a single image to represent all charge states, normally, anyway.
02:30:06LloreanNico_P: Peakmeters, imho, need to be split into right/left individual peakmeters if they're to be made themeable.
02:30:23Nico_PI agree
02:30:36 Quit Genesis ("Quitte")
02:30:48LloreanAnd not only horizontal/vertical, but left-to-right vs right-to-left for the peakmeters at least.
02:31:01LloreanThough you could fake that with clever backdrop use, anyway
02:31:05LloreanSo it's non-essential
02:31:42*Llorean un-says the last few lines, since he realizes that would invert peaks.
02:32:03 Quit jas6180__ (Remote closed the connection)
02:32:40 Join jas6180__ [0] (
02:35:03kugelHas anyone thought about costumizable list again?
02:35:18 Quit Thundercloud_ (Remote closed the connection)
02:35:38 Join jas6180 [0] (
02:36:10kugelbtw, you all have great ideas for future costumizing possiblities of rockbox
02:36:11*Nico_P would love to have enough time to start making the peakmeter themeable
02:36:31kugelNico_P: If it was themable, I'd acutally use it
02:37:03LloreanIf it was themable, it'd probably hurt performance even more. ;)
02:37:25Nico_Phehe, probably :)
02:38:48Nico_Pdoes the forum have viewport-using WPS apart from arboxWidgets?
02:39:20linuxstbI've noticed a few over at, but I'm not sure if they've been updated for SVN implementation
02:40:45kugelI have several themes that use viewports in my build. And I will update most of them to the lasted viewport implementation
02:41:57LloreanThough I do think that right now, viewports "aren't impressive enough." They still, to many people, just appear to be an implementation of what they had in unsupported builds ages ago.
02:42:18LloreanThough if linuxstb's hack becomes a reality, I think people will start wowing at the SVN rockbox, at least for a little bit.
02:42:33LloreanAt least as long as people making the unsupported builds let people know that's a feature that came from us. ;)
02:42:52kugelWell, unless other screens/contexts make use of viewports more, it's not too impressive :)
02:43:07 Part thresh
02:43:22*Llorean finds it pretty impressive, since it lets him make font size independent WPSes.
02:43:52linuxstbHow does it allow that?
02:43:57kugelBut I think it's nice that there's a little competetion between the official builds and the unsupported ones. It speeds developement a bit up ;)
02:44:07scorchethere is?
02:44:41linuxstbI'm just fed up of people coming here complaining that themes don't work...
02:44:42kugelI think so
02:44:59kugelThat's another point imo
02:45:26*pixelma notices that cabbiev2 doesn't take remote hold into account but decides to bring that up tomorrow and find some sleep now, night
02:45:29Lloreanlinuxstb: Because you can set the absolute starting position of a line of text, so if you do a multiple viewports (never depending on line based positioning except when it's "safe") at least all smaller fonts won't screw up positioning
02:45:51linuxstbLlorean: Ah, I thought you were talking about my statusbar hack...
02:46:04Lloreankugel: There's no competition. Anything we can do, they can incorporate immediately, so it's basically physically impossible for us to be "ahead". so why bother?
02:46:27Lloreanlinuxstb: No, I meant viewports itself impresses me because it allows that. I just think that's not something many "oooh, shiny" people notice.
02:46:44scorchenot to mention that most of the unsupported builds just mainly do a bit of compiling..
02:47:30kugelYou want the users to use the official build. That automatically means, that the official build has to follow
02:47:58scorcheno it doesnt...
02:48:40 Quit Nevtus (Read error: 104 (Connection reset by peer))
02:48:53kugelI actually think that viewports came faster, since all these unsupported builds offered features that only viewport could also offer
02:49:14linuxstbviewports took about 2 years...
02:49:37kugeland it took even longer without custom builds
02:49:45jhMikeSwasn't it mostly procrastination though?
02:50:00scorchekugel: that sentence makes no sense
02:50:15scorchenot to mention that most of the unsupported builds were just compiling in a couple patches that were put on the tracker...not much work
02:50:31Lloreankugel: The custom builds didn't do any of the viewports work in advance.
02:50:42 Part pixelma
02:50:44scorcheviewports != scrolling margins patch
02:50:49Lloreankugel: It's entirely possible they made it take longer by providing rejected alternatives rather than encouraging people to work on viewports
02:51:00kugelMeh, you don't have to agree. I see this competition, and I think rockbox benefits from it
02:51:10scorchethere *is* no competition
02:51:27kugelscorche: Okok
02:51:37Lloreankugel: Don't you think we'd know it if we were competing?
02:52:22kugelGood question
02:53:31kugelI think you don't see them, since you are only into the official version and hate custom builds. I am into both a bit, and I see a little competition
02:53:57*Llorean doesn't see how we can compete with something we never look at.
02:53:59kugelIt's not like an official competition of course, it's a bit hidden
02:53:59jhMikeSit does have subtle effect on the subconcious and brings in a fierce urgency of now but I don't know if it's competition per se :p
02:54:16LloreanBy definition competition requires that we either share a goal, or know the other side's status.
02:54:36LloreanAnd if you don't know the other side's status, then its status doesn't affect the competition.
02:55:04*linuxstb greps the logs and finds the first viewport discussion was January 2006
02:55:09jhMikeSone still receives impressions from the collective wisdom though
02:55:19scorcheso what competition do you see exactly?
02:55:26LloreanjhMikeS: True. But I don't think that's enough to say we're in competition with them.
02:55:40LloreanIt's merely a means of benchmarking the wants of the userbase.
02:55:52 Join countrymonkey [0] (n=4b05639a@gateway/web/cgi-irc/
02:59:40 Quit Nico_P ("bed time")
03:00:02countrymonkeyI noticed that there was a russian romanizing table. I also noticed that russian.lang has cyrilic characters in the sysfont strings. Could I write a script to romanize the sysfont stuff bassed on the wikipedea table and get my patch excepted even though I don't speak Russian?
03:01:00***Saving seen data "./dancer.seen"
03:02:00 Quit jas6180__ (Connection timed out)
03:02:26jhMikeSwhich tends to spur ideas about how to make the "neat little patch" that the usebase can't do without into a more generally useable development. my sense that it just kicks a stubborn mule into action sometimes. ;)
03:03:31LloreanjhMikeS: True, but I'd suggest that it also tempts some users who could become action-doers into inaction by providing them instant gratification. I'd imagine they equal out.
03:03:32 Quit jas6180 (Connection timed out)
03:05:21*jhMikeS wonders how to measure the "scrolling margins effect" wherever it may come up
03:08:00 Join jas6180 [0] (
03:11:20LloreanjhMikeS: As far as I know, the only two major cases of there being an official alternative needing implementation have been viewports and "ignore the".
03:11:39 Nick fxb is now known as fxb__ (
03:12:53 Join Ebert [0] (
03:13:01 Part Ebert
03:13:25 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
03:19:13 Quit Rob2223 (Read error: 104 (Connection reset by peer))
03:19:32 Join Rob2222 [0] (
03:21:38 Quit ChAoS- ("CGI:IRC (Ping timeout)")
03:22:49 Join MobileShaid [0] (
03:24:33 Join Llorea1 [0] (
03:24:39kugelJdGordon: I have a question about your list-vp code
03:24:45 Quit Llorean (Nick collision from services.)
03:24:47 Nick Llorea1 is now known as Llorean (
03:25:25kugelJdGordon: I understand that the list is shifted right a bit if the scrollbar is shown and not shown, but with the setting on
03:25:40 Join kushal_12_27_200 [0] (n=kushal@
03:26:06kugelJdGordon: but do we really need to shift the list right, when the user has turned the scrollbar off totally?
03:36:27 Quit Seed (Nick collision from services.)
03:36:30 Join Seedy [0] (
03:37:22 Nick Seedy is now known as Seed (
03:40:03 Quit Shaid (Read error: 110 (Connection timed out))
03:40:04 Nick MobileShaid is now known as Shaid (
03:48:10 Quit piga ("Leaving")
03:49:50 Join csc` [0] (n=csc@archlinux/user/csc)
03:58:07 Quit BlakeJohnson86 (Remote closed the connection)
03:58:38 Join BlakeJohnson86 [0] (
04:02:56 Join ctaylorr [0] (
04:07:56 Join miepchen^schlaf_ [0] (
04:23:41 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:29:02*kugel is working on a custom list solution
04:34:40 Quit ctaylorr (Read error: 110 (Connection timed out))
04:42:51 Quit countrymonkey ("CGI:IRC (Ping timeout)")
05:01:03***Saving seen data "./dancer.seen"
05:08:19Mode"#rockbox +o scorche " by ChanServ (ChanServ@services.)
05:09:33Topic"Please read before speaking: | Please direct offtopic/social chat to #rockbox-community | Student applications for GSoC are open! |" by scorche (i=Blah@rockbox/administrator/scorche)
05:09:43Mode"#rockbox -o scorche " by ChanServ (ChanServ@services.)
05:09:47*scorche shrugs
05:10:56 Quit jas6180 (Read error: 104 (Connection reset by peer))
05:11:55 Join jas6180 [0] (
05:13:34 Join Mouser_X [0] (
05:18:38 Join webguest38 [0] (n=444fbedf@gateway/web/cgi-irc/
05:18:53webguest38are there rockbox manuals in spanish?
05:19:52LloreanThe only official ones are in English, sorry
05:23:53 Quit webguest38 ("CGI:IRC (EOF)")
05:24:23 Quit Horscht ("Snak 5.3.3 Unregistered copy. Evaluation period is over. Program will now quit. Thanks for using Snak.")
05:43:33 Join MobileShaid [0] (
05:46:21JdGordonkugel: it only shifts if the title is shown and yes that is pretty nescacery
05:48:49kugelJdGordon: Hmm...let me look
05:49:12kugelI think it's only when pointer line selector is used
05:49:27kugelthen the list gets shifted for title text, and for pointer selector
05:49:48kugelshifted twice, and a SCROLLBAR_WDITH empty space
05:51:33kugelJdGordon: yea, I was refereing to that situation. Sorry, that I wasn't clear enough
05:52:00kugelEverything is fine with bar selectors
06:02:39 Quit Shaid (Read error: 110 (Connection timed out))
06:02:40 Nick MobileShaid is now known as Shaid (
06:08:23*JdGordon plans to have something more awesomer than that later tonight
06:08:43BHSPitMonkeyscorche, you use facebook at all?
06:09:11BHSPitMonkeyjust wondering :)
06:09:17kugelJdGordon: Noo, don't!
06:09:26scorcheBHSPitMonkey: that is more for the other channel anyway
06:09:31JdGordoni cant help who I am...
06:09:34kugelJdGordon: :)
06:09:41scorcheJdGordon: sadly
06:09:50BHSPitMonkeyscorche, not trying to socialize, I was wondering about something else
06:09:54BHSPitMonkeybut you're probably right still
06:15:39 Join Jon-Kha [0] (
06:37:34kugel FS #8799, going to bed now.
06:38:43Lloreankugel: You should never reset settings on loading a new .cfg file.
06:39:00 Quit kugel ("ChatZilla 0.9.81 [Firefox]")
06:43:34 Quit midgey ()
06:47:35 Join ol_schoola [0] (
06:50:04 Join hannesd [0] (
06:55:33 Quit jas6180 (Read error: 110 (Connection timed out))
07:01:06***Saving seen data "./dancer.seen"
07:01:43 Quit axionix (Read error: 104 (Connection reset by peer))
07:02:26 Join axionix [0] (
07:03:38 Join Infausto [0] (n=c8787895@gateway/web/cgi-irc/
07:03:44 Part toffe82
07:04:23Infaustoanyone in home?
07:04:36scorche130 users...
07:04:46Infaustoyes, but anyona active
07:05:01Infaustoyou, of course
07:05:23scorchejust ask your question and people will see it eventually...perhaps even answer it
07:06:03Infaustook, clear and simple
07:06:14 Nick fxb__ is now known as fxb (
07:06:14Infaustoa litte question
07:06:29 Quit simonrvn ("Sage, Tobacco, Sweetgrass, Cedar")
07:06:32Infaustoy search in ofrum but a not found the answer
07:06:51 Quit hannesd (Read error: 148 (No route to host))
07:07:07Infaustorockbox support in anny level dual core?
07:07:32Infaustowhit a special driver?
07:07:43Infaustoor apply a pathc to toolchain?
07:10:02Infaustoand you can explain to mi how rockbox support that feature... just a litte answer, or give me a link to read about that
07:10:58scorchewell, what are you looking for? said "in any level"...technically sleeping the other core would "support dual core to a certain level"
07:11:21scorcheare you asking for a specific purpose? to do a certain thing?
07:11:52Infaustoyes, you rigth, but a answer for do any task
07:12:11scorche"any task" being...
07:12:46 Join Keypad2 [0] (
07:13:12Keypad2Can any one here tell me if my Ipod nano is second gen or first gen ?
07:13:49scorchewhat is the storage capacity?
07:13:56Keypad24 GB
07:14:23 Nick Keypad2 is now known as Keypad (
07:15:01scorchedoes it have a plastic white or black front with a shiny silver back, or is it one colored metal piece?
07:15:15LloreanInfausto: It uses it for playing video faster, by doing some of the work on the other core.
07:15:48KeypadIts white plastic
07:15:56Infaustoaim looking start with develop of any piece of software for an 5g ipod... and because this player have a dual core arm7, aim wondering how a can program the task in both cpu's
07:16:28LloreanInfausto: You'd have to look at the code in the mpegplayer plugin.
07:16:34scorcheInfausto: ah....that was the type of specific question is was after...have a look at how mpegplayer does it in the code ;)
07:16:58scorcheKeypad: in the front and a shiny silver back or all around?
07:18:13KeypadConfused :(
07:18:34scorcheKeypad: does it look like this?
07:19:02Infaustothx llorean, i take a look for this code
07:19:15Keypadscorche: Looks like this
07:19:16Infaustoand thx scorche too
07:19:26Keypadno silver back
07:19:35LloreanKeypad: Metal like that, with rounded edges?
07:19:35scorcheKeypad: color or monochrome screen?
07:20:15 Quit parafin (Read error: 110 (Connection timed out))
07:22:57 Join ol_schoola_ [0] (
07:22:57 Quit ol_schoola (Read error: 104 (Connection reset by peer))
07:23:33LloreanKugel probably would annoy me a lot less if I didn't know he had a custom build he maintains.
07:24:20LloreanI mean, I know his intentions are good, but evidently he thinks he's competing with us?
07:25:00*Llorean had forgotten he also maintained a build until he checked abi just now.
07:25:44scorcheso you *didn't* know earlier? ;)
07:26:11Infausto.. can i ask what you are talking about?
07:26:20scorcheInfausto: long story
07:26:20LloreanI did know, lots earlier. But I'd forgotten about it recently until I visited ABi, it didn't rank high in my memory.
07:27:13Infaustojeje, i imagine that...
07:30:00*Llorean wonders if there's some way to work "with" the custom build authors.
07:30:12LloreanI still kinda wish they'd stick to patches that were not rejected, at least.
07:30:54scorchegood luck with that
07:31:27 Nick miepchen^schlaf_ is now known as miepchen^schlaf (
07:37:09Keypadscorche: sorry about that dinner was up :), monochrome
07:37:26scorcheKeypad: then you dont have an ipod nano at is an ipod mini
07:38:23 Nick fxb is now known as fxb__ (
07:40:07Keypadscorche: omg. typo...
07:40:36Keypadscorche: sorry about that. I wanted to ask if my "MINI" is 1st or 2nd gen.
07:41:01scorchedoes it say the battery capacity on the back?
07:42:30scorchethen it is a second gen
07:42:37Keypad5-30 Vdc
07:42:46Keypad1.0 A
07:42:50scorchei meant storage ;)
07:42:58scorchemy own typo :)
07:44:25KeypadIts 4 GB
07:44:31KeypadSo That doesnt help :(
07:44:40scorcheyes, but i mean does it say it on the back?
07:44:45 Quit Shaid (Read error: 110 (Connection timed out))
07:45:11scorchethen it is a first gen
07:45:20KeypadThanks for that
07:46:22 Join simonrvn [0] (i=simon@unaffiliated/simonrvn)
07:52:33 Join jas6180 [0] (
07:53:40 Quit jas6180 (Remote closed the connection)
07:54:34 Join jas6180 [0] (
07:57:09 Join jas6180_ [0] (
08:00:29 Quit argumentD (Read error: 113 (No route to host))
08:02:42 Join parafin [0] (
08:07:06 Quit miepchen^schlaf ()
08:08:19 Quit jas6180_ (Remote closed the connection)
08:08:48 Join jas6180_ [0] (
08:09:53KeypadOh FFS
08:09:58KeypadIpod is asking for a wall charger
08:16:40KeypadApple software restore is fail
08:16:58KeypadI just wanted to clean all the songs off it....
08:19:06amiconnjhMikeS: +2KB of code just for the scheduler?
08:21:34amiconnHmm, and very strange: coldfire, PP5002 and PP5020 see ~+2KB while PP5022/5024 and other ARMs see ~+2.7KB
08:23:56 Quit jas6180 (Connection timed out)
08:25:14 Join GodEater [0] (i=c2cbc962@rockbox/staff/GodEater)
08:25:26 Quit JdGordon ("Konversation terminated!")
08:25:40KeypadNevermind ipod stopped being ass after I plugged it in and out of the USB 9 times
08:29:57 Quit Infausto ("CGI:IRC")
08:30:02 Join Rob2223 [0] (
08:30:36 Join JdGordon [0] (
08:34:42jhMikeSamiconn: besides the fact that this +/-1024 thing seems to still be there, the difference is explicable since the inlined swp corelock take more bytes to use than the corelock calls but are much, much faster
08:37:57amiconnI only compared the binsizes, not the total sizes
08:38:27amiconnThe +/-1024 is due to a by-2048 alignment of an usb stack related array
08:38:40jhMikeSI thought that was removed and the audiobuffer used
08:39:07 Quit BigBambi (Read error: 113 (No route to host))
08:39:15amiconnNo, and iiuc that buffer must be always usable
08:39:51amiconnIt could be put near the end of ram though, directly before the codec ram, by means of the linker script
08:40:22amiconnThis way it would always be aligned, and hence not waste 1024 bytes on average
08:40:44amiconnThe audio buffer is used as the transfer buffer iirc
08:41:08jhMikeSwould be better since that's just a monster alignment
08:45:07 Join Mathiasdm [0] (
08:47:31 Join Bagderr [0] (n=daniel@gateway/web/cgi-irc/
08:48:07 Quit Rob2222 (Read error: 110 (Connection timed out))
08:48:32jhMikeSthe main irritant in implementing that is there's three functions with really similar overall procedure but not similar enough to combine them into one and make it fast at the same time
08:51:15amiconnfirmware/target/arm/usb-drv-pp502x.c line 321
08:54:55 Quit Mathiasdm ("Yuuw!")
08:55:01 Join petur [50] (n=petur@rockbox/developer/petur)
08:56:31 Join Mathiasdm [0] (
08:58:03Keypadmy ipod mini is having problems puting music onto the drive : S
08:58:11KeypadStupid micro drive dont die on me.
09:01:08***Saving seen data "./dancer.seen"
09:01:54 Quit cool_walking_ (Remote closed the connection)
09:04:02LloreanjhMikeS: I will say your change seems to have improved responsiveness on my Sansa, though it's purely a subjective observation.
09:05:46amiconnjhMikeS: Regarding sw corelock - didn't you say the PP5002 wouldn't need it?
09:05:55jhMikeSLlorean: quite possible since I'm aiming at that :)
09:06:48jhMikeSamiconn: I thought so but mpegplayer was freezing on me if not using it but the SPC codec was working fine. I'm not sure what was going on there.
09:06:48LloreanjhMikeS: It seems to have fixed (or reduced greatly) the very problem I complained to you earlier about.
09:07:59jhMikeSamiconn: of course I might have neglected something simple like forgetting to update the plugin. it's a simple thing to test by changing the #define in config.h
09:08:43*amiconn should probably test that later
09:10:56*amiconn should also probably test moving switch_thread to iram once more, and remove the sleep alignment hack
09:11:12amiconnMaybe I goofed and the code didn't end up in iram somehow...
09:12:01 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
09:13:17jhMikeSamiconn: given that the problem appeared exactly at the commit that took it out I'm betting it's just a cache line fetch happening at the moment it sleeps
09:13:41 Join ender` [0] (
09:15:53amiconnYeah, and since cache controller problems seem to be common on PP...
09:16:14jhMikeSLlorean: you might try scrolling around really quickly and extensive during playback and it the expense to the UI due to raising the codec priority should happen softly instead of suddenly
09:17:11amiconnjhMikeS: That should then be visible with cube.rock switched to high speed as well
09:17:42jhMikeSit is, and fire and others
09:17:48amiconnWith the previous priority code, it alternated between 2 speeds when run this way during playback
09:18:27LloreanjhMikeS: I scrolled through a thousand element list, from start to finish, while music was playing, and it seemed better than before in that list.
09:18:53amiconnjhMikeS: Did you test on cf?
09:20:03jhMikeSamiconn: tested on everything I've got: SH (no prio of course), CF, ARM
09:21:28jhMikeScube will change speeds but not the jerky stop for awhile and run
09:23:45 Quit OlivierBorowski (Read error: 113 (No route to host))
09:24:28 Nick merbz2 is now known as merbzt (
09:29:47amiconnjhMikeS: test_codec needs to be adapted...
09:30:40*jhMikeS says "oh yeah" and has a look
09:30:59amiconn"test_codec.c:681: warning: assignment from incompatible pointer type"
09:33:56jhMikeSI guess uintptr_t * is the correct change for codec_stack
09:38:01Keypadstupid hitachi drive
09:38:05 Quit Mathiasdm ("Yuuw!")
09:38:08KeypadI fix one thing in the ipod mini
09:38:14Keypadand something else goes wrong
09:39:54 Join Mathiasdm [0] (
09:40:31*jhMikeS thinks the audio thread could just be asked to call a callback instead of this stack stealing business which is safer to do anyway
09:41:24 Join stripwax [0] (
09:43:49jhMikeSwell, codec thread actually...or frankly another core thread could be temporarily abducted by a plugin
09:44:40jhMikeSs/another/any other/ :\
09:46:43 Quit stripwax ("Miranda IM! Smaller, Faster, Easier.")
09:47:12 Quit Mathiasdm ("Yuuw!")
09:50:38*jhMikeS can surely save some bytes by optimizing IRQ level setting and getting better speed to boot (cut 216 bytes off H120)
09:51:39 Join Mathiasdm [0] (
09:52:11markunWhere should STUB be defined which is used in system.c?
09:55:09jhMikeSI suppose in the config*.h. I'm not even sure why I put that in that #define except to use the default "ignore div0 handler" provided by gcc. Where else is it used?
09:56:45jhMikeSI think some interrupt handlers had stubbing in crt0*.S at the time
09:57:46markunI'm trying to compile a meizu m6 sim, but it tries to compile this ARM code. Can you see what I forgot to do?
10:00:31jhMikeSIf the sim is being compiled then the CPU type shouldn't be defined
10:02:21markunit's inside the #ifndef SIMULATOR in config-meizum6sl.h
10:03:35markunah, I think I know what's wrong
10:05:58markunfixed (forgot to define S5L8700, so it defaulted to 0)
10:06:56Bagderra classic...
10:15:34*JdGordon has the menus working in a specified viewport + ability to disable the status/button bar which should be nice in plugins
10:17:40 Join Keypad2 [0] (
10:19:39 Join moos [0] (
10:29:33 Join OlivierBorowski [0] (
10:33:03 Join gevaerts [0] (
10:34:35 Quit Keypad (Read error: 110 (Connection timed out))
10:35:53 Quit Keypad2 (Remote closed the connection)
10:36:07 Join Keypad2 [0] (
10:43:12 Join ManfredMeller [0] (
10:44:57 Quit XavierGr (Nick collision from services.)
10:45:10 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
10:49:41JdGordonamiconn: should the main menu have the buttonbar?
10:51:03amiconnAll core menus should have the button bar
10:51:54amiconn(only if enabled of course)
10:52:28 Join mobin [0] (i=mobin@
10:52:38linuxstb_What does disabling the button bar do? Does it disable the buttons, or just not display the labels?
10:53:47 Join Rincewind [0] (
10:54:09JdGordonjust hides the labe;s
10:54:12amiconnIt disables the bar
10:54:26JdGordondepends who you ask
10:55:06amiconnThe button bar is just a kind of help text for what the soft buttons do
10:55:58amiconnThe button do work independent of button bar state of course
10:56:00 Join pixelma [50] (i=pixelma@rockbox/staff/pixelma)
11:01:01pixelmalinuxstb: good... morning. Did you already start converting the remaining cabbiev2 monochrome wps which are not using the bitmap strip?
11:01:10***Saving seen data "./dancer.seen"
11:02:09linuxstb_pixelma: Not yet, no.
11:02:15linuxstb_(and good morning!)
11:03:22 Join pondlife [50] (n=Steve@rockbox/developer/pondlife)
11:03:43pixelmashall I do that then?
11:04:05linuxstb_Please do.
11:04:15linuxstb_I think there are only two left.
11:04:59pixelmayes, the Archos and the M:Robe100 version. And since I created the two I feel responsible ;)
11:05:38pondlifelinuxstb:Having updated to current SVN my own WPS has parts missing. The sim −−debugwps option doesn't seem to show any obvious errors, but I'm not really sure what to expect.
11:06:18linuxstb_I assume you're not using viewports (%V tags) ?
11:06:18JdGordoncan someone please try on a button bar atrget (or sim..) and tell me if it crashes and/or works?
11:06:36linuxstb_pondlife: Can you pastebin your .wps, and tell me exactly what isn't working?
11:06:36pondlife is the debugwps output.
11:06:43pondlife.wps to follow, hang on a mo
11:07:13linuxstb_And which target is this?
11:07:30JdGordonamiconn: ^ plz
11:08:01pondlifeWPS is at - target is H300
11:08:22pondlifeProblem is that the entire status bar portion is no longer displayed
11:08:51pondlifeI'm not very familiar with WPS syntax, but this one's been good to me for many months
11:09:12linuxstb_pondlife: Your %xl tags are broken - the three extra parameters are illegal...
11:09:23linuxstb_I think one of my commits made the %xl parsing stricter.
11:09:38pondlifeAh, wonder what they were for...
11:10:12*pixelma just wondered about that too
11:10:29LinusNhmmm, i tried the latest build for the X5, and all the images are gone in the wps...
11:10:47pondlifeI'd guess colour-related...
11:10:49linuxstb_LinusN: The _very_ latest? I fixed that bug last night.
11:10:58linuxstb_(or so I thought...)
11:11:17pixelmaworked fine in an X5 sim this night
11:12:03pondlifelinuxstb_: For my info, is there a line in the debugwps output that shows which tag is invalid?
11:12:10linuxstb_LinusN: Does your installed version of cabbiev2.wps contain this change?
11:12:43LinusNhmmm, resetting the settings helped
11:13:22pixelmaresetting settings or a simple reboot?
11:13:30LinusNreboot did not help
11:13:36*pondlife thinks that reset settings should backup the old config.cfg...
11:13:46pondlifei.e. just rename it
11:13:58LinusNgood idea
11:14:15 Part mobin ("Leaving")
11:14:25*Llorean seconds that.
11:14:39*JdGordon 3rds it
11:15:01*Bagderr goes 5th!
11:15:09*JdGordon getting mighty pissed off with this BB code
11:15:12pondlifeBut no-one codes it ... ;)
11:15:14linuxstb_pondlife: I think it was just mis-parsed - treating the images as if they had 255 sub-images, and I don't think debugwps shows the number of subimages for an image.
11:15:21pondlifeAh ok.
11:15:26*JdGordon will code it up a bit later
11:15:32pondlifeI was hoping for the word "error" to appear
11:15:44 Join Ranjith [0] (i=mobin@
11:16:16linuxstb_pondlife: The parser is still quite lax about a few things - e.g. it quite often skips to the end of a line without checking what is there. That's what happened in your case.
11:16:26pondlifelinuxstb_: Thanks, that's working nicely now.
11:17:00*pondlife still wants a themeable status bar common to both menus and WPS.... least.
11:17:21linuxstb_pondlife: Like this?
11:17:44pondlifeYes, but at the top!
11:18:09Lloreanlinuxstb_: Now you know that's not what he wanted. You stated very clearly that allows for a unique one independent of the WPS one as well. :-P
11:18:09linuxstb_Well, it's themeable... ;)
11:18:35 Quit ManfredMeller ()
11:18:44linuxstb_pondlife: You can have one at the top, and one at the bottom if you wish...
11:19:01pondlifeThe important thing is that the WPS doesn't need to have the status bar defined again.
11:19:02peturhmmm pixies...
11:19:22pondlifeWhere Is My Status Bar?
11:19:22*pixelma uses some kind of status box in the right upper part of her WPS... what about that? =)
11:19:29linuxstb_pondlife: You may want to read my description of that patch -
11:19:33Lloreanpondlife: Basically, the idea is that viewports within the WPS can be flagged for display in the WPS, in the list, or both.
11:19:49JdGordonlinuxstb_: is that a mock up? or a patch?
11:19:55linuxstb_JdGordon: A patch
11:20:08pondlifeCertainly I like the idea of the same parser
11:20:11linuxstb_See the IRC log I just linked to.
11:20:17pondlifeYep, reading.
11:20:35 Quit axionix (Read error: 104 (Connection reset by peer))
11:20:37*Llorean wonders when the WIP hack will start showing up in unsupported builds.
11:21:17 Join axionix [0] (
11:21:44pondlifeLlorean: Depressingly soon, I suspect.
11:21:57linuxstb_It's far too broken for someone to include I think
11:22:10pondlifeBest keep it that way until it's ready :)
11:22:30*pondlife points out this is a joke
11:22:32linuxstb_I'll just make sure it conflicts badly with some other patches...
11:22:47pondlifehehe, maybe muitifont
11:22:56pondlifeOr however it's spelt
11:23:52linuxstb_disorganizer has been playing with my patch though -
11:23:59pondlifeOMG! I gotz album art in my status bar!!!!!!
11:25:10pondlifelinuxstb_: Yes, I think you may have just come up with the first good (IMHO) use for viewports.
11:25:56 Join tvelocity [0] (
11:27:06 Quit Keypad2 ()
11:27:10 Join fasmaie [0] (
11:31:25linuxstb_LinusN: So is your X5's theme now displaying correctly after a settings reset?
11:31:50pondlifeJdGordon: Thanks!
11:32:07linuxstb_LinusN: That's odd - I can't think what setting could have affected it...
11:32:09pondlifeJdGordon: Might be better to have it just internal to settings_reset though
11:32:21pondlifeJdGordon: Does that work if I do a boot-time settings reset too?
11:32:31JdGordonno, purposly...
11:32:40pondlifeAlso, no need to expose settings_write_config() then
11:32:41JdGordonI didnt think it made sense to put it there also
11:32:59linuxstb_Why? I thought it was be most useful there - if people accidentally reset.
11:33:07pondlifeMe too
11:33:15JdGordonhmm.. ok maybe
11:33:33pondlifeJust make settings_reset do a rename
11:33:55linuxstb_And why call it ".bak" ? That means you can't select it in the browser.
11:34:17*JdGordon stops volanteering for simple changes
11:35:04*Llorean would call it "previous.cfg" and create it on any load of a .cfg file, not just resets. =P
11:35:25LloreanCovering when you stupidly click on a .cfg file in the filetree too.
11:35:26pondlifeoldconfig.cfg was my vote
11:38:54 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
11:38:58JdGordonok, done again
11:39:25JdGordonjust make sure you dont try loading oldconfig.cfg... it might not like that
11:39:39pondlifeWhy not?
11:39:53JdGordonbecause it will try overwriting oldconfig.cfg before its loaded
11:40:18Bagderruh, but isn't the point with that to allow it to get loaded?
11:40:20pondlifeIt should write oldconfig.cfg with the current config, then reload it.
11:40:24pondlifei.e. no change
11:40:38*Llorean wonders how you're supposed to use oldconfig.cfg without loading it, much like Bagderr
11:40:46Lloreanpondlife: That's useless...
11:40:59LloreanYou need to be able to load oldconfig.cfg without oldconfig.cfg being written.
11:41:03pondlifeIt should only do it in settings_reset IMHO
11:41:05*JdGordon gives up... stupid coding with big bloody headache
11:41:33Lloreanpondlife: I load .cfg files sitting around my filetree far more often than I've reset my settings. UI lag, at least, has caused it.
11:44:01*linuxstb_ suggests thinking first, then coding... ;)
11:44:02Ranjithhello everyone
11:45:00pondlifelinuxstb_: I agree, which is why I never get onto the coding part :)
11:51:18RanjithI'm a student from india im planning to participate in gsoc.I would like to add real audio support in rockbox. I found that the codec mostly used in the real audio container format is "cook" which is already having an open licenced decoder in floating point. why can't we decode this decoded version from the existing one to what, supported in rockbox?
11:51:22linuxstb_How about: 1) Delete config.cfg; 2) Rename oldcfg.cfg to config.cfg; 3) Save settings to oldcfg.cfg; 4) Load config.cfg...
11:52:24 Quit jhulst ("Konversation terminated!")
11:52:24linuxstb_Ranjith: I don't understand your last sentence...
11:53:10RincewindRanjith, rockbox can only do fixpoint math, if the licence of this decoder is compatible with rockbox you would have to convert the floating point routines to fixpoint
11:53:31Lloreanlinuxstb_: Sounds backward...
11:53:37Bagderrbut that's a fine gsoc project I'd say
11:54:54merbztRanjith: the cook decoder in FFmpeg is fixedpoint with the exception of the mdct transform
11:55:25linuxstb_Llorean: Why backwards?
11:55:36merbztRanjith: just replace it and you are set to go
11:55:41Lloreanlinuxstb_: Unless you're talking about special-casing the oldconfig.cfg load, that is.
11:56:31linuxstb_Llorean: Yes, I guess that's what I'm saying...
11:56:37linuxstb_Which doesn't seem nice.
11:56:43LloreanWouldn't it be better to do "write temp.cfg, load (filename).cfg, rename temp.cfg oldconfig.cfg" on settings loads, and not have a special case?
11:56:57Ranjiththats what i had mensioned in the last sentence to convert the very part in the ffmpeg decoder to fixpoint which is supported in rockbox
11:57:00LloreanOr is that extra rename step on every config load a problem?
11:57:55merbztRanjith: the rockbox wma decoder has the mdct already in fixedpoint, so you only need to use that
11:58:38merbztRanjith: the part that is hard is the rm demuxer, it needs to be adapted to Rockbox
11:59:10*linuxstb_ has a half-written (or maybe less...) RM demuxer he intended for Rockbox
11:59:35 Join jas6180 [0] (
12:00:31Ranjithwhy can't we design a demuxer parallel to the one of 'rm'.
12:01:02*amiconn doesn't want an oldconfig :\
12:03:36 Join Thundercloud [0] (
12:04:22*linuxstb_ isn't sure of the need for it either...
12:04:59*pondlife thinks there should be at least 3 stages: consulting, thinking, coding
12:05:09*petur keeps a backup of it anyway
12:05:22pixelmayou can just save your settings...
12:05:25merbztRanjith: ? anyway I wrote most of the FFmpeg cook decoder, there shouldn't be any problems to port it to rockbox
12:05:59linuxstb_pondlife: Plus 4) testing; 5) commit
12:06:13pondlife"at least"... :)
12:06:19linuxstb_The Rockbox 5-step plan.
12:07:13linuxstb_merbzt: Could you predict how efficient Cook is? e.g. compared to WMA.
12:07:37RincewindI propose a step 6: respond to the complaining in IRC ^^
12:07:52linuxstb_Rincewind: Steps 1-5 are designed to avoid step 6...
12:08:29Bagderryou can never avoid that #6... :-)
12:08:37*petur is getting hungry from all this cook business :/
12:08:45*pixelma stress that the order of things is very important (doing 'make && make clean' doesn't help here) ... :)
12:09:11Ranjithmerbzt:so as it is easy to port most of the decoders to rockbox i hope it only matters in designing a demuxer which can properly manage the decoders to give an output efficient as that of 'rm'.
12:09:12merbztlinuxstb_: well cook aka G722.1 is a suitable lowbitrate codec, but as always there are better alternatives
12:09:19*pondlife wants a codec named "biscuit"
12:09:25 Quit fasmaie (Read error: 113 (No route to host))
12:09:37gevaertsYou can't avoid step 6, but you can use the "disabled by default" trick, which allows you to ignore it ;)
12:09:46merbztRanjith: yes and the FFmpeg demuxer is a good reference
12:09:57 Quit jas6180 (Remote closed the connection)
12:10:25 Join jas6180 [0] (
12:10:38 Join PeterMitchell [0] (n=3aa6725e@gateway/web/cgi-irc/
12:11:31 Join XSource [0] (n=XSource@unaffiliated/xsource)
12:11:32 Quit jas6180 (Remote closed the connection)
12:12:21 Join jas6180 [0] (
12:13:04linuxstb_gevaerts: And then slowly remove the #ifdefs by "accident"...
12:13:14Ranjithmerbzt:thanks for your information let me go ahead with my proposal to gsoc, to add a real audio support in rockbox. i will contact 'Davechapman'(corresponding mentor for rockbox) with this proposal.
12:13:17merbztRanjith: hmm, forgot to say, but most parts of the cook fixedpoint code is on the ffmpeg-dev mailing list archives, you need to dig that up
12:13:30BagderrRanjith: that's linuxstb...
12:13:37linuxstb_Ranjith: That's me... I'm not guaranteed to be the mentor, I've just said I would be interested in it.
12:14:34linuxstb_Ranjith: I think it's better for you to just ask questions here.
12:15:28gevaertslinuxstb_: of course :)
12:15:40 Quit jas6180_ (Connection timed out)
12:18:34linuxstb_Ranjith: Have you used Rockbox before, or looked at the Rockbox source code?
12:18:55Ranjithlinuxstb_:sir my plan is to design a demuxer with refference to ffmpeg and one which runs parallel to that of the 'rm'. i believe it can be implemented porting the individual decoders from ffmpeg to the new demuxer
12:20:38Ranjithlinuxstb_:no i havent looked at the source code im interested as i feel im on the right track to get the support for realaudio ready.
12:21:39 Join MethoS- [0] (
12:25:54 Quit Nevtus (Read error: 110 (Connection timed out))
12:30:39 Join DerDome [0] (
12:30:39 Quit PeterMitchell ("CGI:IRC (EOF)")
12:31:50linuxstb_Ranjith: My advice would be to download the Rockbox source and try to understand how codecs are implemented in Rockbox, and what the restrictions are.
12:32:16 Part Bagderr
12:32:20linuxstb_Ranjith: Do you have any experience with C programming, or with embedded systems, or with audio codecs?
12:32:40 Join disorganizer [0] (n=c2785409@gateway/web/cgi-irc/
12:33:32pixelmamy X5 sim doesn't play mp3s anymore ("codec failure") - it worked in the night, made clean even reconfigured... r16791
12:33:35 Quit jas6180 (Remote closed the connection)
12:33:55pixelmacygwin compiled, winXP
12:34:43Ranjithlinuxstb_:i have worked with atmels microcontrollers (mega16) in some projects and have a fair knowledge about the audio codecs.
12:34:51linuxstb_pixelma: I would test r16790...
12:36:00linuxstb_Ranjith: That's good. You should mention that in your application (with more information if you can - e.g. give examples of similar work you have done).
12:36:53Ranjithlinuxstb_:ok for sure.
12:37:39*JdGordon gets fed up with button bar and removes the bloody thing from svn!
12:37:51 Quit jcollie ("Ex-Chat")
12:39:35 Join jas6180 [0] (
12:40:14pixelmalinuxstb: might have been my mistake (again :\ ), trying again now...
12:40:15linuxstb_JdGordon: What's the problem? Isn't it just the same as the statusbar, but at the bottom of the screen?
12:40:37 Part LinusN
12:40:42 Quit jas6180 (Remote closed the connection)
12:41:00JdGordonno, I dont even know what the bloody problem is now... the list vp seems to be overwriting it even with the dimensions set correctly
12:42:03linuxstb_You mean the list items are being displayed over the top of the button bar? Or that something is clearing that part of the LCD?
12:42:57JdGordonits wierd... starting the menu the BB is hidden by the list, but going into a setting and then back into the menu makes the menu show the bb
12:43:24JdGordonits going to be a stupid mistake somewhere.. but my head doesnt want to cooperate tonight
12:44:21JdGordonwe have to get rid of the screens[screen].has_buttonbar? check in viewport.c
12:46:46 Join jas6180 [0] (
12:47:50 Join cool_walking_ [0] (
12:48:25 Join MethoS-- [0] (
12:48:26JdGordonfixed... hopefully...
12:49:09*JdGordon wonders if there really is any point sticking this on flyspray... is anyone going to test it? or just complain after its eventually commited?
12:49:34*disorganizer thinks many people have targets without buttonbar
12:49:35linuxstb_JdGordon: If you don't put it on flyspray, definitely no-one will look at it...
12:49:43gevaertsJdGordon: just follow the 5-step plan ;)
12:50:44 Quit jas6180 (Remote closed the connection)
12:51:31disorganizer@ linuxstb: did you see the crash i posted yesterday in the log with the statusbarportification hack?
12:51:34 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
12:51:41linuxstb_disorganizer: Yes, but I don't care yet ;)
12:52:22disorganizer@ linuxstb : i thought so, just wanted to point it out :-)
12:53:03disorganizerbut to be honest, this thing looks so good "in production" that it would be a shame if its not implemented *g*
12:54:02JdGordonit will be eventaully... one way or another...
12:54:52JdGordondone, finished and fixed!
12:55:30JdGordonamiconn: hopefully the buttonbar issues are all going to be fixed... will you test the patch? (please dont ignore this.. at least say no if your not going to)
12:55:44disorganizerJdGordon : i hope one way (the right one via commit) and not the other (by custom build) ;-)
12:56:16disorganizerafk for some hours
12:56:24 Nick disorganizer is now known as disorganizer_afk (n=c2785409@gateway/web/cgi-irc/
12:56:33 Join jas6180 [0] (
12:56:42amiconnJdGordon: Can't test anything before tonight
12:57:25JdGordonwill you if it goes on flysray?
12:57:43*JdGordon doesnt want to sit on it.... once this is in plugins can get their colours fixed
12:58:07mooshe surely will if svn ;)
13:00:18 Quit jas6180 (Remote closed the connection)
13:01:13***Saving seen data "./dancer.seen"
13:01:29linuxstb_JdGordon: I will probably have a look at it if you post a patch - at least, I'll read the patch...
13:02:57JdGordonfixing up one plugin first to show what needs to be done
13:04:05 Join aliask [0] (n=aliask@rockbox/developer/aliask)
13:04:29 Quit MethoS- (Read error: 110 (Connection timed out))
13:07:12 Quit Rincewind ("bye")
13:08:02 Join jas6180 [0] (
13:08:35 Quit XSource ("Quit!")
13:09:18JdGordonhehe stupid themes
13:09:28 Join webguest59 [0] (n=51858ac3@gateway/web/cgi-irc/
13:09:57webguest59hi there
13:10:58 Join webguest96 [0] (n=51858ac3@gateway/web/cgi-irc/
13:10:58 Quit webguest59 (Client Quit)
13:11:07 Quit cool_walking_ (Read error: 110 (Connection timed out))
13:11:42 Quit webguest96 (Client Quit)
13:12:17 Join MarcinMikolajczy [0] (n=5316e84c@gateway/web/cgi-irc/
13:14:19 Quit XavierGr (Nick collision from services.)
13:14:29 Quit DaCapn (Remote closed the connection)
13:14:32 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
13:14:54JdGordonlinuxstb_: problem: when loading a new theme, every viewport needs reloading (to get the colours)... without explicitly calling and keeping every screens "reload viewport" fucntion how do we tell them to reload?
13:15:34JdGordonI'm thinking a global_status var which would be set to -1 when themes are loaded and each viewport would watch "its" bit and set it to 0 once its been reloaded
13:15:39JdGordonwould that work?
13:16:19JdGordonthat would give us up to 32 viewport parents, and each could then manage its sub viewports easily enough
13:16:41linuxstb_I would have thought that the viewports should initialise themselves based on the current settings when the screen is called.
13:18:01JdGordonthe menu, list, SB and BB viewports would all be running when a theme is loaded
13:18:38JdGordonhmm.. we could use the event system for this I guess
13:18:44JdGordonbit of a waste thoguhg
13:22:30pondlifeJdGordon: An event sounds sensible, better than the bitset.
13:22:50 Quit MethoS-- (Read error: 110 (Connection timed out))
13:26:14JdGordonpatch without that last problem is in 8800
13:26:15linuxstb_Wouldn't it be better to consider the SB and BB as optional parts of the menu widget? That way, the menu just needs to re-init itself.
13:26:58pondlifeCertainly the SB would make sense there
13:27:11pondlifeWhat does the BB display? Is it tied to a menu?
13:27:39pondlifeOK, inclusion sounds sensible then
13:28:38 Join MarcGuay [0] (n=d8ef4bec@gateway/web/cgi-irc/
13:30:10 Quit XavierGr (Nick collision from services.)
13:30:22 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
13:30:50MarcinMikolajczyCould I ask for edit permission of Meizu M6 port page for me (MarcinMikolajczyk in TWikiUsers)?
13:32:05markunhi MarcinMikolajczy!
13:32:09markunI'll give you permission
13:32:12linuxstb_MarcinMikolajczy: Done
13:32:15markunwhat are you going to add?
13:32:28 Nick fxb__ is now known as fxb (
13:32:31 Join CyBergRind|w [0] (n=cbr@
13:32:39MarcinMikolajczylinuxstb_ Thank you
13:33:15MarcGuayJdGordon: That last commit creates a file called ".rockboxoldcfg.cfg" in the root directory... Close, but no cigar...
13:33:29JdGordonyeah, i noticed that
13:33:33 Quit [CBR]Unspoken|w (Read error: 110 (Connection timed out))
13:33:37JdGordonfixing it so its more useful
13:33:51markunMarcinMikolajczy: did you find out something new?
13:35:02MarcinMikolajczyno, i'm just going to add some information about CPU registers and maybe GPIO ports
13:35:45pondlifeNico_P: Is still an issue?
13:36:03Nico_Ppondlife: I couldn't tell you
13:36:10pondlifeOK, just wondered
13:36:37Nico_Pbest is probably to ask on the task
13:36:37 Quit MarcGuay ("CGI:IRC (EOF)")
13:37:29pondlifeNico_P: Have done...
13:37:43Nico_Pabout FS #8601, I think I need to rething the whole track skipping procedure
13:37:56*linuxstb_ wonders what solution JdGordon has planned
13:38:08JdGordonfor which?
13:38:10Nico_Paudio_check_new_track is enough of a mess already
13:38:22linuxstb_JdGordon: "fixing it so its more useful"
13:38:22pondlifeNico_P: Perhaps the track skipping stuff could be scrapped?
13:38:33Nico_Pwhat do you mean?
13:38:37pondlifeNico_P: I mean, so the API becomes "play track N"
13:38:42JdGordonlinuxstb_: trying to get it to use create_numbered_filename()
13:38:47pondlifeAnd the UI determines the value of N
13:39:00 Join PaulJam [0] (
13:39:00pondlifeJdGordon: Urgh, lots of old configs?
13:39:10*linuxstb_ is glad he asked ;)
13:39:14Nico_Ppondlife: it would be good, yes.
13:39:22JdGordonsolves the problem of not being able to load an old one
13:39:25pondlifeNico_P: And, of course, don't clear out the buffer
13:39:37pondlifeJdGordon: KISS
13:39:54*linuxstb_ is starting to think that KISS is reverting this idea
13:40:01*JdGordon tending to agree
13:40:03*pondlife agrees and he had the idea.
13:40:20pondlifeBut I only wanted a rename in settings_reset()...
13:40:38pondlifeNot a config history
13:41:28pondlifeNico_P: Could this also mean that selection from a playlist no longer does a stop and restart?
13:41:47Nico_Pif done right, hopefully
13:41:53JdGordonit also needs to be told not to write out the config on load
13:41:55*GodEater waits for someone to suggest porting rcs to rockbox so the config file can be version controlled
13:42:11*Nico_P had the same idea as GodEater :)
13:42:11*pondlife hopes noone suggests that.
13:42:26linuxstb_How about porting rcs to rockbox so the config file can be version controlled?
13:42:35GodEaterwhat a great idea
13:42:40GodEaterwhy didn't I think of that
13:42:42Nico_Pwhy not git :)
13:42:49*pondlife wondered..
13:42:57GodEatersledgehammer to crack a nut
13:43:02*JdGordon reverted
13:43:07JdGordonsomeone else can do it properly
13:43:16pondlifeGodEater: Sledgehammers are for cracking an iPod, surely
13:43:28*GodEater prefers an axe
13:43:33JdGordonhammer time!
13:43:46pondlifeYou can't touch this iPod.
13:43:49*GodEater looks for his big trousers
13:45:22*JdGordon hopes someone gets a giggle out of his commit message
13:45:37*pondlife giggles
13:46:10*GodEater chortles
13:46:18*Bagder has his stone face on
13:46:25pondlifeJdGordon: Your first green delta coming up?
13:46:25*linuxstb_ goes for lunch
13:46:34*gevaerts wants to know what befroe means
13:46:36*pixelma likes "befroe" most :P
13:47:10*GodEater goes to see how complicated rcs's code base is
13:47:27Nico_P"well we suck!" is going to be on the front page now :)
13:47:52*GodEater can confirm it is indeed on the front page.
13:48:11markunpondlife: ;)
13:48:14BagderI disagree with that statement!
13:48:25gevaertsNico_P: you know what you can do to get rid of it there ;)
13:48:35Bagderof course it doesn't say who "we" are...
13:48:38*pixelma is close to pushing that line a bit further down...
13:48:43*linuxstb_ looks to pixelma for some big commits
13:49:07Nico_Pwe have to hide it befroe anyone realizes it's true!
13:49:12*JdGordon wonders about the lack of a sense of humour in this chan
13:49:18pixelmalinuxstb: almost finished, just testing the Archos version and then do all the add/remove business
13:50:01 Quit MarcinMikolajczy ("CGI:IRC (EOF)")
13:50:04*JdGordon btw blames whoever said the filename stuff should go directly into settings_reset() which for all this
13:50:05 Nick disorganizer_afk is now known as disorganizer (n=c2785409@gateway/web/cgi-irc/
13:50:15JdGordondisorganizer: lier! you said many hours!
13:50:57disorganizerwell, the team meeting was shorter than expected
13:51:51pondlifeJdGordon: I guess I should code up a working one sometime then... although first I need to see exactly how upset people would be about resetting settings dumping an oldconfig.cfg onto their DAP.... amiconn?
13:52:12*gevaerts wonders what a long-term delta line graph would look like
13:52:28pondlifeWe have a lines-of-code graph somewhere.
13:52:37peturohloh you mean
13:52:47disorganizerJdGordon: will #8800 get some more explanation?
13:53:02JdGordonpondlife: just remember to watch out for settings_reset() in main()...
13:53:04JdGordondisorganizer: no
13:53:13 Quit XavierGr (Nick collision from services.)
13:53:20JdGordonits not for users... its only there as a curtosiy
13:53:25 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
13:53:34linuxstb_A what? ;)
13:53:35*JdGordon cant spell tonight
13:53:44*pondlife niether
13:53:54linuxstb_JdGordon: t o n i g h t
13:54:08*linuxstb_ apologises for the very old joke
13:54:11JdGordonhay! i got that right!
13:54:35pixelmapondlife: neither? ;)
13:55:21pondlifeJdGordon: Thanks for the heads-up... that's a slightly odd use of settings_reset()...!
13:55:41JdGordonnot really... it puts it into a nice expected state
13:55:54pondlifeYes, but doesn't settings_load() do the same?
13:56:18pondlifeI suppose some "settings" aren't persisted...
13:56:50JdGordonyeah, config.cfg only saves the changed settings
13:57:02pondlifeAh, yes
14:01:48pixelmaJdGordon: why is the revert not so green? Or didn't you want to revert completely?
14:02:02 Quit jas6180 (Remote closed the connection)
14:02:05pondlifestatus_save() and settings_save() look quite similar. I see the difference in functionality, but wonder why both are needed?
14:02:18 Join jas6180 [0] (
14:03:20*JdGordon directs pixelma to gcc and svn diff
14:04:00*JdGordon patched from the origional diff so should have got everything
14:04:32*gevaerts also recommends diffstat for a quick overview
14:04:34 Quit kushal_12_27_200 ("This computer has gone to sleep")
14:05:40JdGordonpondlife: yeah, they could be merged... but settings_save() is simpler than settings_save(true);
14:05:56 Join michael1486 [0] (
14:06:33pondlifeWhat's the difference in usage? status_save() just forces an NVRAM update, but settings_save forces all to be persisted?
14:06:43pondlifeNeither will trigger a spinup though.
14:06:46*disorganizer wonders if its somehow possible for the lcd-firmware driver to check in which context the user is at the moment, without having to implement new flags or variables
14:07:57michael1486does anyone know if you can change the usb initiation sequence?
14:08:12GodEatermichael1486: to what ?
14:08:32JdGordonpondlife: yes, neither will trigger a spin up
14:08:48pondlifeSo why not just use settings_save for both?
14:09:13JdGordonits slower if you only need to save the status
14:09:20pondlifeHmm, suppose so.
14:09:23michael1486an ipod mini
14:09:42JdGordonpixelma: crap, appologies.. i reverted the wrong revert
14:09:51gevaertsmichael1486: what exactly do you want to achieve ?
14:09:52 Join nplus [0] (n=nplus@141.25.Globcom.Net)
14:10:03JdGordonwe are back to /.rockbox/config.cfg.bak
14:10:40*Bagder ordered a Meizu M6
14:10:41michael1486hardware fuzzing by emulating a usb keyboard before initiating it as a usb drive.
14:11:14pixelmaJdGordon: was about to complain :P
14:12:01*disorganizer thinks JdGordon was right
14:12:01 Quit Galois (Read error: 110 (Connection timed out))
14:13:29gevaertsmichael1486: you could play with the files in firmware/usbstack/, but I still don't really understand what exactly you want. On USB you enumerate as one device class (or several, for composite devices), but you can't change that without a (logical) disconnect
14:14:20*gevaerts should really clean up his local usb changes and commit them
14:15:06michael1486do you know where the initiation sequence is located in the code?
14:16:07gevaertsThere is no single place. usb_core.c basically handles all incoming usb requests, and dispatches them to the correct class drivers (or handles them locally if appropriate)
14:16:37michael1486i need to change what goes out....
14:17:15michael1486i think i found it in usb_class_driver...
14:17:37gevaertsWhat goes out when ?
14:17:38 Join LambdaCalculus37 [0] (n=44a0430f@gateway/web/cgi-irc/
14:17:51pondlifeIs there an option to do a make with "warnings as errors" ?
14:17:52michael1486when you connect to a computer
14:18:38dionoeapondlife: -Werror
14:18:52 Join PeterMitchell [0] (n=3aa6725e@gateway/web/cgi-irc/
14:19:03 Quit PeterMitchell (Client Quit)
14:20:29gevaertsThere is no single "when you connect to a computer" event. The device has to respond to host requests. Where you need to make changes depends on what host requests you want to handle differently
14:21:16markunBagder: great!
14:21:17markunwhich one?
14:21:30BagderSL I beleive
14:21:56Bagder8GB white
14:22:04michael1486when you connect to a computer and it wants to know what drivers to use...i want to add in some usb keyboard stuff to make it seem to be a hub.
14:22:04markunah, I have the black one
14:22:16pondlifedionoea: Isn't that a gcc option, not a make one?
14:22:23Bagderthe best deal here happened to be white so I picked that
14:22:38dionoeapondlife: ah, you meant make warnings?
14:22:46disorganizer@ jdgordon: if you do one more commit you are the only one on the frontpage at the moment :-)
14:22:46Bagderwe'll see, they didn't have it in stock right now so I guess it can still change
14:23:07*dionoea didn't know that make had warnings
14:23:08pondlifedionoea: No, all warnings (cc + ld + make, I guess)
14:23:23pondlifei.e. detect yellow
14:23:36*Bagder runs off
14:23:52dionoeaWell those are cc + ld only I guess. So -Werror should work (in CFLAGS and LDFLAGS)
14:24:17gevaertsmichael1486: that's the configuration descriptor. But a hub and a keyboard are not exactly the same... This is mostly described in chapter 9 of the usb spec
14:24:45dionoeaAh no, looks like it's −−fatal-warnings for ld
14:25:13michael1486no i want to make it seem like a hub with an ipod AND a keyboard on it...
14:25:26pondlifedionoea: Thanks
14:26:39gevaertsmichael1486: you won't be able to do that (harware limitations), but you could make it a composite device that has both a disk and a keyboard. You need to implement a HID driver for that.
14:28:26michael1486couldent i just send the keyboard initiation sequence and data before the ipod initiation sequence?
14:29:04gevaertsThe first problem is that the current stack doesn't support interrupt transfers yet (but I expect to have those pretty soon). After that, I'd suggest to have a look at usb_serial.c for an example of a reasonably simple driver
14:37:00michael1486got to go...
14:42:40 Quit desowin ()
14:46:24 Part Ranjith ("Leaving")
14:53:04 Join My_Sic [0] (n=3ea13b0f@gateway/web/cgi-irc/
14:54:09 Join kugel [0] (n=kugel@unaffiliated/kugel)
14:54:20 Join mf0102 [0] (n=michi@
14:56:57*JdGordon points randomly at FS #8800 then goes to bed
14:57:13 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
15:01:16***Saving seen data "./dancer.seen"
15:02:02 Quit My_Sic ("CGI:IRC")
15:03:12pixelmaRockbox feels more laggy on my Ondio now. My first impression is because (1) for some reason clearing the screen between bootlogo and file browser which I set as start screen is very visible and (2) that delay in the statusbar update when stopping playback (leaving the WPS) is more obvious now because the statusbar area is not blank for a while (as it was before) but shows the part of my WPS before the status bar gets drawn
15:08:57pondlifeWho is "wavey"?
15:09:11pondlifeThe (c) in settings.h is interesting.
15:09:21 Join meck [0] (
15:10:02pixelmaI thing I read the nick once in a very old log (just skimmed the history)
15:10:09amiconnpondlife: There's a very important difference between settings_save and status_save, apart from the latter being faster
15:10:15linuxstb_pondlife: svn blame points at zagor for that line...
15:11:13amiconnYou're right that neither one *causes* a spinup. But the former always *registers* a spinup, while the latter only does that on targets without true nvram
15:11:26peturno, Zagor moved it to the location it is now
15:12:17 Quit XavierGr (Nick collision from services.)
15:12:28 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
15:12:51*linuxstb_ searches the cvs email archive and now blames Stuart Martin
15:14:49*disorganizer still wonders whether there is an easy way for the lcd-firmware-driver to find out in which context the gui is in (to disable backlight for scrollwheel only in certain contexts, aka the wps).
15:15:58amiconnThat has nothing to do with the lcd driver at all
15:16:10amiconnInput is handled in the button driver
15:16:26*amiconn also wonders what disorganizer is trying to achieve
15:17:01 Join sreyas [0] (i=root@
15:17:16disorganizeryes it does. for sansa the scrollwheel enables the backlight from the firmware driver :-) see #8400
15:17:36disorganizeri want to disable the backlight on volume changes, but NOT when using the scrollwheel in the menu ;-)
15:18:53disorganizerand you are right. its the button-firmware-driver :-) must have something to do with not sleeping enough
15:19:37 Quit sreyas (Client Quit)
15:21:16 Quit aliask ("ChatZilla 0.9.81 [Firefox]")
15:23:37linuxstb_disorganizer: I think you simply need an "enable_backlight_on_scroll" variable in the firmware code (and a function to set it), which is set to true by default, and set to false whilst you are in the WPS.
15:27:32pondlifeOne more (c) question - who's hessu?
15:27:51pondlifesettings.c this time
15:28:35linuxstb_My best guess is Heikki Hannikainen - email address was hessuh at sourceforge.
15:29:21linuxstb_And the originator powermgmt.c ;)
15:29:37linuxstb_^add "of" appropriately
15:31:57 Quit jas6180 (Read error: 104 (Connection reset by peer))
15:32:20 Join jas6180 [0] (
15:33:42disorganizerlinuxstb: i searched for an easier way :-) so no additional variables and settings are required. i hoped for some kind of checking the viewports to find out who is active, but obviously there is nothing.
15:34:26linuxstb_disorganizer: No, the firmware code doesn't (and shouldn't) know what the apps code is doing, unless the apps code tells it.
15:35:56kugelHMm I'm getting a weird compiling error:
15:36:03kugelYesterday it worked without problems
15:37:05linuxstb_kugel: Have you done the usual "make clean" (or delete your build directory) and re-run ../tools/configure ?
15:37:29linuxstb_And you're compiling a clean SVN?
15:37:43*pixelma guesses not
15:38:10kugelOnly FS #8799 applied, I doubt it's causing that
15:39:25kugelI even did "rm -dfr *" in my sim dir, and re-run tools/configure
15:40:08linuxstb_Did you change anything in your environment? e.g. upgrading packages?
15:40:53 Quit Mathiasdm ("Yuuw!")
15:41:09kugelIt worked yesterday, just before I shut down my box. Today, after turning it on, it doesn't work anymore :(
15:41:33linuxstb_I can't find any references to "midiparser" in the Rockbox source - so my best guess is that it's something local in your setup which is broken.
15:41:39linuxstb_When did you last reboot your box?
15:41:53kugelan hour ago?
15:42:02linuxstb_I meant before yesterday...
15:42:26kugelYesterday about 15:00 (cet)
15:43:24 Join Mathiasdm [0] (
15:45:18 Quit Mathiasdm (Client Quit)
15:45:41 Join Mathiasdm [0] (
15:46:22kugelI installed teamspeak yesterday, together with some lib32 (including lib32asound2), can that be the reason?
15:46:29 Join Mathiasdm2 [0] (
15:46:43linuxstb_kugel: You said you didn't change anything...
15:46:52kugelI didn't change anything today
15:46:56linuxstb_As the error relates to libasound, then yes, I would say that could be the issue.
15:47:04kugelI installed that yesterday, when it still worked
15:47:23linuxstb_Until you rebooted I'm guessing.
15:48:05kugellinuxstb_: Well it seems to work now. Weird that teamspeak messes up my rockbox enviroment
15:49:46 Quit corevette (Remote closed the connection)
15:53:07 Quit Nevtus (Remote closed the connection)
15:54:28 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
15:55:50 Join barrywardell [0] (
15:57:51 Join jgarvey [0] (
16:03:38 Join jas6180_ [0] (
16:04:38 Quit Mathiasdm (Connection timed out)
16:14:54*disorganizer wishes he could write c programs
16:15:26markundisorganizer: what would you write?
16:16:38*gevaerts recommends decimal, and starts writing 12 programs
16:18:20 Join SirFunk [0] (n=Sir@
16:18:31gevaertsAssuming "c" is hex
16:18:41 Join Infausto [0] (n=c8787895@gateway/web/cgi-irc/
16:19:48*markun is slow ;)
16:19:57 Quit jas6180 (Connection timed out)
16:20:22*disorganizer writes c now
16:20:56markundisorganizer: what kind of programs would you like to write?
16:22:32*markun feels ignored.. :)
16:22:35disorganizerah well, i only would like to get #8400 to work on my sansa :-) but i doubt im able to do it ... with my cut&paste approach to programming *g*
16:23:03*disorganizer is almost on the way home, so dont take him too serious
16:24:26 Join webguest41 [0] (n=4ea571a5@gateway/web/cgi-irc/
16:24:31 Join Galois [0] (
16:25:42 Quit webguest41 (Client Quit)
16:27:45 Quit Nico_P (Read error: 113 (No route to host))
16:29:10 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
16:33:10petur2798 points ;)
16:33:20pondlife64-bit fun?
16:33:22peturwell 2790 really
16:33:52amiconnpondlife: Looks like you didn't test-compile for any target...
16:34:10pondlifeI did tests for H300 and Archos Player
16:34:36pondlifeBut forgot a file
16:36:40gevaertsAt least the bootloaders still compile
16:36:48pondlifehehe, not too many option menus in there
16:37:12Infaustoanyone knows the Erlang's programming language?
16:37:47krazykitInfausto, does that have anything to do with rockbox?
16:38:16michael1486does anyone know if you can change the usb initiation sequence?
16:39:01scorche|shmichael1486: didnt you ask that a couple hours ago and got an answer?
16:39:27michael1486yes but i am wondering if anyone else knows about it...
16:39:45Infaustoi don`t know, but elang is a concurrenci oriented language, specially optimized for a multicore envioroment... just a think how that look in an ipod...
16:39:47scorche|shthe person you talked to would know the most about it...
16:40:19scorche|shInfausto: it would probably look the same in an ipod once compiled...
16:41:36michael1486does anyone know the best way to compile rockbox in windows (im using dev-c++)
16:41:53scorche|shthere are many wiki pages about that
16:41:56pondlifemichael1486: Cygwin, or use a Linux VM
16:41:56krazykitmichael1486, by reading the relevant wiki pages
16:42:13*scorche|sh wouldnt count cygwin as one of the "best"
16:42:33*PaulJam likes cygwin
16:42:36*pondlife too
16:43:15pondlifeIt's unintrusive
16:43:18linuxstb_michael1486: Rockbox requires gcc and a unix-like environment to compile
16:43:23pondlifeVery very slow though
16:45:44 Join webguest [0] (n=55496e7d@gateway/web/cgi-irc/
16:46:02 Quit disorganizer ("CGI:IRC")
16:47:17 Quit webguest (Client Quit)
16:47:49LambdaCalculus37michael1486: Useful information here:
16:48:43LambdaCalculus37From there, choose your path: Cygwin, Linux VM, or a Linux install.
16:48:54scorche|shor colinux
16:51:05*LambdaCalculus37 points michael1486 to the Wikipedia page he just linked right now
16:52:16LambdaCalculus37michael1486: When in doubt, consult the Book of Wikipedia. :)
16:52:32LambdaCalculus37Or the Deity known as Google.
16:53:25LambdaCalculus37Just follow the DevelopmentGuide page, and you'll have a Rockbox dev environment in no time.
16:54:57 Quit Ave (Read error: 110 (Connection timed out))
16:59:26linuxstb_michael1486: Why do you want your ipod to emulate a keyboard?
16:59:44 Join ryran [0] (n=42b13a58@gateway/web/cgi-irc/
17:00:01peturbecause it has lots of buttons? ;)
17:01:12michael1486i want to do hardware level fuzzing (among other things)....
17:01:16 Join Ave [0] (
17:01:19***Saving seen data "./dancer.seen"
17:02:27LambdaCalculus37michael1486: Define "Hardware level fuzzing".
17:03:00 Quit Ave (Client Quit)
17:03:00 Quit XavierGr ()
17:03:50 Join Ave [0] (
17:03:57michael1486fuzz usb input
17:04:24scorche|shfor what purpose though?
17:04:26*LambdaCalculus37 is still 404'ing here
17:04:26michael1486(correction) fuzz usb keyboard input
17:04:53michael1486debugging ;) :D
17:04:53scorche|sh< scorche|sh> for what purpose though?
17:05:03scorche|sh....debugging what?
17:05:47michael1486winlogon.exe :D
17:06:04*LambdaCalculus37 smacks his forehead in disbelief
17:06:27michael1486and other things like a new os im trying to make
17:07:40scorche| know how fuzzing works, right?
17:08:08LambdaCalculus37Well, explain it to us, then.
17:08:55michael1486you enter in random data and see what happens. :D
17:09:35 Part meck
17:10:29scorche|sh...and what are you expecting to happen?
17:11:11scorche|shi take it that "fuzzing" is just some cool word you heard thrown around and wanted to try?
17:11:40michael1486nope i actually know what to look for...
17:12:11michael1486like buffer overflows and random crashes....
17:15:39 Join Nate123 [0] (n=465c9f89@gateway/web/cgi-irc/
17:15:41michael1486im running immunity debugger to find out what happens in the memory...
17:16:30Nate123can somebody tell me where to place the rockbox stuff for my ipod mini gen 2?
17:16:48Nate123i dont know where to find the place to put it
17:17:06scorche|shNate123: the manual can tell you
17:17:11michael1486what type of stuff?
17:17:21scorche|shmichael1486: well, this is getting a bit offtopic for here...
17:17:59michael1486what nate123 or fuzzing?
17:18:21 Quit kugel ("ChatZilla 0.9.81 [Firefox]")
17:18:27LambdaCalculus37The fuzzing.
17:18:48michael1486ok ok ok.....
17:20:32Nate123im kinda confused what rockbox does, does it change the skin for your itunes or your ipod?
17:21:12michael1486nate123: get the installer
17:22:09linuxstb_Nate123: Rockbox completely replaces the software on your computer - kind of like replacing Windows with Mac OS X.
17:22:42linuxstb_I meant it replaces the software on your ipod...
17:23:56scorche|shreplaces in a functional sense that is, as the apple firmware is not touched
17:24:36LambdaCalculus37So you can always go back to it if you ever need to.
17:26:05Nate123oh alright, one more question, if i were to have the ipod mini ( i beleive thats what i have, but i thought it was ipod nano gen 2 but its colored, so ill try mini), and normally u cant put in games for it right? so rockbox would put in games? (im reading that /whyrockbox article)
17:26:10michael1486nate123: get the installer (i dont know the link)
17:27:01linuxstb_Nate123: Firstly, you need to confirm what kind of ipod it is - see here
17:28:52michael1486he said mini 2
17:29:06krazykitmichael1486, he said he wasn't sure.
17:29:24LambdaCalculus37michael1486: Some people have confused the nano 2nd Gen with a mini.
17:29:42linuxstb_Nate123: When did you buy your ipod, and was it brand new when you bought it?
17:29:50 Quit ryran ("CGI:IRC (EOF)")
17:33:58 Join kugel [0] (i=kugel@unaffiliated/kugel)
17:34:13 Join kushal_12_27_200 [0] (n=kushal@
17:34:36kugelLlorean: I don't stick to rejected patches (answering to
17:36:07 Join Rincewind [0] (
17:36:13 Quit Rincewind (Client Quit)
17:36:45Nate123linux, i bought it...i wanna say in 2006, it was new as in i bought it new, but i dont think it was too new at the time, i beleive its a ipod nano (when i remember buying) but then again i dont know if it were a ipod mini cuz it looks more like that
17:36:46linuxstb_kugel: I don't think that comment was directed at you specifically - just custom builds in general. So things like the custom-line and scrollmargins patches get widely used by theme authors, even though it's known they will never be officially included.
17:36:50 Join Rincewind [0] (
17:37:13linuxstb_Nate123: What colour is it, and what capacity (GB) ?
17:37:20LambdaCalculus37Nate123: Is your iPod very thin as well?
17:37:49kugellinuxstb_: Rejected means to me, that the task got closed with the reason rejected. So, even you told "this patch is most likely gonna rejected soon" but the task is still open, it's not rejected for me
17:37:50pondlifeI wish custom builds wouldn't extend the WPS.
17:38:05kugelpondlife: hehe
17:38:09pondlifei.e so all WPS would work with SVN
17:38:19linuxstb_kugel: The tasks weren't closed to help custom builds...
17:38:26 Join ryran [0] (n=42b13a58@gateway/web/cgi-irc/
17:38:34Nate123ummm, its decently thin, i would say less then the FIRST gen ipod nano, and its a curved edge on the side not a straight flat one
17:38:42linuxstb_Which is probably our mistake then - we should reject patches earlier...
17:38:44pondlifeAlso the artists could concentrating on making WPSes that are good for all.
17:38:48scorche|shNate123: so it is a color screen?
17:39:17michael1486look here to identify it...
17:39:19kugellinuxstb_: Yea, I actually think you should. Rejecting patches should happen as soon as it's clean they are not gonna go into svn
17:39:21Nico_Pwhat are the current incompatible WPS patches that custom builds use?
17:39:28Nate123the ipod itself is blue, the screen has color to it (the album stuff shows up in color and games etc that are preloaded on it)
17:39:32LambdaCalculus37Nate123: You're out of luck. 2nd Gen nano = not supported.
17:40:30pondlifeNico_P: I'm not sure of the current status, but multifont comes to mind.
17:40:34michael1486you might be able to use
17:40:52 Join ladnaa [0] (
17:41:00LambdaCalculus37michael1486: Okay, enough already.
17:41:04krazykitmichael1486, please keep in mind that this is an on-topic channel. ipodwizard is not on-topic.
17:41:12linuxstb_Nico_P: Lots will still be using custom-line and scrollmargins, but those can be converted to viewports. AFAIK, it's just multifont.
17:41:23ladnaahey guys, is there a way to randomly play ALL songs in ALL folders with rockbox, like with the old system on the gigabeat F10 ?
17:41:51ryranhey all. I was wondering if one of the devs could edit the title of FS #8597 (which I created) to make it more descriptively accurate..something like "stuttering & undefined instruction errors on codec-switch w/crossfade".
17:41:57pondlifeladnaa: Sure. Go into Database > Tracks > All
17:41:59michael1486krazykit: how else could i tell him?
17:42:13krazykitmichael1486, doesn't matter, it's not on-topic.
17:42:14Nico_Pis the syntax of the multifont patch incompatible with the SVN WPS syntax?
17:42:14ladnaaok one sec lemme check
17:42:25pondlifeladnaa: Sorry, just Database > Tracks
17:42:36LambdaCalculus37michael1486: This is a channel about Rockbox. iPodwizard is not on-topic, nor is it supported here.
17:42:45pondlifeladnaa: Then play and shuffle current playlist
17:42:48linuxstb_Nico_P: I think it was tied to the custom-line patch, but I've tried to write viewports in such a way as the multifonts patch can use it, but in a compatible way.
17:43:06ladnaaDatabase -> Track
17:43:15pondlifeladnaa: Or, go into Database, highlight Track then hold SELECT and go for Playlist > Insert Shuffled
17:43:17ladnaathen shuffle current playlist alright cool
17:43:41 Quit kugel ("ChatZilla 0.9.81 [Firefox]")
17:44:51Nico_Pso the syntax differences are getting smaller, that's the important part IMO
17:45:35ladnaafor some reason, i only have 60 mp3s in my track selection
17:45:41ladnaaand i should have over 800
17:46:40 Quit petur ("*real life*")
17:47:12ladnaaah ok now i see all
17:48:32Nate123if i used the ipod nano first generation wouldnt it not matter?
17:48:47ladnaaok sweet thanks pondlife
17:48:54LambdaCalculus37Nate123: No, it won't work. The hardware inside is completely different.
17:49:45 Join kugel [0] (i=kugel@unaffiliated/kugel)
17:51:18kugelNico_P: The syntax different is very small. In the latest multifont versions, you can only define userfonts, which can be used in the V% tag
17:51:30kugelwpsfont,tunerfont etc have been dropped
17:51:40Nico_Pso the WPS are compatible?
17:52:07kugelWhat WPSs? Those using multifont to the SVN version? or the other way around
17:52:23scorche|shwhat else would we be discussing...
17:52:59kugel"compatible" by the means that the default wps screen doesn't show up?
17:53:52kugelIf the WPS has %V|..|..|..|..|2|..|..| (2 being userfont1), and it's used on a SVN version, the font will fall back to FONT_UI
17:54:09kugelso, the wps is still shown, of course the font isn't the same
17:55:03kugellinuxstb_ coded the %V tag so, that it rejects any other value than 0 or 1, and defaults to 1 in such a case
17:55:15Nico_Pthat's much better than the situation used to be with customline and scrollmargins :)
17:55:51kugelNico_P: Indeed. Even though very little wps have been converted to the new multifont syntax. Those in my build will be using the new syntax
17:57:20kugelNico_P: I can only talk for me, but I like the current situation much better. They were nice themes back in time when customline and scrollmargins actually, but all those dependencies and the bad coding in general were a pain.
17:57:35kugelPlus, those themes can be converted
17:57:37linuxstb_And with careful use of viewports, WPSs should be able to be designed to be readable with a variety of fonts.
17:58:39scorche|shthey were patches that should have been rejected..."dependencies and bad coding" should be expected...
17:58:46Nico_Pand if/when multifont is implemented correctly, the conversion process should be straitforward if not nonexistent :)
17:59:36*kugel hopes that someone does the job of implementing a more %V like syntax in FS #8799
18:00:27linuxstb_kugel: Didn't JdGordon's original patch already do that?
18:01:52kugellinuxstb_: Not that I know (see He started with a .vp that contains a similar syntax like the actual one of FS #8799
18:02:01 Part michael1486
18:02:18 Quit Nate123 ("CGI:IRC (Ping timeout)")
18:02:56kugel.vp file, rather
18:03:16linuxstb_But then a later patch (or maybe just his reverted commit) moved it to the .cfg
18:03:55kugelI didn't read the whole task, but I think the syntax was the same
18:04:36 Join phinze [0] (
18:05:01 Join lymeca [0] (i=lymeca@unaffiliated/lymeca)
18:05:04lymecaI mounted a FAT32 partition on an iPod and tried to delete a folder but it jsut sits there unresponsive nto returnign me to a new command prompt. I can write other data to the HDD partition though, but viewing this one particular folder or writing to it or deleting it just freezes bash
18:05:11lymecathe folder is important though
18:05:17lymecait's the .rockbox folder
18:05:18linuxstb_kugel: See list_w_custom.patch at the very end of the task
18:05:43lymecawhenever I try to boot up rockbox it sends me to that black screen at the bootloader
18:05:50lymecasaying rockbox.ipod was not found
18:05:57lymecarecent problem
18:06:25kugellinuxstb_: Gonna look at it later, gotta fo
18:06:28linuxstb_lymeca: Just sounds like a corrupted filesystem - have you tried checking it for errors? Or you could just reformat that partition.
18:06:40lymecalinuxstb_: how should I check it?
18:07:32linuxstb_lymeca: chkdsk on WIndows, fsck on Linux, I don't know about Mac OS X
18:07:57lymecaokay so fsck.vfat /dev/sde2 ?
18:08:03lymecashould I umount it first?
18:08:21linuxstb_Yes, you should umount it.
18:08:45 Join webguest27 [0] (n=4ba5e59c@gateway/web/cgi-irc/
18:08:49linuxstb_I don't know the exact fsck command (check the manpage)
18:09:29 Quit webguest27 (Client Quit)
18:11:01 Join disorganizer [0] (
18:11:29disorganizerdingelidang, hello@all
18:12:44Rincewindhere we go again....
18:12:53 Quit gevaerts ("going home")
18:13:48lymecaFATs differ but appear to be intact. Use which FAT ?
18:13:48lymeca1) Use first FAT
18:13:48lymeca2) Use second FAT
18:13:49DBUGSent KICK lymeca to server
18:13:50Kick(#rockbox lymeca :No flooding!) by logbot!n=bjst@gateway/web/cgi-irc/
18:13:50 Join lymeca [0] (i=lymeca@unaffiliated/lymeca)
18:14:05*disorganizer does not want to be "on-topic'ed" again ;-)
18:14:40 Quit DerDome ("Leaving.")
18:15:47linuxstb_lymeca: I don't know - toss a coin...
18:15:56Rincewindfor an online browser game I played I wrote an IRC bot named Disorganizer, it said "dingeldong" a lot and said the time randomly among other things ^^
18:16:01linuxstb_lymeca: If I was you, I would just reformat though.
18:16:03*Rincewind stays on topic for now
18:16:08disorganizer@kugel: i meant linuxstb's dirty-statusbar-hack
18:16:16*scorche|sh glares at Rincewind
18:16:56*disorganizer grins at rincewind
18:17:28*Rincewind freezes and waits for his punishment
18:18:21 Quit jas6180_ (Remote closed the connection)
18:18:42 Join jas6180_ [0] (
18:19:24 Join equant [0] (
18:20:17pondlifeHmm, I've had 2 hard lockups today (on H300) - both times with the disk icon displayed.
18:20:17 Quit ryran ("CGI:IRC (EOF)")
18:21:26 Quit jas6180_ (Remote closed the connection)
18:21:45equantanyone know how to force gnupod to remove tracks from the ipod's database for tracks that aren't actually on the ipod?
18:21:46Lloreankugel: Just as a note, your post at ABi does explicitly say you're including patches that have been formally rejected.
18:21:48 Join jas6180_ [0] (
18:22:04Lloreanequant: That would be a question to ask the gnupod guys...
18:22:26 Join bertrik [0] (
18:23:24equantLlorean, oh reeeealy?
18:23:50LambdaCalculus37equant: Yes, reeeealy.
18:23:55*LambdaCalculus37 points to the channel topic
18:24:04disorganizeranyways, i think that when mf is implemented and viewports offer nice new functionality, the custom builds will propably not be SO common any more.
18:25:09pondlifeWhen or if?
18:25:18 Quit axionix (Read error: 104 (Connection reset by peer))
18:25:49 Quit Infausto ("CGI:IRC (EOF)")
18:25:52disorganizerat the time when (if) it is implemented ;-)
18:26:00pondlifeThat's better :)
18:26:10disorganizeror from then on, or whatever you like
18:27:16*disorganizer wonders if we could implement mf with the old memory-wasting cache and just ifdef it out on targets low on memory
18:27:56pondlifeAll targets are low on memory :)
18:28:07Rincewinddisorganizer: memory wasting is always bad because the memory is needed for buffering
18:28:25pondlifePeople still want pictureflow to work during playback...
18:28:35 Quit phinze ()
18:28:39*disorganizer wonders why this channel does not recognize sarcastic comments *g*
18:29:25scorche|shyou lacked the <sarcasm> tag
18:29:28Rincewindif the buffer is reused when pictureflow is exited then I wouldn't have problem with it.
18:30:14disorganizer<s>oh really?</s>
18:30:33*disorganizer waits for the slap with the channel topic
18:30:43 Join kopf [0] (n=kopf@
18:31:04pondlifeI've not looked at the multifont code yet - does it just extend the current caching to work across multiple fonts, so a single cache is used?
18:32:07Lloreanpondlife: One cache per font, iiuc.
18:32:16pondlifeAh, that's not great.
18:32:28*Llorean leaves
18:34:00kopfi've got an iPod 4G grayscale with r16800. Two questions - How do I turn on gapless audio? (It was enabled on a previous rockbox release I had installed, but I can't find it now) and Is it possible to activate the backlight with a certain keypress (like holding Menu) only?
18:34:16 Quit ladnaa ()
18:34:26disorganizerthats the main problem of mf at the moment. no good algorythm on how to use one cache for all fonts :-) and the only reason to not implement it imho would be not having a good caching algorythm
18:34:26disorganizer@kugel: i tested the latest vp-list patch and you clear the complete screen when the list is activated. is this on purpose?
18:34:26 Quit Mathiasdm2 ("Yuuw!")
18:35:03pondlifekopf: Gapless doesn't need to be enabled, assuming your music is encoded correctly.
18:35:37LambdaCalculus37kopf: And as for your second question, yes.
18:35:39pondlifekopf: You may need to enable dircache though - under System > Disk. There's a bug causing disk access between tracks if that is not enabled, which can cause gaps.
18:35:52kopfpondlife: i think that's it.. Nice one!
18:36:15pondlifeThat's just a workaround for a tricky bug though, hopefully not needed forever.
18:36:15kopfLambdaCalculus37: whereabouts do I go for the second question? I've looked through Settings -> Backlight and all I seem to be able to configure is the duration it stays on
18:36:35LambdaCalculus37kopf: Settings > General Settings > Display > LCD Settings > First Buttonpress Enables Backlight Only
18:36:39 Quit kushal_12_27_200 ("Leaving")
18:36:57LambdaCalculus37If that's what you're looking for.
18:36:59 Join argumentD [0] (
18:38:09 Join Arathis [0] (
18:38:16kopfLambdaCalculus37: don't think so.. What I'm looking for is an emulation of the original ipod backlight system
18:38:17disorganizerkugel: and when going down in a list, the last few lines are "missing" from display
18:38:33kopfwhere you can use the ipod without the backlight turning on automatically, but if you hold Menu, the backlight comes on
18:38:49 Join Mathiasdm [0] (
18:39:20LambdaCalculus37kopf: In that case, I'm not sure.
18:39:38disorganizerpropably a line-number miscalculation from font-height
18:40:19equantIf the Apple firmware and the rockbox firmware disagree on what mp3s are on my ipod, what can I do? Apple see's all of the tracks that gnupod does, and rockbox only sees a few tracks.
18:41:00equantusing 16797 on a 5th generation ipod
18:43:52scorche|share they mp3s?
18:44:32equantYeah, I can listen to them with the Apple firmware.
18:44:42equantBut they don't exist via Rockbox.
18:45:01equantWhen I use gnupod, it can find them via gnupod_search.
18:45:08 Quit Thundercloud (Remote closed the connection)
18:45:31scorche|shhave you deleted the *.tcd files and recreated the database?
18:46:01equantNo. Not on purpose.
18:46:08equantI don't know what .tcd files are.
18:46:16 Join Thundercloud [0] (
18:46:41scorche|shthey are in your .rockbox directory...try doing that if updating didnt help
18:46:59equantok, thanks.
18:49:35 Join MarcinMikolajczy [0] (n=5316e84c@gateway/web/cgi-irc/
18:50:30equantThanks scorche, that did the trick.
18:51:05 Join gevaerts [0] (
18:53:05 Join axionix [0] (
18:53:31preglowmy h120 still hangs if i insert usb while it's booting :/
18:57:17disorganizergtg, cul8r
18:57:17 Quit disorganizer ("ChatZilla 0.9.81 [Firefox]")
18:57:25 Join miepchen^schlaf [0] (
18:57:38 Quit suom1 (Read error: 131 (Connection reset by peer))
19:01:21***Saving seen data "./dancer.seen"
19:01:26*scorche|sh sighs
19:03:14 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
19:05:13*LambdaCalculus37 hopes that he finally got his modified disktidy.c working right so he can post the new patch already
19:07:29 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
19:11:05 Join RexDart [0] (n=cd9cbcfe@gateway/web/cgi-irc/
19:11:42RexDartHmm hmm hmm
19:12:18LambdaCalculus37disktidy.c :389: warning: excess elements in array initializer
19:12:31pondlifeHow many is too many?
19:12:33LambdaCalculus37Can someone take a look at this pastebin:
19:12:39RexDartCan rockbox be prevented from auto-shutoff when connected to a power supply? It's kinda gimpy when it shuts off and comes right back on and shuts off a few minutes later and comes right back on and shuts off a few minutes later and comes right back on and shuts off a few minutes later and comes right back on. Then guess what it does.
19:13:01bertrikI'll have a look
19:13:15pondlifeLambdaCalculus37: There are 4 entries
19:13:26pondlifesystem_option[4] should work
19:13:43 Quit equant ("Leaving")
19:14:13pondlifeAlso, is "Both" correct for 3 options?
19:14:30LambdaCalculus37pondlife: I fixed it just now. It's "All". :)
19:14:34bertrik[] would also work
19:15:04LambdaCalculus37pondlife: Line 378 of disktidy.c reads int selection, ret = 2; should I change that?
19:15:11LambdaCalculus37Or is it fine?
19:15:33pondlifeSorry, I don't have the source here
19:15:41pondlifePaste it?
19:15:51LambdaCalculus37One moment; will do.
19:17:25bertrikif it was my code, I would change it
19:17:57LambdaCalculus37I was think it should be 3 instead of 2.
19:18:08pondlifeAh, that bit.
19:18:22pondlifei.e. default to ALL
19:18:31LambdaCalculus37Done too!
19:18:44*LambdaCalculus37 gives gVim the magic command to save
19:18:44pondlife rb->set_option("Files to Clean", &ret, INT, system_option, 4, NULL);
19:19:36bertrikoh I thought you were talking about coding style
19:19:55*LambdaCalculus37 fixes that last bit.
19:20:04LambdaCalculus37All right gentlemen... let's pray... :)
19:20:11*LambdaCalculus37 compiles
19:20:24*pondlife faces Sweden and gets on his knees
19:20:39*gevaerts spots a warning and edits LambdaCalculus37 to fix it
19:20:58LambdaCalculus37gevaerts: Where?
19:21:12gevaertsLambdaCalculus37: Just joking...
19:21:12pondlifeLambdaCalculus37 failed to compile?
19:21:21*pondlife complies instead
19:21:29LambdaCalculus37pondlife: Still working.
19:21:36LambdaCalculus37I'm watching it.
19:21:47*gevaerts promises to add more smileys (or make fewer jokes)
19:22:01*pondlife tried the latter but failed
19:22:07pondlifeAh, not the LATTER
19:22:50LambdaCalculus37disktidy.c:339: error: expected expression before 'else'
19:22:50 Quit MarcinMikolajczy ("CGI:IRC (EOF)")
19:24:06 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
19:24:27LambdaCalculus37Also: disktidy.c:468: error: expected declaration or statement at end of input
19:24:52bertrikcan you paste the entire source code, it's a bit hard to figure out which line is line 339
19:27:25 Quit linuxstb (Read error: 104 (Connection reset by peer))
19:28:09pondlifeYou're missing a } at line 219 ?
19:28:10bertrikyou're missing a } somewhere
19:28:33pondlifeAnd perhaps 151 too?
19:28:43pondlifeOr the indenting is a bit odd
19:29:06pondlifeLooks like you might have tabs in there, naughty...
19:29:27LambdaCalculus37pondlife: They were already in there. :)
19:29:28bertrikand what
19:29:34bertrik's up with the \ ?
19:30:01pondlifeAh, line 313 is the one
19:30:29LambdaCalculus37Line 313 is blank.
19:30:35pondlifeShould be }
19:30:52pondlifeJust like line 293
19:32:20LambdaCalculus37Line 151 is return TIDY_RETURN_USB;
19:32:41LambdaCalculus37Ditto Line 219.
19:33:21LambdaCalculus37Should there be a new line after Line 468?
19:35:05bertrikCan't the functionality be done in a much more clean way? same piece of code is not essentially duplicated three times.
19:36:02 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
19:36:14pondlifeIt could be a lot neater
19:36:40domonokyLambdaCalculus37: always place a newline at the end of a .c file, or the compiler will warn.. :-)
19:36:48 Quit barrywardell ("Konversation terminated!")
19:37:03pondlifeLambdaCalculus37: Especially put a new line at the end of a .h file... :)
19:37:20LambdaCalculus37domonoky, pondlife: Right!
19:37:39*LambdaCalculus37 takes the advice of domonoky and pondlife
19:39:46bertrikis there a rockbox style of indenting?
19:40:10*LambdaCalculus37 compiles once more
19:40:13 Join lee-qid [0] (
19:40:17pondlifebertrik: Kind of
19:40:33Beta2KI've seen a couple styles so far
19:40:34*domonoky thinks the is something about intending in the rockbox coding guidlines..
19:40:35pondlifebertrik: No tabs, and follow the style you find on a per file-basis
19:41:12 Join stripwax [0] (
19:41:16pondlifeWe have both" if () {" and separate lines for brackets (which I personally prefer)
19:41:19 Quit stripwax (Client Quit)
19:41:23bertrikI was asking because disktidy.c itself looks like it could be tidied up a bit
19:42:26LambdaCalculus37betrik: Those tabs were there when I opened the file, and that's a clean version from SVN.
19:42:32LambdaCalculus37I already eliminated the tabs.
19:42:45bertrikindeed, it's not that bad actualyl
19:43:46 Quit Mathiasdm ("Yuuw!")
19:43:59 Quit FOAD (Read error: 104 (Connection reset by peer))
19:44:10 Join FOAD [0] (
19:44:24 Join XavierGr [0] (n=xavier@rockbox/staff/XavierGr)
19:44:32LambdaCalculus37Yes! \o/
19:44:51bertrikit compiles? ship it!
19:45:02pondlifeFlyspray it!
19:45:07*LambdaCalculus37 is going to test it before making the patch
19:45:16*LambdaCalculus37 throws a few .dolphinvew files into a folder
19:45:22Beta2KHurry up and commit it before there's a chance to find bugs in it
19:45:33pondlifeTalking of coding standards, do we have a policy on use of const?
19:45:53*pondlife tends to use it everywhere possible, but knows some people who don't like that so much.
19:46:07linuxstbpondlife: I think it's a good thing because of rombox
19:46:22LambdaCalculus37ARGH... segfault on the sim!
19:46:24pondlifeI've held off doing it in Rockbox because it wasn't much used
19:46:26bertrikI like to use it for structs and lists etc., but I hate it when used in function prototypes
19:46:34pondlifebertrik: Why?
19:46:39pondlifeI like it everywhere
19:47:07 Quit OlivierBorowski (Remote closed the connection)
19:47:42LambdaCalculus37No good. It's segfaulting when I try to enter the Files to Clean menu.
19:47:44bertrikonce you change it in a prototype, the requirement for const ripples all the way up to other functions that use the prototype
19:47:58bertrikbut maybe I'm not always using it correctly
19:47:59pondlifebertrik: Which is useful, IMHO.
19:48:22pondlifeI've seen code that boils down to strcpy("HELLO", "WORLD!!");
19:48:29pondlifeconst prevents that
19:48:32 Join jas6180 [0] (
19:49:10LambdaCalculus37pondlife: Celebrated too early. =/
19:49:13*gevaerts points to ioccc for that sort of code
19:49:51gevaertsMy only submission there didn't win :(
19:50:07pondlifeMaybe usb_core.c can enter #20 ?
19:51:24gevaertsIt's not ready yet :)
19:51:33*amiconn has the greyscale library running on M3 :D
19:52:21*pondlife forcibly removes a \o/ from LambdaCalculus37 and gives it to amiconn
19:52:47bertrikconst for storage of stuff has direct impact on bin size, while const for function prototypes is only useful to prevent incorrect use but doesn't have much impact on generated code
19:52:50LambdaCalculus37Awww, my cheering man! :(
19:53:19*gevaerts hands LambdaCalculus37 a \☺/
19:53:27pondlifeLambdaCalculus37: I'm sure you'll get a new one soon.
19:53:43LambdaCalculus37So, any ideas as to what's causing the sim to segfault on disktidy?
19:53:53pondlifeLambdaCalculus37: gdb is your friend
19:54:53pondlifegevaerts: I fear your code breaks several Rockbox guidelines.
19:54:57gevaertspondlife: it needs to compile
19:55:43 Quit Thundercloud (Read error: 104 (Connection reset by peer))
19:55:57 Join Thundercloud [0] (
19:56:30*gevaerts wouldn't know what's wrong with that code ;)
19:56:53linuxstbamiconn: Nice ;) Is that the first remote greyscale?
19:57:07amiconnIt's quite demanding on the CPU though
19:57:12*LambdaCalculus37 is running rockboxui through gdb
19:57:28pondlifeLambdaCalculus37: You should see what's up when you run disktidy
19:57:36amiconnThe iaudio remote uses serial transfers and the TL0350A lcd controller can't be switched to mono mode
19:58:02amiconnThat means I have to transfer 24Kbit per frame, or ~1.8MBit/s
19:58:37 Join toffe82 [0] (
19:58:46 Join disorganizer [0] (
19:58:47amiconnI managed to squeeze ~4MBit/s out of the plain update (at 124MHz)
19:59:20amiconnFor grey it's a little lower, because the phase calculation has to happen interleaved with the bit transfers
19:59:52amiconnSo the greylib causes 54% cpu load for fullscreen
20:00:24*amiconn wants an elephantsdream_128x96.mpg
20:00:56markunlinuxstb: do you still have the original ED file?
20:01:52 Join barrywardell [0] (n=barrywar@rockbox/developer/barrywardell)
20:01:58bertrikI read another open source animation movie was coming out soonish
20:02:48 Quit jas6180_ (Connection timed out)
20:04:40linuxstbmarkun: I think so, but it was just the HD version from their website. I need to find the disk containing it and my encoding scripts...
20:05:33linuxstbamiconn: There's already one that size -
20:05:58 Join bluebrother [0] (n=dom@rockbox/staff/bluebrother)
20:06:03*amiconn would also like to see better scaled versions for the various lcds now that we know that the dimensions just need to be even
20:06:24amiconnlinuxstb: Ah, the small H10 4:3 version :) No widescreen though...
20:07:07 Join corevette [0] (
20:07:45 Join Mathiasdm [0] (
20:11:20 Quit SirFunk (No route to host)
20:15:00linuxstbAnyone have any ideas about how to handle conditional viewports?
20:15:29 Quit Mathiasdm ("Yuuw!")
20:15:53 Join Mathiasdm [0] (
20:17:32LambdaCalculus37pondlife: gdb gave me the message 0xb7dbbc23 in strlen () from /lib/tls/i686/cmov/
20:17:48pondlifeUse the bt command in gdb
20:17:59pondlifeThat will show you where in disktidy you are
20:18:45RexDartWhat oh what does 'car adapter mode' do for a 5th gen ipod?
20:19:03RexDartanything besides powering the ipod on/off according to the power supply's state?
20:20:07*domonoky thinks a segfault in strlen sounds like a unterminated string.. :-)
20:21:01amiconnOr like using strlen() instead of the correct rb->strlen()
20:21:07LambdaCalculus37domonoky: bt in gdb reports 0xb199e32b in tidy_lcd_menu () at disktidy.c:402
20:21:55LambdaCalculus37Just before that: 0xb199e366 in plugin_start (api=0x80b95c0, parameter=0x0) at disktidy.c:424
20:22:52*disorganizer just thinks of conditional viewports with a conditional based on the context! like if in wps show this viewport, if in list show that viewport... *g*
20:23:25*disorganizer also plans to demoralize the devs and conquer the world, but thats another story
20:23:41domonokyLambdaCalculus37: can you give us the output of the bt in a pastebin ? (if you can solve the problem yourself) :-)
20:24:07 Join desowin [0] (n=desowin@atheme/developer/desowin)
20:25:00LambdaCalculus37domonoky: Sure, one minute.
20:25:19 Quit kopf ("leaving")
20:26:29amiconnlinuxstb: While the M3 LCD is technically a remote LCD, it is handled as the main LCD in rockbox, and hence gets greyscale support
20:27:04domonokyLambdaCalculus37: have you changed anything in tidy_lcd_menu() till you last pastebin of the file ?
20:27:12amiconnWhile it would be possible to make a secondary greylib for true remote LCDs, there's quite some nastiness I have no idea how to solve
20:27:21LambdaCalculus37domonoky: No, nothing.
20:27:31amiconnOn H300 and X5 it wouldn't be problematic, because their main LCDs have no greylib
20:28:39amiconnBut on H1x0 and M5, there is the problem that both the main and the remote LCD could have greylib support - *but* they cannot be run at the same time
20:28:54domonokyLambdaCalculus37: the size of opt_items struct seems wrong.. its defined as system_option[3] but is inited with 4 elements ?
20:29:22amiconnThis is for at least 2 reasons: CPU load, and the availability of only one user timer
20:30:44*RexDart lets a juicy one
20:30:55*RexDart has soiled the floor
20:31:16scorche|shRexDart: this is an on-topic room..
20:32:27domonokyLambdaCalculus37: the segfault is in the rb->setoption(..) on line 402, so i suspekt gcc is droping the forth line of the opt_items struct, (because the size specified is too small), and the set_option goes beyond the struct because you tell it it has 4 entrys..
20:33:42Beta2KThis is why C needs to keep track of array dimentions...
20:33:47LambdaCalculus37domonoky: So I have to fix the system_option to [4]?
20:34:06domonokyjup, or better to [] (without number)
20:34:18LambdaCalculus37Will do.
20:34:34domonokythen gcc will take the size from the number of element you use to initialise
20:35:49linuxstbpixelma: Do you have any plans to look at any of the other WPSs in SVN? I remember that iCatcher for example doesn't always display the correct codec icon (we reached the 52-bitmap limit).
20:37:21 Quit Horscht ("We don't make mistakes, we just have happy little accidents")
20:37:23LambdaCalculus37Compiling now!
20:37:35pixelmalinuxstb: I thought that one should look at it too (in the not so distant future) but I also wanted to do other things... for example draw at least 2 new svgs, just started
20:39:06linuxstbpixelma: OK, I may look at iCatcher then...
20:39:10RexDartscorche|sh: This is an on-the-floor mess
20:39:25RexDartIs there another room where Rockbox is perhaps the topic?
20:39:32scorche|shthis is it
20:39:35LambdaCalculus37RexDart: Yes, this one.
20:39:49RexDartseems otherise
20:39:54 Quit RexDart ("CGI:IRC")
20:40:07scorche|shif it is, i would love to know about it...
20:40:39pixelmalinuxstb: but it's really worth it to clean up the wps bitmap folders
20:42:37LambdaCalculus37WOOHOO!! \o/
20:42:53*LambdaCalculus37 gives pondlife and domonoky beers
20:43:10*LambdaCalculus37 takes a beer
20:43:31disorganizerpsssst. ontopic?
20:43:39scorche|shbeer *is* ontopic
20:43:44*pondlife must do more test builds...
20:43:45domonokybeer is also ontopic for this channel :-)
20:43:50*disorganizer swipes away the beer from the topic
20:43:53LambdaCalculus37So is success in getting disktidy to finally work! :)
20:43:56*amiconn seems to have an automatic enter key :(
20:44:04*LambdaCalculus37 preps a patch for the tracker
20:44:26pondlifeThat can't be correct - PortalPlayer firmware code relies on an apps variable declaration....
20:44:55Beta2KBeer is always on topic here disorganizer
20:45:28linuxstbpondlife: ?
20:45:37pondlifeMy red...
20:45:55pondlifeAh, it's all SWCODEC
20:46:04disorganizerbut thats disgusting. who is always spiling beer on the topic? its better to drink it!
20:47:24 Quit desowin ()
20:47:28pondlifelinuxstb_: Do you think it correct that firmware code should rely on playback.c like that?
20:47:49markundisorganizer: it's better not to mention the *b* word or else you'll wake up petur and preglow..
20:48:11pondlifeaudio_is_initialized is really "playback is initialised", not "sound is initialised".
20:48:56 Join chrisjs169 [0] (n=chrisjs1@unaffiliated/chrisjs169)
20:49:03disorganizerbeer beer beer *g*
20:49:06pondlifeamiconn: Ah, now I understand your gah!!
20:49:22*amiconn can't find the reversal command :(
20:49:24pondlifeMakes my 2790 look poor
20:49:45 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
20:49:47amiconnI know it was in the irc logs, but the download link for the raw longs vanished, so I can't get them for greping
20:49:57 Join orsonj [0] (n=orson@
20:50:14 Join mirak [0] (
20:50:54amiconn*raw logs
20:52:01 Quit mirak (SendQ exceeded)
20:52:31 Join plb [0] (
20:53:21plbhm interesting d2 has 'mostly' usable menus now?
20:53:28 Quit barrywardell ()
20:53:31 Join mirak [0] (
20:53:39Nico_Pamiconn: svn merge
20:53:51 Quit axionix (Read error: 110 (Connection timed out))
20:53:54plbcan't wait to slap rockbox on my d2 ;]
20:54:03amiconnyeah, found it (going through revision history and finding the date of my first goof :\
20:56:42LambdaCalculus37pondlife: Patch for disktidy in Flyspray; see above.
20:56:59 Quit LambdaCalculus37 ("CGI:IRC 0.5.9 (2006/06/06)")
20:59:40*jhMikeS sees r16802 and wonders how about the commit message :)
20:59:46 Join SliMM [0] (n=stefan@
20:59:47 Join donutman25 [0] (n=chatzill@
21:00:45SliMMhow do i compile rbutilqt in linux?
21:00:51amiconnjhMikeS: svn commit -m "message" <accidental_enter_before_specifying_files> :(
21:01:08SliMMon ubuntu
21:01:19domonokySliMM: take a look at rbutil wikipage..
21:01:24***Saving seen data "./dancer.seen"
21:01:35domonokyif that doesnt hepl ask more specific questions :-)
21:01:44SliMMdomonoky: ok, thanks
21:01:46 Quit plb ("Lost terminal")
21:01:50 Quit Arathis (Read error: 110 (Connection timed out))
21:01:53Nico_Pamiconn: that's why you shouldn't use -m :)
21:02:14amiconnWhy not? Hoe else would I enter the message?
21:02:33jhMikeSamiconn: I'm always ready to hit ctrl-C since there's time
21:02:46Nico_Pan editor is opened to let you enter the message, and there is a list of touched files in it
21:02:49amiconnI tried. Too late...
21:02:50*pondlife puts the -m at the end of the line
21:03:09pondlifei.e. svn commit <file> <file> -m <message>
21:03:14amiconnNico_P: on cygwin?
21:03:21bertrikthat has happened to me too, the " is too close to the enter button
21:03:35Nico_Pamiconn: if there is something like nano or vi on cygwin, then I guess so
21:03:42pondlifebertrik: Depends on your keyboard layout.
21:03:51amiconnbertrik: " is far away on a german keyboard (shift-2)
21:03:59pondlifeAnd a UK one
21:04:04jhMikeSI've hit enter just to make the shell remember the command line I was typing because I had to do something else first :p
21:04:50*amiconn thinks he should at least adopt the "enter -m last" method
21:06:04amiconnNico_P: No nano or vi...
21:06:21amiconn(they might exist, but I don't have them installed)
21:06:47*amiconn prefers user friendly editors
21:07:03Nico_Pthere is evim :)
21:07:04 Part pondlife
21:07:06*gevaerts agrees. That's why he uses vim ;)
21:07:27Nico_Pbut yeah I see what you mean... I've started to learn vim only quite recently
21:07:53*scorche|sh wonders how nano isnt certainly is nicer to new folk
21:07:55amiconnI know how to use vi. It's just a really cumbersome editor
21:08:03*amiconn prefers gui editors
21:08:18*bluebrother points to gvim ;-)
21:08:24 Join desowin [0] (n=desowin@atheme/developer/desowin)
21:08:30scorche|shah...that makes sense then
21:08:51scorche|shnano always seemed like the simplest terminal editor
21:09:14bluebrotherI never understood nano ... :o
21:09:28SliMMdomonoky: what is this libusb thing?
21:09:56Horschtbluebrother, do you prefer vim?
21:10:12domonokya usb library you need, use synaptic to get libusb-dev :-)
21:10:39bluebrotherHorscht: yes. After getting used to it ;)
21:10:40SliMMoh, so -dev
21:10:44pixelmalinuxstb: in case you're working on it - iCatcher.128x96x1.wps could be renamed to represent that it's for a greyscale display (though I think it only uses monochrome graphics), same with UniCatcher
21:11:07domonokyits used for the player detection... you need the devel files, normal lib-usb is probably already installed on your pc..
21:11:13bluebrotherSliMM: the -dev (or -devel on other distros) package usually holds the development headers, i.e. the stuff you need for building
21:11:40bluebrotheryou don't need devel packages for using a library (or whatever), only if you plan building software that links against it.
21:11:50SliMMdomonoky: and after compilation, how do i install this (i.e. integrate it into the package manager)
21:12:04SliMMbluebrother: i understaind
21:12:22domonokySliMM: no need to integrate it, just run it, its just one file..
21:12:36 Quit PaulJam (".")
21:12:38SliMMwhere will that file be?
21:12:41bertrikgevaerts: do you have any plans regarding charging now that USB works?
21:12:58 Join midgey [0] (
21:13:03domonokywhere you have built it.. named rbutil or rbutilqt ?
21:13:26SliMMwhat's the usb stack doing at the moment?
21:13:40gevaertsbertrik: not really. I don't know enough about that (yet?) to have any kind of opinion on it
21:14:03BigBambilinuxstb: Is the following possible with viewports - I would like to display album art, but if it doesn't exist, position text instead using say three different viewports in it's place
21:14:19SliMMis it working for regular users?
21:14:49gevaertsSliMM: you can find status information at
21:15:03SliMMgevaerts: thank you :)
21:15:17bertrikgevaerts: I think it's not that hard, the hard part is integrating it in the existing API's I think. For example, I think we need some kind of indication from USB to the charging subsystem to tell when high-current charging is allowed and when it is not allowed.
21:15:18linuxstbBigBambi: No... I havent' worked out how to incorporate conditionals in viewports yet - the existing mechanism won't work (viewports are a level higher than all other WPS elements)
21:15:55BigBambiSo I need to have one viewport for the album art, then position the replacement text within that based on the line method?
21:16:31gevaertsbertrik: I can do that indication easily, but I'd really like to leave other power-related stuff to others for now.
21:17:25linuxstbBigBambi: For now, yes, but I would like to have some kind of conditional viewports, even if it's only for album-art
21:17:31BigBambiOK, cheers
21:18:07SliMMdomonoky: i want to move rbutilqt away from the rockbox source
21:18:07BigBambiI think album art is the important one, as I can see if not having one big image, wanting multiple other replacements
21:18:10linuxstbThe left-margin tag should be considered deprecated (once the functionality is fully replaced with viewports - i.e. conditionals)
21:18:11SliMMhow do i do that?
21:18:47RincewindLlorean: are closed entries still kept in flyspray or are they removed completely after a while?
21:18:54domonokySliMM: you mean the newly built binary ? just copy and paste it somewhere..
21:19:07SliMMjust the binary?
21:19:36bluebrotherRincewind: you can still browse closed tasks.
21:19:40domonokyrbutil is built as just one file, you can copy this were you like..
21:19:48bluebrotherjust expand the search box using the +
21:20:01kugeldisorganizer: I don't quite get what you want in FS #8799
21:20:16bluebrotherthe only external files rbutil uses are the language files. You _could_ add those to the binary as well
21:20:23Rincewindbluebrother: and closed tasks don't get deleted some time after the close?
21:20:30 Join gregzx [0] (
21:20:58SliMMuh, ugly theme that doesn't integrate into the current gtk theme
21:21:12 Quit nplus (Remote closed the connection)
21:21:18bluebrotherRincewind: not that I know of (how could we point users to searching closed tasks in such a case?)
21:21:53SliMMis there a way to make qt integrate with the current theme?
21:22:33 Quit csc` ("Powering Off")
21:23:03domonokySliMM: rbutil uses Qt which is KDE style.. so it doesnt fit your gnome style, but it should be somehow customiceable.. but that goes beyond the scope here ..
21:23:05disorganizer@kugel: i defined the list viewport to be x=14,y=14,width=145,height=145. then when i for example enter the context menu during playback, the list shows up inside the area (except the other error), but the space outside the defined area gets cleared instead of still having its old content
21:23:19bluebrotherSliMM: Qt supports theme plugins.
21:23:32kugeldisorganizer: and where do you edit it?
21:23:33bertrikhmm, my sansa e260 hangs after installing v16800 and newer, worked fine yesterday, I'll try an official build.
21:23:36bluebrotherdomonoky: there's a difference between KDE and Qt styles
21:23:37SliMMwhy qt?
21:23:49Rincewindbluebrother: thanks. Then I can still point people to my patch if they ask for it.
21:24:00SliMMand not GTK :>
21:24:16bluebrotherSliMM: why Qt in which aspect?
21:24:16kugeldisorganizer: I think your issue s, that you edit the parent viewport. I edit the child (list) viewports, so that the parent vp is still fullscreen
21:24:28domonokySliMM: because Qt is nice and really multiplatform..
21:24:52SliMMso gtk isn't?
21:25:21bluebrothergtk is a bit more a hassle on windows.
21:25:35SliMMok then
21:25:38disorganizerkugel: if i use your patch the outside-area gets cleared. when i used the "old" patch the outside-area maintained its content, where outside-area is the rest of the screen outside the list-viewport
21:25:38bluebrotherit works well, but it requires another install.
21:25:55bluebrotheralso, the old wx version of rbutil was already cpp −− gtk is c.
21:26:28bluebrotherfor your style issue, you might want to look for a Qt style that picks up your gtk settings.
21:26:47kugeldisorganizer: and whats the problem?
21:26:59SliMMrbutilqt freezes for a couple of seconds, before finishing the installation and after unpacking
21:27:26bluebrotherSliMM: there is −− I guess there is also something the other way round
21:27:45bluebrothermight this be the moment it writes the unpacked data to disc?
21:28:02disorganizerkugel: should the area outside be cleared or not? as far as i understood it should not get cleared.
21:28:20SliMMit might be, but it was annoying :p
21:28:38SliMMi don't care anymore, i use ubuntu now
21:28:42kugeldisorganizer: of course it should get cleared. That was the main problem of the old vp-list patch, that the screen was full of wps garbage
21:28:52domonokySliMM: nothing was wrong, rbutil was just working :-)
21:29:00kugeldisorganizer: Or are you keen on not-updating wps parts in the main menu?
21:29:24disorganizer@kugel: the question is: what happens if a, lets say status bar viewport, writes there too?
21:29:36SliMMare there new colours of cabbie?
21:29:52kugeldisorganizer: huh?
21:30:13domonokybluebrother: maybe we should change the mouse to the hourglass while installing ? then the freeze shouldnt this nasty.. as you know its doeing something :-)
21:30:14kugelwrite what_
21:30:42disorganizer@kugel: see logs yesterday :-) for example use the area above and below the list viewport to display data when in a list-context
21:31:31kugeldisorganizer: that's another issue. I think it should be possible to combine linuxstb's wip patch with my one
21:32:14kugelI don't see a problem, we could just add another viewport to the list, which contains technically any content
21:32:29 Join crzyboyster [0] (n=6210f0cd@gateway/web/cgi-irc/
21:32:54 Quit jas6180 (Remote closed the connection)
21:33:01*kugel haven't looked into linuxstb patch too much though
21:33:05crzyboysterCan anyone look at my wps code with viewports and tell me why it draws up the default wps and doesn't work? >
21:33:18 Quit crzyboyster (Client Quit)
21:33:24 Join jas6180 [0] (
21:33:38kugeldisorganizer: Just take a deeper look in apps/gui/bitmap/list.c to see what I've exactly done
21:33:46 Nick thegeek_ is now known as thegeek (
21:34:05 Join crzyboyster [0] (n=6210f0cd@gateway/web/cgi-irc/
21:34:39 Join helseekspressen [0] (
21:35:26disorganizer@kugel: as i said on fs its just a misunderstanding / misinterpretation :-)
21:35:47SliMMis there a way to build the database on the pc?
21:35:48linuxstbcrzyboyster: Failed parsing on line 68 : Unclosed conditional
21:35:55 Join abrenk [0] (
21:37:02SliMMi mean the [former] tagcache
21:37:47abrenkHi! I just want to say a quick thank you and "Nice work!" for your accomplishments on rockbox and especially the updated default theme.
21:37:48crzyboysterStill won't work. Here's the new code:
21:37:57*domonoky thinks you cant built tagcache on pc... this is a missing tool :-)
21:38:02 Join phinze [0] (
21:38:13crzyboysterabrenk: Just wondering, what player do you own?
21:38:34abrenkcrzyboyster, ipov video 80gb
21:38:59SliMMdomonoky: and where is tagcache in the rockbox sourcecode?
21:39:14linuxstbcrzyboyster: That WPS should load OK...
21:39:18SlasheriSliMM: please check tools/database.c
21:39:21SliMM(a script may be enough)
21:39:36abrenkcrzyboyster, tried a release a couple months ago and kept the default theme which everybody else found quite stupid. ;-)
21:39:37Slasheriunfortunately, that doesnt compile atm
21:39:39*disorganizer thinks a custom peakmeter which works vertically and where you can position the r and l channel freely would be awesome.
21:40:01 Quit mirak ("Ex-Chat")
21:40:11SlasheriSliMM: but that skeleton tool uses the core tagcache engine, so it will give the best compatibility when working
21:41:08SliMMSlasheri: i'll take a look; how should i.. compile it?
21:41:18crzyboysterlinuxstb: It still doesn't seem to load and it still displays the default theme. Will having too many images cause the wps to not work?
21:41:45domonokySlasheri: the best would be to build a lib out of this, and use it in rbutil, like we do with rbspeex :-)
21:42:11scorche|shcrzyboyster: he just told you why...
21:42:12disorganizer@ crzyboyster: use bitmap-strips
21:42:21SliMMwhat is rbspeex?
21:42:52domonokyrockbox modified speex codec for voice in rockbox..
21:45:18disorganizer@kugel: the list-length (in lines) seems to be calculated with screenheight, not vp-height. at least the list is as long as it would be without list-vp before scrolling (that was regarding 1 in the fs entry)
21:45:38crzyboysterSo having the image buffer overflow causes the wps to not work at all?
21:46:56linuxstbcrzyboyster: Yes, that's possible.
21:47:06 Quit spiorf (Remote closed the connection)
21:47:13linuxstbdisorganizer: That won't help - they will still use the same space once loaded
21:47:42disorganizer@kugel regarding 2: i compiled your patch with the sb-hack, and your patch overwrites the area the statusbar uses
21:47:48disorganizer@ linuxstb: ?
21:48:00 Join spiorf [0] (
21:48:03linuxstbdisorganizer: "use bitmap-strips"
21:48:22 Part SliMM
21:48:28disorganizer@linuxstb: i meant to reduce the bmp# in his wps
21:49:58 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
21:50:02 Quit Slasheri ("Reconnecting")
21:50:06 Join Slasheri [0] (
21:50:52Slasheridomonoky: indeed
21:51:10Lloreankugel: Seriously, why are you so against the settings simply working like *every* other setting in .cfg files in Rockbox works?
21:51:12Slasheridomonoky: i will try to remember that and take a look at some point
21:51:13disorganizer@linuxstb: interesting enough, when using kugel's patch and your hack together (so the wps tags dont get displayed any more) then selecting from the playlist wont result in a crash *eg*
21:51:17linuxstbkugel: IMO you should wait for JdGordon's work on the lists - I would expect him to add customisablity once the core functionality is working well.
21:52:02 Quit donutman25 ("ChatZilla 0.9.81 [Firefox]")
21:52:30disorganizer@linuxstb: i didnt even understand a bit of what his patch is about :-) but i can wait
21:53:41*amiconn calibrates the M3 greylib
21:54:32 Join mirak [0] (
21:56:11 Quit abrenk ("Leaving")
21:59:22disorganizer@ kugel: if you wait for jdgordon's basic list functions and have time, you could try to make #8400 work *g*
22:01:44 Quit phinze ()
22:02:34 Quit crzyboyster ("CGI:IRC (EOF)")
22:03:16 Join phinze [0] (
22:06:42 Quit Mathiasdm ("Yuuw!")
22:06:55 Join Mathiasdm [0] (
22:10:57 Join shotofadds [0] (
22:15:32 Quit bertrik (Read error: 104 (Connection reset by peer))
22:17:23 Join simonrvn_ [0] (i=simon@unaffiliated/simonrvn)
22:18:13jhMikeSamiconn: does this look like a reasonable alternative to the overly general set_irq_level? Saves 80 bytes OndioFM, 488 for H120, 736 for e200
22:18:21 Quit simonrvn (Read error: 104 (Connection reset by peer))
22:18:26 Nick simonrvn_ is now known as simonrvn (i=simon@unaffiliated/simonrvn)
22:21:08 Quit phinze ()
22:23:56 Part helseekspressen
22:25:52 Quit jgarvey (Read error: 110 (Connection timed out))
22:26:24 Join jgarvey [0] (
22:27:45 Quit mf0102 ("Verlassend")
22:30:07kugeldisorganizer: The viewport length (in lines) is done in apps/gui/list.c
22:30:37 Quit simonrvn (Read error: 104 (Connection reset by peer))
22:30:43 Join simonrvn [0] (i=simon@unaffiliated/simonrvn)
22:30:52disorganizer@kugel: could you reproduce 1) ?
22:31:07kugeldisorganizer: I'm not on my linux back atm, can't compile
22:31:41disorganizerok. theme uses nimbus-12. maybe its enough to set the font manually to provoke the error
22:33:05kugeldisorganizer: I don't have a build with the latest version
22:33:55 Quit lee-qid (Read error: 110 (Connection timed out))
22:33:59kugellinuxstb_: IMO I can start with it when I want, and JdGordon can get into the task when he wants to
22:34:05 Quit midgey ()
22:34:34Bagdermarkun: how far are you down the meizu road in regards to trying something on the target?
22:35:44 Join simonrvn_ [0] (i=simon@unaffiliated/simonrvn)
22:36:07 Quit simonrvn ("Reconnecting")
22:36:13 Nick simonrvn_ is now known as simonrvn (i=simon@unaffiliated/simonrvn)
22:36:31disorganizer@kugel: did you see jdg's patch on fs? he is working on the viewport framework atm, so you may need to do a lot of resyncs soon.
22:36:36kugelLlorean: As I said in the tracker, I can't think of a solution without editting all existing themes. Please do not think that I want the current "solution" desperately
22:36:52kugeldisorganizer: probably
22:36:59Lloreankugel: And as I said, editing all the existing themes would take less than a few minutes and a one line shell script.
22:37:24LloreanWhether you want to be the one who does the job or not, it's not an excuse to break existing functionality.
22:38:09 Join crzyboyster [0] (n=6210f0cd@gateway/web/cgi-irc/
22:38:21kugelLlorean: I'm open to better solutions. It's work-in-progress, and just a first shot.
22:38:36 Quit Jon-Kha (Remote closed the connection)
22:38:49 Quit gevaerts ("feeling tired")
22:38:49Lloreankugel: Well, remove the hack, then.
22:39:17crzyboysterOK, now I seriously have no idea why my wps isn't working. I already converted 10-15 images into bitmap strips, but it still won't work! Here's the code:
22:40:08linuxstbcrzyboyster: If your bitmaps are too big for the image buffer, converting to bitmap strips won't help
22:40:26linuxstbcrzyboyster: Try temporarily removing some images, and see if it works then.
22:40:39 Join suom1 [0] (
22:41:08kugelcrzyboyster: You didn't convert repeat bmp's ;)
22:41:54crzyboysterI know. I only did battery and volume. So can I just cut and paste half of the images out from the wps directory and see if it works then?
22:42:18 Join Shaid [0] (
22:43:03crzyboysterIt still won't work... can someone check it for me in checkwps?
22:43:11kugelLlorean: So the best way would be to remove the hack, and edit the themes (i.e. cabbiev2) to contain listxpos: 0, listypos: 0, listwidht: LCD_WIDTH and listheight:LCD_HEIGHT?
22:43:36markunBagder: I would like to try and run some test code this week
22:43:42markunbut I don't know if I will have time for it
22:43:48Lloreankugel: Yes.
22:44:03Lloreankugel: Or find another way that doesn't involve introducing awkward special cases.
22:44:07markunI would prefer to get the bootloader to compile and then just add the test code in there
22:44:37Bagdermarkun: sounds like a fair approach yes. Do we know the memory address ranges in use?
22:44:42Lloreankugel: But you could make a shell script that simply concatenated those lines onto the end of every file with .cfg extension, then run it on a list of all the old themes
22:45:24markunBagder: yes,
22:45:44kugelLlorean: Sure it wouldn't be hard for integrated themes, I just thought of not breaking compability with existing theme, like those from the wps gallery
22:45:48Bagderah, missed that
22:46:04Lloreankugel: The existing WPS gallery is close to being trashed anyway, so don't worry about those.
22:46:11kugelLlorean: I just got an idea.
22:46:22LloreanWith the new themes site, we're going to be a lot stricter on broken themes as it is, so a very, very large portion of them will need fixes anyway
22:46:25pixelmacrzyboyster: in the %?pv you're referencing an image with a "p", I don't see an %xlp. Not sure if this is all
22:46:42markuntoday kgb2008 spotted a mistake in the datasheet, I'm very happy that he didn't stop reading the disassemblies now that we have so many docs.
22:47:10Bagder0x08000000 sounds like an odd SDRAM start
22:47:21kugelLlorean: How is this: I planned to implement "List dimensons" settings anyway, so that the user doesn't have to edit the .cfg allways (and to avoid hidden cfg settings). There could be an option to reset the dimensions
22:47:42markunBagder: I can check the datasheet again, but this is the starting address of the part of the firmware which gets loaded into SDRAM
22:47:44Lloreankugel: If you can set them manually, why do you also need a reset?
22:48:00LloreanAs in, if there's settings for each dimension, you just set the setting to its max or min respectively
22:48:05kugelFor doing 4 jobs in one
22:48:49kugelDoes that sound good to you?
22:48:53LloreanNo sense in wasting binary on that, though.
22:49:09markunBagder: I wonder what the battery life with rockox will be since we will not be using the Calmrisc initially
22:49:41Bagdermarkun: I don't think it'll be that bad as I doubt they even use that for audio
22:50:01Lloreankugel: Remember, one goal is to not waste space. Resetting them ought to be so rare that it's not worth adding another setting just for resetting those four features. If you REALLY want to add a reset, add a "resetlist.cfg" to the normal configs folder on your player. =P
22:50:01Bagderunless perhaps samsung provides codecs that does use it
22:50:14markunit wouldn't surprise me
22:50:38markunsince they also use CPU for their own mp3 players
22:50:51Bagderah right
22:51:04kugelthe bin delta would be approx. 1 byte
22:51:14Bagderbut still the dsp makes little point for audio but a lot of point with video
22:51:22markunI found a site about calmrisc optimisations for tremor, but of course I forgot to bookmark it
22:51:38markunwe'll see if we can/have to use it
22:51:50Bagderreally? what compiler did they use?
22:51:56Lloreankugel: Little wastes here and there add up. And I don't think you're basing that approximation on any real data, anyway. Just don't do it, okay?
22:52:39markunBagder: it's not very clear to me (and the page doesn't load right now)
22:52:46markunah, there it is
22:52:54 Join bertrik [0] (n=3e2d0fb6@gateway/web/cgi-irc/
22:53:08kugelLlorean: Settings should done in plugins imo anyway
22:53:50Lloreankugel: There's some good reasons for not doing it. But whether they should or shouldn't is irrelevant.
22:54:00LloreanYou can add it in later if they get moved to plugins. Until then, it's wasteful.
22:55:10BagderI built the dfu-util from openmoku for the fun of it anyway
22:56:44kugelLlorean: I will do something about it. The resetlist.cfg sounds actually nice
22:56:58Lloreankugel: I didn't mean "create one by default"
22:57:05LloreanI meant "if you have bad themes, you can use one"
22:57:18LloreanThemes accepted to the new theme gallery shouldn't be bad like that, and if they are, we'll take 'em down.
22:57:44LloreanAssuming this feature gets completed before the gallery does, etc.
22:58:04markunBagder: yes, me too
22:58:23kugelWhat's the reason against creating it by default?
22:58:37markunI hope that gevaerts can figure out the changes to the protocol so we can use it for our firmware updates
22:59:51Lloreankugel: We don't need more clutter, and it's absolutely useless for people without broken themes.
23:00:25LloreanAnd even with broken themes, someone can quickly and easily create their own one for future use.
23:00:37Bagdermarkun: is it known to be changes? the wiki page says dfu-utils haven't been "confirmed" to work?
23:00:37kugelThey are not broken until the patch is committed imo
23:01:02markunwell, they can 'see' the player
23:01:04kugelFor now, it's just a patch. And I don't think that themes, which do not take the patch into account are broken
23:01:05markunbut that's it
23:01:17Lloreankugel: They will be broken once the patch is committed though, as you said.
23:01:27***Saving seen data "./dancer.seen"
23:01:35Lloreankugel: Meanwhile, the patch tracker is specifically for "working on patches with the intent to be committed." Are you saying that's not what you're doing?
23:01:57Llorean"it's just a patch" should mean "so I'm going to make the changes that will make it most suitable for inclusion in an official build"
23:02:04kugelI'd gladly break compability upon committing. But until then, I have my doubts
23:02:36LloreanSo, you want to make a patch that will be different from the committed version, intentionally?
23:02:42kugelLlorean: The patch isn't in a committable state, with or without the hack
23:03:06Lloreankugel: But it should work TOWARD a committable state. That means removing things that aren't suitable for SVN as they crop up.
23:03:20LloreanNot "keeping them in so it's easier on my unsupported build"
23:03:28kugelLlorean: Sure, but I'd rather break compability at last
23:03:35LloreanBreak it FIRST
23:03:43LloreanSo people start writing themes that will be compatible with it when it goes in SVN
23:03:58LloreanI know you don't want to change all the themes in your build, but frankly, the patch tracker isn't there so you can work on features for your build.
23:04:50LloreanYou should strive for the first iteration of the patch to feature the syntax that you intend to be the final syntax, if at all possible, so that people testing in or unsupported builds eager to use it will be getting closer to compatible with SVN when it's committed, not becoming schismed further
23:04:51 Quit jgarvey ("Leaving")
23:05:01kugelLlorean: the themes in my build aren't the problem. As you said converting them is fastly done
23:05:08LloreanThen make the change.
23:06:19LloreanThere's no reason for waiting for the last minute to break compatibility if you know it's going to anyway. People testing the patch for SVN inclusion should be testing it specifically with viewport-list enabled themes anyway.
23:07:55linuxstbkugel: Please correct me if I'm wrong, but you seem to be doing two different things - 1) Working on a patch you hope will get included in SVN; 2) Making your own build. IMO things like your theme reset belong in your custom build, but not in the patch you're promoting for inclusion.
23:09:15 Join phinze [0] (
23:09:37crzyboysterlinuxstb: Can you take a look at this and tell me if there's still something wrong with the code:
23:10:32linuxstbcrzyboyster: It validates OK (using the checkwps tool)
23:11:45 Quit domonoky (Read error: 104 (Connection reset by peer))
23:11:55crzyboysterI have no clue what I'm doing wrong...
23:12:26crzyboysterIt works with the same amount/size of images in the older version without viewports
23:13:04disorganizertry commenting out everything and then activate each viewport after the other.
23:13:42linuxstbcrzyboyster: Can you upload the entire wps (with bmps) somewhere?
23:14:20crzyboysterYes. Just wait a minute.
23:14:49 Quit phinze ()
23:16:48 Quit bertrik ("CGI:IRC")
23:18:09linuxstbcrzyboyster: Which device is it for?
23:18:55crzyboysteriPod Nano
23:19:56BigBambiI've finally mastered viewports in a WPS :)
23:20:29crzyboysterI'm still getting there!
23:21:02*disorganizer wonders whats so complicated about viewports.
23:21:38crzyboysterWell, it's a lot different from the old line based wps coding
23:21:55BigBambiFor me, it was just not missing one argument from each of 8 or so viewports, then getting confused between the different ways of positioning things
23:22:19disorganizeryes, but if you read the examples and look into other wps's with vp its easy to understand it
23:22:39*amiconn found an obscure bug in greylib vertical unbuffered scrolling
23:23:12BigBambiFor instance viewports are x|y|width|height and progress bar is height|leftpos|rightpos|toppos so calculating one relative to the other, whilst not difficult, leaves room for confusion
23:23:14 Quit Mathiasdm ("Yuuw!")
23:23:17crzyboysterIt still takes a while though. It is a very nice feature I have to admit..
23:23:19*disorganizer needs to admit the change from screen relative to viewport relative positioning between patch and commit came with nice surprises. and of course the suddenly strict checking of parameters
23:23:25BigBambidisorganizer: I didn't say it was hard
23:23:29LloreanBigBambi: I've never liked the order of parameters in teh progress bar.
23:23:39LloreanThey were basically tacked on as they were added, to avoid breaking compatibility.
23:23:48BigBambidisorganizer: Just it has taken me ages to work out all the little typos and bugs
23:24:16BigBambiLlorean: I think it just gets confusing that everything with position parameters uses different parameters and different orders
23:24:18*Llorean wonders if he could get people to agree on standardizing that token, then running a quick fix on any "old" WPSes we decide to carry over.
23:24:19disorganizerLlorean: so hopefully this will be gone with viewportified progressbar (hopefully also vertical *g*)
23:24:47LloreanBigBambi: I think they should always be x,y,x2,y2 or x,y,width,height. One or the other, but all of them use the same one.
23:25:03linuxstbdisorganizer: What viewportified progressbar?
23:25:10disorganizerpreferrably x,y,width,height,font,bgcolor,fgcolor
23:25:32BigBambiLlorean: I agree. Trying to work out e.g. progress bar positioning when it is relative to the current viewport, whilst very doable, is much more confusing than it could/should be
23:25:33*disorganizer just likes pinging people by feature requests
23:25:39Lloreandisorganizer: Font, bgcolor, fgcolor are somewhat useless for things like progress bars, peakmeters, and images. (x2,y2 are useless for images, mind you)
23:26:34disorganizerso it would be x,y[,width,length][,font],[bgcolor,fgcolor]
23:26:53linuxstbcrzyboyster: Your last %V definition is wrong - the Nano's LCD is only 132 pixels high...
23:27:02disorganizerah. -,[+[,
23:27:06Lloreandisorganizer: As I said, those are irrelevant. I'm just discussing standardizing the positioning elements.
23:27:21LloreanRight now some onscreen elements use really awkward positioning.
23:27:50Lloreanor have the positioning in strange order, because for example, they were originally based on line-based positioning and compatibility was valued.
23:28:29disorganizerand of course everything should be vp-relative. but are there other elements with non-x-y positioning left except the pb?
23:28:32crzyboysterlinuxstb: It seems ok to me. It only goes 100 pixels if I looked properly...
23:29:03linuxstbcrzyboyster: Sorry, I meant the one before last...
23:30:02crzyboysterWow. It was screwing up because of 1 pixel....
23:30:29Lloreandisorganizer: Not as far as I know.
23:30:31linuxstbcrzyboyster: But your images are not displaying - you're still not using positions relative to the viewport
23:31:45crzyboysterI can fix that up soon. I never got a chance to tweak it like that because the whole thing wasn't showing up before. And thanks for being so helpful linuxstb
23:32:03 Quit Nevtus ("Gone")
23:32:05disorganizerLlorean: so this change should be easy enough, shouldnt it? another question would be: do we still need left-margin?
23:32:32Lloreandisorganizer: Yes. There's a hard limit of 15 viewports. It's a lot, but you can always lessen the number you need by using left margin instead in some places.
23:33:18LloreanThe problem with the change to %pb is that it would break all existing viewports.
23:33:22LloreanErr, WPSes
23:33:23disorganizerwouldnt we then also need right margin and vertical centering of text? if i see my wps, i could really need vertical centering of text in a viewport. not to mention a right margin
23:33:32Lloreanif it were done it should be committed in time to coincide with the new gallery.
23:33:44disorganizerLlorean: i could quote your own comment from the irc log :-)
23:33:55LloreanThis is a different case entirely
23:34:10LloreanI propose scripting a conversion for all new WPSes for the gallery if it's to be done like this.
23:34:22LloreanSo we specifically don't want people using the new syntax, or that script will break their new-syntax WPSes
23:34:31linuxstbLlorean: IMO we don't want to encourage the use of the left-margin. We should aim to get rid of it (viewports should be able to do the same)
23:34:45amiconnCentering is possible afaik
23:34:49disorganizerjust kidding ;-) anyways it should propably be committed togethe with other changes (statusbar? unlikely. list-viewports? multifont?) so people only need to convert their wps'es once
23:34:50Lloreanlinuxstb: Ah, I thought it was going to stick around for the simpler cases (one or two intended lines, etc)
23:35:07Lloreandisorganizer: *most* changes won't break backwards compatiblity
23:35:13disorganizeramiconn: only horizontally, not verticall ;-)
23:35:27 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
23:35:27Lloreandisorganizer: Kugel's proposal was to create a new standard, for the explicit purpose of breaking it.
23:35:40JdGordondid anyone look at 8800?
23:35:50disorganizeryes, me :-)
23:36:01disorganizerbut i dont count, do i?
23:36:01linuxstbJdGordon: I looked at it, but all I could see was a load of NULLs added to the list init function...
23:36:03amiconndisorganizer: Vertical? Why would you ever need that? Lines may vary in width, but certainly not in height...
23:36:38JdGordonlinuxstb: menu.c is the interesting bit
23:37:49disorganizeramiconn: using a viewport to draw a blue box with text vertically centered :-)
23:38:30amiconndisorganizer: You know how high the viewport is, and you know how high a line is. Place the line appropriately so that it's centered...
23:38:49disorganizeramiconn: also, you may want to have a text centered inside a viewport without knowing the font (for example if you like to allow the user to change the font)
23:39:58 Quit crzyboyster ("CGI:IRC (EOF)")
23:40:07Lloreandisorganizer: Unfortunately, a lot of Rockbox depends on line based drawing, which basically means the screen (or viewport) needs to be treated as a series of lines, iiuc.
23:40:10disorganizeramiconn: im talking about text, which i cant position x/y inside a viewport. thats what viewports are for, which brings us back to: why use leftmargin? *g* but i do see a sense in it now :-) nevertheless rightmargin would also make sense for scrolling lines
23:40:13 Join barrywardell [0] (n=barrywar@rockbox/developer/barrywardell)
23:40:17markunThe S5L8700 runs in big-endian mode. Do you think we should switch to little-endian?
23:40:29markunthat was for Bagder
23:40:54Bagdermarkun: yes I do, unless some peripherals use big-endian or something
23:40:55disorganizerwell, in fact we do have viewports for all those cases, so they should be used. though 15 viewports is tight if we also use them for statusbar and other screens.
23:41:16*disorganizer 's wps already uses 11-13
23:41:38*Llorean thinks if he designed a WPS, including statusbar for the menu, it would use 6 or 7 viewports.
23:41:48linuxstbViewport don't take up that much RAM (now...), so I don't think it's a problem to increase the number if needed.
23:41:52LloreanAnd that's assuming I decided to integrate AA support.
23:42:07 Join |AhIoRoS| [0] (n=ahioros@
23:42:40 Join Nevtus [0] (n=Nevtus@unaffiliated/nevtus)
23:43:23linuxstb52 bytes for the viewport struct itself, plus 8 bytes (two ints) to indicate the first/last lines of the WPS that belong to that viewport - so 60 bytes per viewport.
23:43:24disorganizerLlorean: depends on the target and how small the display area is. big screen = easy to fit information in it without using many fonts and much positioning
23:43:54linuxstb(less on mono targets)
23:44:07*disorganizer doesnt want to do his own build just for increasing vp count
23:44:22markunBagder: btw, this is what dfu-util gives me: Found DFU: [0x0419:0x0145] devnum=0, cfg=0, intf=0, alt=0, name="UNDEFINED"
23:44:34Lloreandisorganizer: I said for me. I have targets ranging from a 112x64 mono screen, to a 240x320 16-bit screen, and this WPS would require that many (or less on the mono target)
23:45:08Lloreanlinuxstb: We definitely should add 2-4 more for the status bar, at least.
23:45:17BigBambiI've used 6 (7 including the main viewport) for the gigabeat, and if I could have used conditional viewports for the no AA case, it woul have been 9 (10)
23:45:45disorganizerit also depends on what you want to display, though. nevertheless i think 20 would be a good raised value for sb-viewporting.
23:46:15disorganizerBigBambi: non-overlapping viewports?
23:46:27BigBambidisorganizer: eh?
23:46:41LloreanI think 2-4 added for statusbar (I think we should assume users will have info both above and below the list, but whether they want two viewports for each, or just one, is the question)
23:46:41disorganizerdo your viewports "overlap" each other?
23:46:50LloreanAnd then another 2 added when "conditional viewports" happens.
23:46:51BigBambiNo, they can't
23:47:13LloreanBecause, I suspect that means that everyone will be adding 1 or 2 viewports for display when album art isn't used.
23:47:21BigBambiYeah, I wanted three
23:47:32BigBambiBut no go atm
23:47:35LloreanYeah, but you have some to spare anyway. :-P
23:47:47BigBambiYeah, is is plenty for me
23:47:58disorganizerguess how many vp's this wps needs:
23:48:21LloreanWay too many because of multifont use
23:48:23disorganizernote that 1/3 is a conditional changing with a scrolling playlist-name
23:48:23*linuxstb used 12 for the jClix theme
23:48:30 Join matsl [0] (n=matsl@
23:48:45disorganizerLlorean: only font 0 and 1 is used. pure svn wps
23:49:23 Join jas6180_ [0] (
23:49:25disorganizerand of course bmp-stripes @ linuxstb
23:49:25Lloreandisorganizer: yeah, but I still call that multifont
23:49:37Lloreandisorganizer: I think I could do that one in 4.
23:49:43disorganizerbut we need to calculate wps-needs with that ;-)
23:49:52disorganizerproove it
23:49:53LloreanOr one visually *almost* identical.
23:50:05disorganizer;-) no, fonts need to be used that way :-P
23:50:34Lloreandisorganizer: The volume, battery, playlist position, track time, Next, and album art can all belong to the "main" viewport, no?
23:50:52LloreanMeaning that you only need individual viewports for RTC, and the three lines squeezed between the blue bars.
23:51:08Bagdermarkun: the dfu-util page mentioned a problem on big-endian, could they possibly mean with big-endian targets?
23:51:15LloreanYou might have to adjust where the blue bars are slightly to accept the reality of line based positioning, but it'd be a tiny cosmetic difference.
23:51:20linuxstbLlorean: There shouldn't really be a "main" viewport - that can easily cause problems by overwriting the other viewports.
23:51:35disorganizerplaylist position must be its own vp because it changes with playlist name which scrolls and only takes the middle part. this also prevents the time displays from using the main viewport
23:51:37Lloreanlinuxstb: I thought the WPS itself was the "main" viewport
23:51:48LloreanAs in, "anything not described within a %V"
23:51:51 Join Chaotic [0] (n=0cb2250b@gateway/web/cgi-irc/
23:52:13markunBagder: no, that's about the server. I asked in the openmoko channel
23:52:21disorganizernext could be moved from viewport text to background image though, i must admit. just wanted it in text for easy changing :-)
23:52:40linuxstbLlorean: Yes, but only for backwards compatibility. I wouldn't recommend it's use for a viewport-ified theme, But it may work - it just depends on what you put there.
23:52:50Lloreandisorganizer: Well, all I said was that I could create one visually like the one you showed me. You didn't show me extra features.
23:52:59ChaoticSorry for a really newbie question, but I can't find the answer in the forums nor docs. My 4G iPod Color (photo) is essentially rebooting and not charging
23:52:59Lloreanlinuxstb: Non-scrolling text.
23:53:12Lloreanlinuxstb: And an album art image that does not share any lines with any text.
23:53:19ChaoticI have it hooked up to the apple ipod charger that came with the ipod
23:53:28disorganizerLlorean: i was just teasing ;-)
23:53:33Lloreanlinuxstb: And the only viewports used don't share lines with any non-viewport lines. :-P
23:53:36LloreanBut it was just a hypothesis.
23:53:39linuxstbLlorean: IMO, it's simpler and cleaner to put everything in their own viewports - you know where you are then.
23:53:58Lloreanlinuxstb: Probably very true. I was just trying to make a case for the possibility to be *very* frugal with them, if necessary.
23:54:07BigBambiChaotic: I'm not sure I quite understand the question/situation
23:54:09LloreanI'd say, use all 15 if you need 'em.
23:54:14ChaoticI'm running r16796-080325. Anyone know what's up? I played songs for about 5 hours, and then the battery basically died
23:54:41Chaoticnow, when I hold Play/Pause to shut down, it says that the battery is low, shuts down, then immediately reboots
23:54:43disorganizer@ Llorean : but i must admit i learned that you CAN use the main viewport :-)
23:55:16Chaoticit's back in the main screen, and says it's charging, but it's been that way for a couple hours and hasn't charged any it seems.
23:55:28BigBambiChaotic: Using a wall charger?
23:55:37ChaoticYes, using a wall charger
23:56:02 Join DerDome [0] (
23:56:32ChaoticI had car adapter mode turned on, so I disabled that and the problem is still happening
23:56:47Lloreandisorganizer: Also, using a backdrop image for static text is a time honoured means of faking an extra font. :-P
23:56:52BigBambiChaotic: Can you shut down without the charger?
23:57:12ChaoticNo, I can't shut down without the charger. It has the same looping problem
23:57:21BigBambiThat is odd
23:57:31Chaoticbah, have to go to a meeting. I will be back later to beg for more help =)
23:57:32disorganizer@ Llorean : i still like the idea of having a customizable vertical peakmeter :-P each channel in its own viewport
23:57:38BigBambiChaotic: OK :)
23:57:40 Quit Chaotic ("CGI:IRC")
23:58:35 Join waldo [0] (
23:58:52disorganizerLlorean: i used this in simple-aa wps to change between a vertical "NEXT" and "NOW" :-) because i couldnt write a text vertically *g*

Previous day | Next day