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

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

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

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

#rockbox log for 2009-02-16

00:00:28saratogalinuxstb: I see what you mean about other flac decoders being slow, take a look at the Clip battery life for FLAC vs. MP3 in the retail firmware:
00:00:47saratogathey must be clocking that thing at 100+ MHz for flac playback
00:01:07pixelmaon the wiki page there is a note that it says it is on OSX. What OS are you running ferric84?
00:01:25 Quit perrikwp (" ajax IRC Client")
00:01:34ferric84i'm on ubuntu 8.10
00:01:45pixelmahmm, don't know then
00:04:10ferric84yeah. who knows
00:04:44linuxstbsaratoga: I can't see any recent flac benchmarks on the SansaRuntime wiki page, but there is one from Jan 2008 (with patches) that gives > 20 hrs. So I'm guessing mp3/flac are similar in Rockbox now?
00:05:19saratogalinuxstb: I think flac is actually a lot faster then MP3
00:06:02kugelFlynDice: I found something interesting. Have a look how GPIOB differs depending on the background color
00:06:20saratogalinuxstb: wiki says 13MHz for flac playback
00:07:15saratogathe best part is thats an ARMv5E CPU with 512KB of IRAM in the second bench, so they've got single cycle MAC and memory access
00:07:32 Quit Llorean ("Leaving.")
00:07:51 Join flx_ [0] (
00:08:04 Part piotrek2234
00:08:08 Join pixelma_ [50] (n=pixelma@rockbox/staff/pixelma)
00:08:08 Quit pixelma (Nick collision from services.)
00:08:11 Quit flux (Read error: 104 (Connection reset by peer))
00:08:23 Nick pixelma_ is now known as pixelma (n=pixelma@rockbox/staff/pixelma)
00:08:30kugellinuxstb: I've posted results to the CodecPerformanceComparision page some month ago
00:08:33 Quit itcheg (" ajax IRC Client")
00:08:35 Quit mipfi ("Get MacIrssi -")
00:08:39kugelthat was after the mp3-on-cop commit
00:09:03 Quit amiconn (Nick collision from services.)
00:09:04 Join amiconn_ [50] (n=jens@rockbox/developer/amiconn)
00:09:10ferric84how do you exit an application?
00:09:13saratogaspeaking of which, I should update those with WMA results
00:09:15ferric84like metronome for example..
00:09:24 Nick amiconn_ is now known as amiconn (n=jens@rockbox/developer/amiconn)
00:09:29linuxstbkugel: Yes, but how does that translate into battery life differences?
00:09:56kugellinuxstb: well, the initial question was whether flac and mp3 are comparable now
00:10:00linuxstbferric84: Mostly MENU+SELECT. The Rockbox manual should describe it.
00:10:07saratogalinuxstb: on flash targets, it probably won't since we clock the CPU at 30MHz
00:10:11ferric84gotcha. thanks
00:10:13linuxstbkugel: Yes, _runtime comparable_
00:10:23saratogaat best you'll get a tiny savings from sleeping the core
00:10:28saratogabut thats extremely small
00:10:44 Join Llorean [0] (
00:11:01kugellinuxstb: well, flac perfoms very faster, but both don't boost, so they are comparable
00:11:16kugelyou can get above 20h with both
00:11:52saratogai think we're somewhere around 25 hours now for a brand new battery based on the current draw measurements
00:12:28saratogaonce all the current improvements go in, and the core clock is lowered, and wps updates on screen off are disabled, we'll probably be just short of 30 hours
00:12:34kugelthat's a wild assumption :)
00:12:55saratogawell thats assuming the OF actually gets 20 too
00:13:02kugelI'd say it's nearer to 25h than to 20h, but I doubt it's above 23.xh
00:13:14saratogalets say on a battery that gives the OF 20 hours, I think we'd be around 25
00:13:52kugelthe OF isn't so bad
00:13:54saratogai'll do another round of measurements once all the work in progress power savings stuff goes through
00:14:05kugelI don't think we can get 30h
00:15:34 Quit matsl (Remote closed the connection)
00:15:35 Join moos [0] (i=Mustapha@rockbox/staff/moos)
00:15:42saratoga30 hours means 25mA draw, Toni claimed he got 28mA at 24MHz core clock
00:16:00 Join rakslice [0] (
00:16:00saratogaso we'd have to find another 3 mA somewhere else
00:16:27saratogathe playback DMA stuff gives us 300uA, so we're 10% there
00:16:47saratogadisabling updates when the screen is off is probably another 1-1.5mA
00:17:02kugelit's a bit less than 1h
00:17:38 Quit ferric84 ("Leaving")
00:17:53saratogathats about 1.5mA then
00:18:42saratogai guess someone using FLAC with the unboosted clock at 16MHz would save another 2.5mA, so they'd probably break 30 hours, but that'd be a trick with MP3
00:19:30saratogathough maybe we'll get lucky and someone will show up with another hard ware device we don't properly power down like Toni did this fall
00:21:22kugelsaratoga: UI will be a pain though @16MHz ;)
00:21:58kugelI'm still unsure why the zero boost wait is rotting. The UI wasn't that bad, and it would allow shrinking the pcmbuf a bit
00:22:53saratogakugel: from what i understand, it could be commited with the UI left at 30MHz
00:23:16saratogabut if we want to lower the clock to 24, it'd still need more boost calls added to the actual UI code so that it doesn't lag
00:23:30saratogaI've actually been meaning to ask Zagor about that
00:23:51saratogai'm unsure if he still has other plans before its committed
00:24:05kugelwe could just boost lcd_update for the beginning, until some proper ui boost is worked out
00:24:09kugelthat would help alot already
00:24:15amiconnDidn't someone test zero-boost wait and found that it uses a bit more battery than classic boosting?
00:24:21 Quit flydutch ("/* empty */")
00:24:42amiconnAlso, zerowait boost isn't possible on coldfire, so we'd have to maintain two different methods
00:25:42stripwaxamiconn - seems to help on pp if these results (at end) are to be believed
00:26:21saratogaamiconn: the benches are exactly the same, which is as expected since the average clock is also unchanged aside from the tiny speed up from not taking so long to boost
00:26:38saratogatheres only a speedup once the core clock is lowered
00:26:46kugelit *is* possible
00:26:52kugeland we concluded that a dozen times now
00:28:40kugel"22:28 without this patch, 23:21 with this patch." −− looks noticeable to me
00:29:01***Saving seen data "./dancer.seen"
00:29:16*amiconn is very sceptical about this zero-wait boost
00:30:00saratogakugel: thats from adjusting the core clock though, you can get that right now just by lowering it without zero wait boosting
00:30:04amiconnSure lowering the unboosted clock will increase runtime, but at the price of laggy ui. And if you do ui boost, it can quite easily reduce runtime below what we started
00:30:28saratogaamiconn: the idea of zero wait boost is to avoid that problem
00:30:40saratogayou simply boost in the required UI code, and then unboost when done
00:30:50kugelamiconn: I don't think so
00:30:52amiconnUI boost would not only need to kick in when the user is actively doing something, but also e.g. for scrolling
00:31:19saratogaamiconn: of course, but it doesn't have to stay boosted, even when scrolling the GUI only uses a fraction of the CPU's time
00:31:41saratogaso scrolling would probably still result in a boost ratio only somewhat higher then normal
00:31:55kugelonce the user is doing UI stuff, the power consumption is higher anyway (backlight, lcd updates, additiona calculations), the boost won't be that noticeable in the end
00:32:10amiconnI know. I still think it's not worth the hassle, and will make things worse
00:32:11kugelbut the lowered clock will give better runtime when you don't use the player
00:32:40pixelmayes, runtime is best when I don't use my player ;P
00:32:41saratogaamiconn: I don't understand your reluctance here
00:33:06saratogai assume improved battery life is worth any increase in complexity so long as someone is willing to work on it
00:33:08kugelpixelma: particulary with lowered clock (e.g. 22MHz instead of 45MHz), which is the point
00:33:19kugelat least for high-effecient codesc
00:33:45saratogaand frankly, the increase here some quite small
00:34:02saratogawe'll need to add a couple calls to boost, and fix up the PP clocking, which probably should be done regardless
00:34:19saratogaon Coldfire, I suppose it depends on if we can really get away zero wait or not
00:34:39amiconnsaratoga: The improved runtime part is what I don't believe until proven. And I mean not as an experiment, but with proper gui speed
00:35:01 Quit bmbl ("Woah!")
00:35:03saratogaamiconn: I don't see why GUI speed should have any impact on battery life
00:35:03amiconnWe can not get zero-wait boost on coldfire without risking freezes. Period.
00:35:16amiconnsaratoga: I mentioned it. Scrolling.
00:35:24saratogawe're literally adding a couple mA of current load while the backlight uses 100mA
00:35:29saratogaits literally irrelevent
00:35:51saratogahmm i used literally twice
00:36:07amiconnWhat does the backlight have to do with boost & scrolling?? And if it's irrelevant compared to backlight, you can stop anyway...
00:36:09 Nick JdGordon|zzz is now known as JdGordon (n=jonno@rockbox/developer/JdGordon)
00:36:23JdGordonamiconn: why cant a track change event based follow playlist work?
00:36:41JdGordonbareing in mind i just woke up
00:37:03amiconnJdGordon: "Follow Playlist" is supposed to put you on the currently playing track whenever you leave the wps. So it's a status based operation, not a transition based one
00:37:30 Quit stripwax (Read error: 104 (Connection reset by peer))
00:38:07kugeluhm, but the directory changes only at track change
00:38:32kugelso, the patch only needs to be updated during track change
00:38:49amiconnTry the follwoing in current rockbox. Start playing a track ('Folow Playlist' enabled of course). Navigate to the browser, move the cursor to a different track. Go back to the wps, and the again go to the browser. The track must not have changed yet. Note at which track you end up.... the one you browsed last.
00:39:07amiconnYou *should* end up on the currently playing track though....
00:39:47amiconnkugel: Yeah, but why copy the path on track change to yet another buffer? The wps knows the path anyway...
00:40:01 Join iamben_ [0] (
00:40:16saratogaamiconn: the only way a properly implemented zero wait boost system can decrease battery life is if it increases average clock [we know this from measurements in fs9800]
00:40:20saratogathe only way the average clock can increase is if the system decides that a higher clock is needed for smooth UI updates
00:40:32amiconnIt's easier to just set the browser path whenever you leave the wps (if follow playlist is enabled)
00:40:42kugel"is supposed to put you on the currently playing track " −− I thought just to the folder where the track is (which I'd like more too)
00:41:00saratogafrom my point of view, if the system is properly designed and decides that a higher clock is needed for what the user wants, then I think giving up some battery life is the right thing to do rather then lagging the UI
00:41:06amiconnsaratoga: Of course - and I think that it most probably will
00:41:10kugelbut your example breaks that too, I think
00:41:18saratogahowever I think in practice it will probably be impossible to do that
00:41:27saratogascrolling uses very little extra CPU time
00:41:27pixelmaI believe it's been this way (putting you into the directory you browsed last unless track changed in the meantime) for quite a while...
00:41:38amiconnkugel: What's worst is that all this used to work nicely not too long ago :\
00:41:58saratogaits on the order of 10MHz in my testing, which is comparable to what is saved by lowering the CPU clock
00:41:59JdGordonworkking, and working nicely are not the same
00:42:18saratogarealistically you'd have to scroll for hours and hours to cancel out the savings from the lowered clock when you weren't scrolling
00:42:29saratogathe battery would die first from the back light
00:42:39kugelsaratoga: I think he means scrollin lines, but I'm not sure
00:42:46 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
00:42:48amiconnWhat does the backlight have to do with scrolling?
00:42:58amiconnYes I mean scrolling lines of course
00:43:06saratogaaside from ipod users, most people use both at the same time
00:43:12saratogaand even then, most ipod users use both
00:43:30raksliceperhaps scrolling turns the backlight on?
00:43:30pixelmaIpod users?
00:43:43saratogaindeed it does, at least by default
00:44:11kugelscrolling lines don't turn the backlight on
00:44:16pixelmascrolling lines in e.g. the WPS don't turn the backlight on, otherwise it would almost constantly be on
00:44:24saratogai mean scrolling lines in lists
00:44:45saratogascrolling lines in WPSes isn't all that CPU intensive
00:44:48kugelYou're obviously not talking about the same scrolling here ;)
00:45:02pixelmathey don't either if you just stay in the list or menu without moving up or down...
00:45:19amiconnI mean scrolling text in the wps of course, which can (depending on display and font size) scroll for hours and hours without any backlight involved (on transflective displays)
00:45:20kugelsaratoga: it's more cpu intensive as you think
00:45:34saratogayour benchs put it at 1.5mA
00:46:04saratogasmall compared to scrolling in lists anyway
00:46:18saratogathe ipod video probably needs 5+ mA to do that
00:47:18amiconnAnd scrolling text already looks odd at 30MHz on G5 iirc. If you want to lower unboosted clock (which would be the only reason to introduce zero-wait boost), you need gui boost, including for whenever the scroll thread kicks in. And then I doubt that any savings remain, unless your wps never scrolls
00:48:07 Quit domonoky (Read error: 104 (Connection reset by peer))
00:48:28amiconnAlso, you then have to maintain two strategies, zero-wait boost and classic boost. It *may* become necessary to have different strategies anyway though (thinking i.MX31 here, with its dynamic voltage adjustment)
00:49:01JdGordonamiconn: your recipe 5min ago works fine on h300... the problem is not in that diff... its in hwcodec doing something differently
00:49:19*JdGordon disagrees that using the track change event for follow playlist is bad
00:49:26 Join SoapWork [0] (n=42c07542@gateway/web/cgi-irc/
00:49:49kugelis it better?
00:51:16amiconnJdGordon: The wps already knows the current track's path? So why copy it somewhere else, using an event? Sounds unnecessarily complex to me
00:52:00JdGordonamiconn: either way it gets copied, at the end of the playlist the id3 pointer is invalid so the wps loses the position
00:53:14JdGordonthe change is when its saved, previously it wasted the MAX_PATH in the wps struct which needed the cleanup and so moved the buffer to where its actually used (i.e root menu)
00:53:16amiconnI don't understand....
00:53:49kugelhow's the id3 struct of the current track invalid?
00:54:02JdGordonat the end of the playlist
00:54:08amiconnWithout events, the wps would copy the path to the browser when leaving the wps. With events, the event will cause copying the path to an extra buffer. When leaving the wps, it needs to be copied again - to the browser
00:54:39kugelJdGordon: how?
00:54:40JdGordonthat second copy doesnt happen
00:54:54 Quit SoapWork ("CGI:IRC (Ping timeout)")
00:55:06JdGordonkugel: at playlist_stop() (or maybe audio_stop()) the current track info is \0'ed
00:55:15amiconnHmm? So how is the scenario I described earlier supposed to work?
00:55:31amiconnIiuc, you mean that tthe track change event copies the path directly to the browser?
00:55:48amiconnSo the screnario I described *cannot* work
00:56:06JdGordonwell, not entirely... its copied to the root menu which checks if it should use it based on which screen was the previous one
00:56:09amiconn...because if you go to the browser and browse somewhere else, the current path will be changed
00:56:35amiconnI don't go to the root menu - I use ON to go directly from WPS to browsxer
00:56:46amiconnAnd then the scenario I described breaks here
00:56:49kugelJdGordon: why is it \0'd? is there any advantage to not keep the id3 pointer after stopping playback?
00:56:56amiconn(would be SELECT on irivers)
00:57:32JdGordonkugel: I dont know and it caught me out elsewhere... it just does
00:57:41*amiconn tests on iriver
00:57:48 Quit ender` (" We're not lost. We're locationally challenged. -- John M. Ford")
00:58:18kugelI'd just stop the \0'ing, it sounds unnecessary to me anyway. the event stuff sounds more complex than needed to me too
00:58:41 Quit tessarakt ("Client exiting")
00:58:44amiconnHmm, it does indeed work on iriver. So why does it?
00:58:53JdGordonamiconn: just tried again... the only time it might not work is if you press navi before the event gets fired (so while its buffering very quickly after pressing on)
00:59:17JdGordonbecause there is a difference in the way swcodec and hwcodec handles end of playlist...
00:59:24amiconnThe first-time "follow playlist" *does* work on hwcodec, but neither does it work at end-of-playlist nor when trying the described scenario
00:59:51amiconnThe scenario has nothing to do with end-of-playlist
01:00:03 Part iamben_
01:00:08JdGordonno, im surprised that doesnt work on hwocdec
01:01:45amiconnHmm, why does it work now?????
01:02:00amiconnI tried a few hours ago, over and over. It never worked....
01:02:33amiconnMeh, I hate bugs that change behaviour when trying to hunt them down :\
01:03:48amiconnReally. I tried dozens of builds (past-r19747), and none of them worked in the described scenario
01:04:22JdGordonif you go wps->browser->menu->browser then what you said should happen (i.e look broken) but if you always go wps->broswer then it shold work fine
01:04:58JdGordonend of playlist *is* definetly broken though yeah?
01:05:00amiconnBrowser-menu-browser shouldn't put you on the currently playing track. Just leaving the wps should
01:05:18amiconnIt even causes a null-pointer access in r19747 and later
01:06:50JdGordona quick fix for that could be just check id3 (root_menu.c:82) is not null?
01:07:36amiconnThat wouldn't fix the problem though - only the null pointer access
01:09:22JdGordonI think it actually will...
01:09:26kugelI'm still not convinced of the event thing, I agree that the wps should handle that
01:09:51JdGordonthe previous track path will stay untouched so you'll end up at the last track of the laylist
01:09:57kugelsince it's only supposed to happen upon leaving the wps anyway, so it's more a wps feature
01:11:27amiconnJdGordon: If it does, the underlying problem is probably that hwcodec also sends a track chnage event at the end of playlist, while swcodec doesn't
01:11:34amiconnNow which is the correct behaviour?
01:11:43JdGordonswcodec is correct in that case
01:11:55JdGordontrack change is for when the new track actually starts playing
01:12:35amiconnNow I only need to understand why the mpeg thread doe sthat
01:15:03kugelwhen the event won't be sent, you can safe the check for NULL, not?
01:17:00amiconnNot sending the event would be correct iiuc
01:21:39 Quit BHSPitLappy (Remote closed the connection)
01:26:11 Quit dfkt ("-= SysReset 2.53=- Ph'nglui mglw'nafh Cthulhu R'lyeh wgah'nagl fhtagn.")
01:29:48amiconnJdGordon: r20014 :)
01:30:10JdGordonnice work :)
01:30:21*JdGordon heard the disks churning and wondered who commited something
01:31:03JdGordonhopefully that doesnt break anyting :D
01:31:10 Join SoapWork [0] (n=42c07542@gateway/web/cgi-irc/
01:32:01 Join topbloke [0] (
01:32:26 Quit SoapWork (Client Quit)
01:33:24kugelJdGordon: I think the actual bug is line 843
01:33:46JdGordonof which file? there are quite a few in the repo :p
01:34:27kugelqueue_post(&mpeg_queue, MPEG_TRACK_CHANGE, 0);playing = false;
01:34:52kugelthat's on stopping, this post should be removed, not check for playing in the queue
01:35:31kugelamiconn: queue_post(&mpeg_queue, MPEG_TRACK_CHANGE, 0); will cause track_change() to be called
01:35:42kugelI noticed the difference...
01:36:13kugelamiconn: and track_change is sending the event
01:36:30JdGordonits probably also doing important stuff like cleanups
01:36:55amiconnYes, but not only. It's also doing some other important things
01:38:23*kugel wonders if stop works
01:39:26amiconnThat said, mpeg.c is kind of a mess. The question is whether it should be cleaned up, or hwcodec should be made to use playback.c, or yet a third solution (e.g. write a new, lightweight playback engine based on finding from both mpeg.c and playback.c)
01:39:35 Join nuonguy [0] (
01:40:09amiconnBut first of all, mpeg.c needs to be split into the recording and playback parts
01:41:24kugelsounds wrong to me that track_change() is called on playlist _end_, but yea, it's a mess ;)
01:43:14amiconntrack_change() *is* what detects end-of-playlist (by calling update_playlist()) - transfer_end() merely indicates that it is out of mpeg data due to a track end (with potentially more tracks to follow)
01:44:25 Join cool_walking_ [0] (i=cb3b81c3@gateway/web/ajax/
01:46:36 Join jordoex [0] (
01:49:29kugelamiconn: why is num_tracks_in_memory() even returnin > 0?
01:49:50amiconnBecause there is a track in memory?
01:50:04kugelthe one that just ended, you mean?
01:50:19kugelhow's that one interesting when the playlist ends?
01:50:56amiconnremove_current_tag() is what removes the current one from the list, so num_tracks_in_memory() yields 1 before
01:51:15kugelah ok
01:52:06kugelshouldn't it remove the track before?
01:52:39amiconnWhat I don't understand is why the test is even there. It should always be true, otherwise something's seriously broken, and playback would never stop
01:53:01kugelyep, that's what I'm thinking too
01:53:38JdGordonamiconn: I thought ou were against major work in mpeg.c because it has been pretty stable?
01:54:00amiconnIt is pretty stable in general, but it's also quite messy
01:54:30amiconnI always said that I would like to see the playback engines unified, but then playback.c used to be far less stable than mpeg.c
01:55:04*amiconn btw thinks that playback.c isn't really clear either, although it became a lot better already
01:55:58amiconnIt's also quite long - almost as long as mpeg.c even though the latter contains both playback and recording
01:56:51JdGordonpart of that is the codec stuff which isnt in mpeg.c isnt it?
01:58:26JdGordonnot completly relevant, but we havnt decided what so of api playlist<->playback should use... once thats set there is a chance playback can be cleaned up more
01:58:38 Quit Thundercloud (Remote closed the connection)
01:59:09amiconnIt might be a good idea to split the codec handling stuff from playback.c
01:59:11 Quit topbloke ("bye")
01:59:56 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
02:01:47JdGordoncould the MAS be thought of as just a hardware codec? so if things were usnified we could use the same api for it and the sw codecs?
02:03:01kugelwhy is the SIM #ifndef'd so hard in mpeg.c?
02:04:50JdGordonbecause its not closyly simulated like swcodec
02:04:57kugelit doesn't even have the thread
02:05:07amiconnSame api will be very difficult, as it really has different constraints (e.g. no overlapping), but the api should become similar
02:05:14 Join shadearg [0] (
02:05:18JdGordonactually... it is simulated alot, whereas swcodec runs the same code
02:10:50 Quit shadearg ("goodbye.")
02:11:16amiconnBtw, playback.c and mpeg.c are not the only playback engines in rockbox. There are at least 3 more: one in mpegplayer, one in video.rock, and another (simple) one in wavplay
02:11:55 Join shadearg [0] (
02:11:59amiconnThese are a lot simpler as they don't have to handle track changes though
02:18:18 Quit kugel ("ChatZilla 0.9.84 [Firefox 3.1b2/20081201080242]")
02:20:21JdGordonsure, but is there any point trying to merge them in also?
02:27:30amiconnNo, but they can be used for comparison
02:29:05***Saving seen data "./dancer.seen"
02:57:11 Quit mirak (Remote closed the connection)
03:07:26 Quit Nico_P (Remote closed the connection)
03:17:45 Join SoapWork [0] (n=42c07542@gateway/web/cgi-irc/
03:19:35 Quit SoapWork (Client Quit)
03:27:34 Nick fxb is now known as fxb__ (
03:35:06 Quit moos ("Rockbox rules the DAP world")
04:07:05 Join blkhawk- [0] (
04:13:07 Join AndyIL [0] (i=AndyI@
04:23:12 Quit blkhawk (Read error: 110 (Connection timed out))
04:24:03 Nick blkhawk- is now known as blkhawk (
04:24:40 Join SoapWork [0] (n=42c07542@gateway/web/cgi-irc/
04:26:47 Quit AndyI (Read error: 110 (Connection timed out))
04:27:15 Quit SoapWork (Client Quit)
04:29:08***Saving seen data "./dancer.seen"
04:39:20 Quit miepchen^schlaf (Connection timed out)
04:44:27 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
04:44:36*JdGordon so loves bugs which dont effect the sim :(
04:45:09 Join Barahir_ [0] (
04:45:44 Quit Barahir (Read error: 60 (Operation timed out))
04:58:49 Join likemindead [0] (
05:02:23 Part likemindead ("I'm out--like a boner in sweat pants... 0_o")
05:09:16 Quit Llorean ("Leaving.")
05:12:58 Join Llorean [0] (
05:15:40 Quit daurnimator (Read error: 60 (Operation timed out))
05:15:54 Join daurnimator [0] (
05:28:21 Part Aurix_Lexico
05:28:21 Join `Assassin [0] (
05:37:28 Join webguest09 [0] (n=ad167099@gateway/web/cgi-irc/
05:38:57 Quit webguest09 (Client Quit)
05:39:10 Join webguest49 [0] (n=ad167099@gateway/web/cgi-irc/
05:40:10 Quit webguest49 (Client Quit)
05:40:33 Quit Horscht ("Verlassend")
05:46:27 Join SoapWork [0] (
05:53:41 Join Acky [0] (
05:53:47 Quit rocko ("Leaving")
05:56:10 Join rocko [0] (
06:01:01 Quit rocko (Client Quit)
06:04:56 Quit CaptainKewl (Read error: 60 (Operation timed out))
06:08:09 Quit Acksaw (Connection timed out)
06:17:59Unhelpfulhere's my bufalloc-like allocator so far, with some very simplistic tests. i'll try to tackle compaction in the next few days, it should be pretty simple, just walk the list of blocks from first_free_block up, accumulating the value of each free block, shift used blocks down by that ammount, and update their handle table entries.
06:20:59 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
06:24:54 Join Tuplanolla [0] (n=jani@unaffiliated/tuplanolla)
06:25:31 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
06:27:25 Join _Auron_ [0] (
06:28:40 Quit perrikwp (" ajax IRC Client")
06:29:11***Saving seen data "./dancer.seen"
06:33:14 Join RoC_MasterMind [0] (
06:33:32 Quit RoC_MasterMind (Client Quit)
06:55:19JdGordonoh bugger.... I've got a bug which freezes up my DAP's and with a bit of adding and removing nothing but some spalsh()'s its gone... :p
06:57:13 Quit GodEater_ (" ajax IRC Client")
07:01:05 Quit gevaerts (Remote closed the connection)
07:10:14 Quit saratoga ("CGI:IRC (EOF)")
07:14:36 Quit daurnimator (Read error: 60 (Operation timed out))
07:14:49 Join daurnimator [0] (
07:22:11 Join rocko [0] (
07:22:21 Join einhirn [0] (
07:27:13 Quit _Auron_ ("Infinity repeatedly denies rumours of plotting with zero to bring down the Universe.")
07:29:48 Join _Auron_ [0] (
07:41:50 Join blkhawk- [0] (
07:44:59 Nick flx_ is now known as flux (
07:52:48 Quit Llorean (Read error: 104 (Connection reset by peer))
07:53:30 Join Llorean [0] (
07:54:38 Quit blkhawk (Read error: 110 (Connection timed out))
07:54:48 Nick blkhawk- is now known as blkhawk (
07:59:34 Quit rocko ("Leaving")
08:09:06 Nick Beaver`away is now known as Beaver`univ (
08:09:58 Join GodEater_ [0] (i=c2cbc962@rockbox/staff/GodEater)
08:10:51 Quit GodEater_ (Client Quit)
08:11:07 Join GodEater_ [0] (i=c2cbc962@rockbox/staff/GodEater)
08:12:09 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
08:19:57 Join Buschel [0] (
08:22:40 Quit perrikwp (" ajax IRC Client")
08:22:40 Quit Buschel (Client Quit)
08:25:03 Join ender` [0] (
08:27:13 Join perrikwp [0] (i=18ac0c41@gateway/web/ajax/
08:29:13***Saving seen data "./dancer.seen"
08:32:36 Join lymeca [0] (
08:40:37 Join Rob2223 [0] (
08:43:58 Join timc [0] (n=aoeu@
08:54:20 Join kugel [0] (n=kugel@rockbox/developer/kugel)
08:54:35kugelJdGordon: data abort?
08:55:02kugel(re: I've got a bug which freezes up my DAP's and with a bit of adding and removing nothing but some spalsh()'s its gone..)
08:55:21 Quit cool_walking_ (" ajax IRC Client")
08:58:59 Quit Rob2222 (Read error: 110 (Connection timed out))
08:59:12 Quit gromit` (Read error: 110 (Connection timed out))
09:04:12 Join robin0800 [0] (
09:04:33*pixelma also guesses that it happens with the e200
09:06:08 Join miepchen^schlaf [0] (
09:06:11 Join Zagor [242] (n=bjorn@rockbox/developer/Zagor)
09:11:32 Join petur [50] (n=petur@rockbox/developer/petur)
09:24:21 Quit robin0800 (Read error: 104 (Connection reset by peer))
09:25:33 Join robin0800 [0] (
09:29:50 Quit robin0800 (Client Quit)
09:30:37 Join robin0800 [0] (
09:36:52 Quit jordoex (Read error: 110 (Connection timed out))
09:38:06 Join {-phoenix-} [0] (
09:42:46 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
09:47:49 Quit BHSPitMonkey (Remote closed the connection)
09:55:22 Join Thundercloud [0] (
09:55:24 Quit kugel ("ChatZilla 0.9.84-rdmsoft [XULRunner]")
10:02:03 Quit miepchen^schlaf ()
10:02:19 Quit kachna|lappy (Read error: 113 (No route to host))
10:05:20 Join ap0 [0] (i=51fc5204@gateway/web/ajax/
10:05:58`AssassinWhat possible reasons might there be for song files not showing up in the database but showing up in the file list fine? I have to go out of the normal way of finding these files in the database by going to the <Untagged> section. The files in question have all the ID3 info filled in, but all the database will show is a partial title track name when playing.
10:06:22`AssassinI reinitialized the whole database to see if this might fix the problem but nothing changed.
10:12:48Zagor`Assassin: try playing it, then go to "view id3 tags" in the context meny and see how the tags look
10:15:11`AssassinI don't know where to go to view all the ID3 tags without connecting my iPod to my computer. >_<
10:16:01Zagorit's in the the wps context menu. I don't remember the key for ipod, try holding select for a few seconds when in the wps.
10:16:02`AssassinI know at the very least the track title ID3 tag is messed up in almost all these files.
10:16:45Zagor`Assassin: well then it's not surprising that the database gets confused
10:16:45`AssassinBut it's of course not messed up when I right click the file to go to properties while it's connected to my computer.
10:17:31 Quit kadoban (Remote closed the connection)
10:18:12`AssassinThat makes the problem mysterious to me. It's like the database just decided it didn't want to bother with these particular songs for no reason.
10:18:28 Join msou| [0] (
10:18:55Zagornot "no reason". you said yourself the tags are not read correctly. that is the reason.
10:19:38`AssassinWell yeah. So any idea why they aren't being read correctly?
10:21:25BigBambiWhat format are the tags, and what format are the music files?
10:21:45BigBambiSome encoders put the wrong format tags on for the type of music file
10:21:59BigBambie.g. ape tags on MP3
10:22:45`AssassinI probably tagged them with iTunes. I'll check on the format...
10:23:06BigBambiiTunes doesn't always change the actual tags as well, it uses its own database
10:23:32BigBambiSo you really need to check with a different program that reads only the actual tags, and not a database
10:24:03 Join NanoBoy [0] (n=55e65f04@gateway/web/cgi-irc/
10:25:51 Quit NanoBoy (Client Quit)
10:27:31`AssassinOkay, one of the files in question was encoded with LAME3.90. It's an mp3 and it's ID3 tag version is 2.3.
10:29:17***Saving seen data "./dancer.seen"
10:35:25 Quit Thundercloud (Remote closed the connection)
10:35:44 Join pyro_maniac [0] (
10:37:15 Quit ap0 (" ajax IRC Client")
10:46:39 Join kadoban [0] (
10:47:38 Join kachna|lappy [0] (
10:54:20 Join miepchen^schlaf [0] (
10:56:56 Join kugel [0] (n=kugel@rockbox/developer/kugel)
10:58:37 Quit {-phoenix-} (Remote closed the connection)
10:59:18JdGordonkugel: if it was a data abort at leats i would know where to start debugging :/
10:59:29JdGordoni tihnk its a freeze/infinite loop more than a crash
11:01:28 Join blkhawk- [0] (
11:09:26 Quit einhirn (Read error: 54 (Connection reset by peer))
11:11:39 Quit nuonguy ("This computer has gone to sleep")
11:14:36 Join CaptainSquid [0] (
11:15:26 Quit blkhawk (Read error: 110 (Connection timed out))
11:16:26 Nick blkhawk- is now known as blkhawk (
11:17:30 Join ap0 [0] (i=51fc5204@gateway/web/ajax/
11:24:49 Quit robin0800 (Read error: 104 (Connection reset by peer))
11:29:28 Join gevaerts [0] (n=fg@rockbox/developer/gevaerts)
11:31:58 Join gromit` [0] (
11:32:01 Quit blkhawk (Remote closed the connection)
11:32:04 Quit rakslice (Read error: 104 (Connection reset by peer))
11:32:13 Quit Rob2223 (Read error: 104 (Connection reset by peer))
11:32:17 Join rakslice [0] (
11:32:17 Join Rob2222 [0] (
11:32:22 Join blkhawk [0] (
11:43:11pixelmaJdGordon: reproducable by playing a certain directory? And does it happen with backlight off so you can't watch a possible error displayed on the screen and which DAP?
11:43:19 Quit z35 ("Leaving")
11:43:47pixelmareproducible even
11:44:01JdGordoni dont know about it being only certain directories, and its with the backlight on
11:44:13*JdGordon assumed it was his patch...
11:44:26JdGordonI havnt reviously had problems with the folder im using to test
11:44:29pixelmaok, probably something else than what I suspected
11:44:41pixelma(guess kugel too)
11:45:22kugelyep ;)
11:45:30 Quit kugel (Remote closed the connection)
11:46:03JdGordonyeah, na its definetly my patch which is causing it... not your bug
11:46:23 Quit `Assassin (Read error: 54 (Connection reset by peer))
11:51:07 Join blkhawk- [0] (
11:51:23 Quit msou| ("Bbl.")
12:05:00 Quit blkhawk (Read error: 113 (No route to host))
12:05:05 Nick blkhawk- is now known as blkhawk (
12:07:27 Join domonoky [0] (n=Domonoky@rockbox/developer/domonoky)
12:14:13 Join blkhawk- [0] (
12:14:35 Join webguest74 [0] (n=1813fbae@gateway/web/cgi-irc/
12:21:34 Join kugel [0] (n=kugel@rockbox/developer/kugel)
12:24:39 Quit webguest74 ("CGI:IRC (EOF)")
12:24:48 Join ThoDin [0] (n=1813fbae@gateway/web/cgi-irc/
12:27:08 Join __lifeless [0] (n=lifeless@
12:27:42 Quit blkhawk (Read error: 110 (Connection timed out))
12:28:11 Nick blkhawk- is now known as blkhawk (
12:28:39 Quit kugel (Remote closed the connection)
12:28:54 Join moos [0] (i=Mustapha@rockbox/staff/moos)
12:29:19***Saving seen data "./dancer.seen"
12:29:38ThoDinGot a "vid pid" issue with a 4th gen color ipod, ipodpatch looking for 0x05ac 0x1203 but have 0x05ac 0x1263 in XP home. the question is would I be beter off at my debian box(not convinient) or is there a workaround for the vid-pid's in XP home sp3?
12:30:34 Quit linuxstb (Remote closed the connection)
12:30:59gevaertsThoDin: that may be a bug in ipodpatcher I guess. Can you submit a bug report?
12:32:01ThoDinprobably should, just registered with th wiki page that lists th id's so cant author yet
12:32:50ThoDini mis mt *nix (sob)
12:33:53ThoDini miss my unix (cant type, 3 am)
12:35:13domonokyipodpatcher doesnt look for vid/pid.
12:35:25domonokyit detectes the drive by the content.
12:36:52domonokyyes. rbutil might uses the vid/pids, but not ipodpatcher itself.
12:38:03ThoDinmaybe i Have over looked somethin then, but i read this url and thought different ;
12:38:04domonokyand a pid of 1263 sounds very new, ipod classic has 1261, and nano 3th gen has 1262, is it really a ipod color ?
12:38:25ThoDini'll keep reading, thanks
12:39:36domonokythis wiki page, doesnt say what method ipodpatcher uses, we collect those ids, to help rbutil detect the device.
12:39:58 Join gnuppla [0] (
12:39:58 Quit ThoDin ("CGI:IRC (EOF)")
12:40:41 Join ThoDin [0] (n=1813fbae@gateway/web/cgi-irc/
12:40:56 Nick ThoDin is now known as thodyn (n=1813fbae@gateway/web/cgi-irc/
12:41:45thodynsorry, got kicked by cheap hardware
12:42:43domonokythodyn: so what is the real issue ? does ipodpatcher detect your ipod ?
12:42:49 Quit _lifeless (Read error: 110 (Connection timed out))
12:42:55thodynyes, it's shaped like an aeroplane wing looking at either end and it's fusia in color ( i think )
12:43:08 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
12:43:30thodynno ipod detected by ipodpatcher or rbutil
12:44:02thodynitunes got it of course, although i have disabled it now
12:44:44thodyngot it in emergency mode just fine
12:44:50advcomp2019that is the 4th gen nano for what you said
12:45:00domonokymaybe take a look here to see if you really have a ipod color
12:45:22 Quit ap0 (" ajax IRC Client")
12:45:54thodynfor sure, 8 gig version
12:46:33BigBambiThere is no 8GB ipod color
12:46:45domonokyif it is a new nano, its not supported.
12:46:51BigBambiThat is some form of ipod nano, I'd guess 4th gen
12:46:52thodynidentical to the pink one
12:47:07BigBambiIt is not an ipod color, and it is not supported
12:47:24domonokyand 4th gen nano, would make sense with this usb pid. (one higher the 3th gen Nano)
12:47:48 Join robin0800 [0] (
12:48:12thodynsorry did I say it was a color, my bad, it's a 4gen
12:48:28BigBambithodyn: Have a look at
12:48:55thodyngot it on my sansa e250
12:48:59BigBambithodyn: It is a 4th gen *nano* - the nano bit is important - and as states is not supported
12:49:07thodynworks great!
12:50:06BigBambiyes, my point in saying look at was that the front page of very clearly states that the 4th gen Nano does not work
12:50:20thodynyou know I read that too, but they seem to go on as if it works, and Im a bleeding edge kinda guy
12:50:36BigBambiWhere does it suggest it works?
12:51:22thodynthats good to know, feel like a shot glass of a mind just read a beer mug of info (spilled some)
12:51:37BigBambi"It runs on a wide range of players: (...) not the 2nd/3rd/4th gen Nano" seems pretty unambiguous to me
12:51:54thodynthanks alot guys, apreciate your patience
12:52:48thodynmaybe i was confusing the issue some how, no woories i'll read for a while more\
12:52:49BigBambino problem
12:53:14BigBambiIf you can find anywhere that suggests the 4th gen nano works please tell us and we will change it
12:53:21BigBambiBut I don't think there is anywhere
12:53:34thodynu got it :)
12:55:27thodynlater (tommorow maybe) Ive got an interstingly "refurbished" third gen I'd like an opinion on. Take it easy .
12:55:34 Quit thodyn ("CGI:IRC")
12:58:34BigBambiIf he means third gen nano I'm going to smack him with the frontpage :)
12:59:11 Join dfkt [0] (i=dfkt@unaffiliated/dfkt)
13:00:11 Nick fxb__ is now known as fxb (
13:01:32 Quit gnuppla ("...dignity and pain for all your beloved...")
13:11:40 Join kugel [0] (n=kugel@rockbox/developer/kugel)
13:14:40 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
13:16:49 Quit bmbl (Read error: 110 (Connection timed out))
13:19:17 Quit kugel (Remote closed the connection)
13:27:05 Quit Seed (Read error: 110 (Connection timed out))
13:31:58 Join kugel [0] (n=kugel@
13:34:06 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
13:38:22 Join {-phoenix-} [0] (
13:39:15 Quit faemir (Connection timed out)
13:47:21 Join Seed [0] (
13:52:58 Quit kugel (Remote closed the connection)
13:53:14 Join kugel [0] (n=kugel@rockbox/developer/kugel)
13:57:14 Quit robin0800 (Read error: 104 (Connection reset by peer))
13:57:17 Join robin0800_ [0] (
14:00:06 Quit FOAD (Read error: 110 (Connection timed out))
14:01:14 Part LinusN
14:06:06 Quit kugel (Remote closed the connection)
14:06:33 Join kugel [0] (n=kugel@rockbox/developer/kugel)
14:07:07 Quit kugel (Remote closed the connection)
14:07:56 Join piotrek2234 [0] (
14:08:49 Quit robin0800_ (Remote closed the connection)
14:11:29 Quit Seed (Nick collision from services.)
14:11:34 Join Seed [0] (
14:16:21 Quit daurnimator (Read error: 60 (Operation timed out))
14:17:58 Part piotrek2234
14:18:10 Quit Seed (Nick collision from services.)
14:18:15 Join Seed [0] (
14:20:11 Join piotrek2234 [0] (
14:24:25 Join kugel [0] (n=kugel@rockbox/developer/kugel)
14:28:12 Join daurnimator [0] (
14:29:21***Saving seen data "./dancer.seen"
14:30:34 Quit Seed (Nick collision from services.)
14:30:38 Join Seedy [0] (
14:35:51 Join robin0800 [0] (
14:35:54 Join Sedgewick [0] (n=Sedgewic@
14:37:25 Quit Seedy (Read error: 104 (Connection reset by peer))
14:44:28 Join Seed [0] (
14:45:52 Nick JdGordon is now known as JdGordon|zzz (n=jonno@rockbox/developer/JdGordon)
14:46:29 Quit robin0800 (Read error: 104 (Connection reset by peer))
14:47:38 Quit kugel (Remote closed the connection)
14:47:51 Join robin0800 [0] (
14:48:49 Quit robin0800 (Read error: 104 (Connection reset by peer))
14:49:44 Join robin0800 [0] (
15:05:37 Quit bmbl ("Woah!")
15:13:25 Join xxYO [0] (n=d4f4ad7d@gateway/web/cgi-irc/
15:14:19 Quit xxYO (Client Quit)
15:14:37 Join mrkiko [0] (
15:15:37 Part piotrek2234
15:15:42 Join gregzx [0] (
15:29:01 Join Rob2223 [0] (
15:29:35 Join Acksaw [0] (
15:32:43 Quit Acky (Read error: 60 (Operation timed out))
15:47:15 Quit Rob2222 (Read error: 110 (Connection timed out))
15:47:38 Quit kachna|lappy (Read error: 110 (Connection timed out))
15:55:00 Join jaykay [0] (
15:56:22 Join CaptainKewl [0] (
15:57:48 Nick Barahir_ is now known as Barahir (
15:59:51 Join midijunkie [0] (
16:01:52 Join Zoxc [0] (
16:04:42 Join Aurix_Lexico [0] (
16:07:36 Join LambdaCalculus37 [0] (
16:09:54 Join Chris_Black [0] (n=Sedgewic@
16:25:36 Quit Sedgewick (Read error: 110 (Connection timed out))
16:26:21 Join kugel [0] (n=kugel@rockbox/developer/kugel)
16:28:59 Quit jaykay (Read error: 110 (Connection timed out))
16:29:25***Saving seen data "./dancer.seen"
16:32:35 Join gartral1 [0] (n=Gartral@
16:33:04gartral1the Ipod FAQ page has some horribly outdated information
16:34:31LambdaCalculus37gatrall: What needs to be fixed up?
16:36:36kadobanit says the Rockbox Utility is experimental...
16:36:50kadobanthere was a broken link i just fixed up a few minutes ago
16:37:38gartral1at the bottom of functionality questions, it sais that rockbox does not support any accessory...
16:38:13gartral1for Ipods, obviously
16:38:43LambdaCalculus37I'll fix that up, since we have accessory support, but not every accessory has been tested.
16:40:07gartral1yea, but i figured saying "some work" is better than the misleading "none work"
16:44:55gartral1and this isnt really a bug, as it is just plain funny, if you go into the plugins>applications menu and select lamp with long press and go too its properties, rb reports it has a play length of 8:32
16:45:49*linuxstb wonders about the answer to "My iPod doesn't show up as a USB drive in Windows Explorer, why not?"
16:46:23gartral1i was wondering about that too, but its kinda funny
16:47:40kadobanit's pretty unprofessional, and probably pretty confusing or anyone who would actually be asking that question...
16:48:12kadobani thought i remember reading a section like that somewhere else that was nice..i wonder if it's in the manual
16:49:09 Join fyrestorm [0] (
16:49:45gartral1is there way to "force" i file too be played in rockbox as a different file type?
16:49:52gartral1a file*
16:50:01 Join Nico_P [50] (n=nicolas@rockbox/developer/NicoP)
16:50:13kadobangartral1: in the context menu with 'Open With...'?
16:52:12gartral1kadoban: i said play, as in take an MP3 and play it as another type of file, say, a WMA or WAV
16:53:12 Join toffe82 [0] (
16:53:32 Quit miepchen^schlaf ()
16:54:49 Join bmbl [0] (n=Miranda@unaffiliated/bmbl)
16:57:44 Join nuonguy [0] (
16:59:09 Join webguest67 [0] (n=ad167099@gateway/web/cgi-irc/
17:00:29webguest67Hey I really want to upload a theme I made for the sansa c200 as there are not very many of them but I need write permission. Any help there?
17:01:03domonokywebguest67: what is your wikiname ?
17:02:17 Quit obo (Remote closed the connection)
17:04:07 Join CaptainKwel [0] (
17:05:16 Quit Zagor ("Client exiting")
17:09:36 Join obo [0] (
17:11:51 Quit lymeca (Connection timed out)
17:14:35 Quit webguest67 ("CGI:IRC (Ping timeout)")
17:15:03 Quit CaptainKewl (Read error: 110 (Connection timed out))
17:16:03 Join saratoga [0] (n=9803c6dd@gateway/web/cgi-irc/
17:16:05 Quit Chris_Black ("off")
17:17:06saratogadoes anyone know Aoyumi's [the vorbis developer] full name?
17:17:14saratogai want to commit his patches
17:19:22moosask him in a patch comment?
17:21:18kugelsaratoga: I can only find Aoyumi using google, seems he hides his last name
17:23:39 Join avis [0] (n=ident@pdpc/supporter/student/avis)
17:24:15aviscan anyone recommend a nice rockbox compatible player, that has a very good long battery life ? (and hopefully doesn't cost an arm and leg)
17:25:12moosavis: check this out first:
17:25:16SoapWorkwhat do you consider long life?
17:25:29avisi consider 24 hours long life
17:26:20*domonoky is playing again with the e200v2 button driver, and its really funny that a red backdrop makes the scrollwheel work :-)
17:26:37kugeldomonoky: not funny at all, imo
17:26:50SoapWorkIpod Video 60/80GB with MP3/MPC _should_ be able to acheive that if/when the BCM disable code becomes part of SVN.
17:27:03moosdomonoky: hehe :)
17:27:44rob_SoapWork: whats BCM disable code?
17:27:45SoapWorkand they are cheap on Craigslist / eBay. Else you're looking towards a more expensive option - Like an Iriver Hxxx or Iaudio X/M
17:27:59SoapWorknot just the backlight, the entire LCD controller.
17:28:10SoapWorkmassive power consumption decrease.
17:28:15rob_hmm :>
17:28:31kugeland massive is not even exaggerated
17:28:58avisooh ok
17:29:08kugelthough you can go 22+h with the sansa e200 too, and we also have found some ways to increase that
17:29:09rob_i bought a second hand ipod video the other day for 75GBP
17:29:11avisi'll look for something cheap :)
17:29:19rob_60gb model
17:29:33gartral1as of right now, the e200s have about 20-25 hours
17:29:35SoapWorkOf course there is nothing on the screen when the backlight is off, and this is different than stock behavior (for iPods) but not different than many other players with their not transflective screens.
17:29:37rob_though oddly on the BuyersGuide page theres no mention of it...
17:29:44moosavis: sansas seems to are the way to go for you
17:30:28*gartral1 loves his e200
17:30:29avisi can live with 14-17 hours
17:30:33SoapWorkyes, the 60/80GB models have more RAM and larger batteries than their 30GB counterparts, as well as dual-platter HDDs.
17:31:00rob_SoapWork: so the 60gb is effectively the same as the 80gb (rather than 30gb)?
17:31:08SoapWorkIf you can live with 14-17, then a sansa E2x0 is a very inexpensive option.
17:31:24rob_do sansa do any high-capacity models?
17:31:32SoapWorkrob_, for sake of this discussion, yes.
17:31:52rob_i mean in general, such as 30gb+
17:31:55gartral1SoapWork: i get 20+ on my sansa, what are you trying to play?
17:32:07rob_i saw a sansa the other day and it looked really nice (i like that it has external memory slot too)
17:32:25gartral1rob_: unfortuently, no, but they do have micro SD ports
17:32:32SoapWorkgartral1, I do not follow your question.
17:33:12gartral1SoapWork: i get 20 + hours playtime on my sansa, why is your batt life so much shorter?
17:34:13gartral1rob_: the highes internal storage sansa is the e280 (8 gb internal) but good luck finding a V1...
17:34:31rob_whats a V1?
17:34:44SoapWorkI don't recall saying it was - I directly responded to his (avis') mentioning of 14-17.
17:34:59aviswhat model number is the sansa e2x0 ?
17:35:08SoapWorkrob_, there are two totally different sets of hardware both having been sold as Sandisk Sansa E2x0s
17:35:12kadobanrob_: there are two versions of the e200 series, versions 1 and 2. only version 1 is currently supported
17:35:33avisso its a hit or miss thing ? no guarantees unless stated what revision ?
17:35:37gartral1the V1 sansas are older, and have different hardware than their v2 counterparts, currently, rockbox only works on the V1s
17:35:49LambdaCalculus37You can tell one from another by the firmware revision.
17:35:56SoapWorkThe first version of the hardware (V1) uses hardware very much like iPods and is fully supported by Rockbox. The second revision of the hardware (the AMS versions) is in developement.
17:36:04LambdaCalculus3701.XX.XX means it's a v1, and 03.XX.XX means it's a v2.
17:36:19gartral1avis: E2x0 is a placeholder for e250 e260 e270 e280
17:36:23avisooh thank you LambdaCalculus37 thats wonderful information
17:36:47gartral1avis: all of which are the same DAP with different sized drives
17:36:54avisooh ok.
17:37:59LambdaCalculus37There are also some e200 models that are known as "Rhapsody" models. They're largely the same, but have a slightly different original firmware, and thus a slightly different install method for Rockbox.
17:38:06LambdaCalculus37Those would have an "R" in their model name.
17:38:17LambdaCalculus37e.g. 250R, etc.
17:38:25gartral1there also somewhat rare outside of the US
17:38:42gartral1from what i hear >.>
17:38:55LambdaCalculus37Yeah, it was mostly sold in Best Buy stores to be used as part of their Rhapsody music service.
17:39:10 Quit thegeek (Read error: 104 (Connection reset by peer))
17:39:12gartral1which was bought by Real Inc
17:39:16LambdaCalculus37But I digress...
17:39:27 Join Rich^ [0] (n=rich@
17:40:29 Join thegeek [0] (
17:40:58avisthe e270 looks alright to me. if i heard you correct LambdaCalculus37, the series original firmware is what i need in that model ?
17:41:26LambdaCalculus37You need a 01.XX.XX firmware, yes.
17:41:42avisthank you :) save me alot of heartache knowing that info
17:41:58gartral1so whats the extra letter by the processor name in hardware info for?
17:42:06 Quit Richlv (No route to host)
17:42:37domonokyhm, e200v2 button driver also works fine if i just write out one red pixel before reading the dbop. maybe we can place this pixel out of the visible range ?
17:42:40gartral1avis: yea, but its still somewhat hit/miss with buying used/refurbished ones, and dont buy a new one
17:42:56aviseven with 01.xx.xx firmware gartral1 ?
17:43:28gartral1no, but you have too be holding the DAP in your hands to figure out which firmware version it is
17:43:29 Quit chrippa (Read error: 110 (Connection timed out))
17:43:30kadobanavis: if you can get the answer to that question it's always correct i believe, but often it's not a choice
17:43:43LambdaCalculus37If you buy on eBay, you can always ask the seller to check for you.
17:44:04avisthats what i'm doing
17:44:19*gartral1 got his refurbished, thankfuly it was a v1
17:44:47avisi had a cowon iaudio 7 16gb, but it stopped working on me, and i can't find my paypal receipt. there is no way to search transactions by string i dont think either :( for warranty
17:45:38 Join webguest06 [0] (n=ad167099@gateway/web/cgi-irc/
17:45:45gartral1avis: bummer, but also off topic, take that kinda stuff too #rockbox-community
17:46:46domonokyha, e200v2 button driver works fine, if i write the red pixel to LCD_WIDTH/HEIGHT+1 (out of the visible area) so thats a solution to our button problem :-)
17:46:49gartral1np, its just this is a technical channel, and we dont like the logs "cluttered" so to speak
17:47:44kugeldomonoky: I already tried that ;)
17:48:03kugeloh, not that. how did you do?
17:48:07moosdomonoky: wee \o/
17:48:35 Quit krazykit (Read error: 104 (Connection reset by peer))
17:49:05kugelI was having problems with lcd_window_x/y (in order to tell the driver where to draw), changing that messes normal usage up (at least what I tried)
17:49:20 Quit webguest06 (Client Quit)
17:49:28 Join krazykit [0] (
17:49:29gartral1so wait, e200 v2 buttons work?
17:49:32 Join webguest78 [0] (n=ad167099@gateway/web/cgi-irc/
17:50:09domonokykugel: dbop reading function: and dummy lcd_update function:
17:50:20domonokygartral1: we are working on it..
17:51:14kugeldomonoky: and that doesn't cause problems?
17:52:03 Quit webguest78 (Client Quit)
17:52:09 Join webguest95 [0] (n=ad167099@gateway/web/cgi-irc/
17:52:09 Quit webguest95 (Client Quit)
17:52:09kugelthe dbop reading is in an interrupt, and may be during an lcd update (and you're changing the addresses), that's surely problematic isn't it? It was here
17:52:14 Quit petur ("*plop*")
17:52:14kugelsince I was trying similar things
17:52:22domonokyseems to work fine
17:53:22 Join webguest64 [0] (n=ad167099@gateway/web/cgi-irc/
17:53:24 Quit webguest64 (Client Quit)
17:53:33 Join webguest36 [0] (n=ad167099@gateway/web/cgi-irc/
17:53:41kugelwell, I need to try that then :>
17:53:52domonokyi also print out the dbop values in the button ticktask, and it doesnt cause any problems.
17:54:25kugelwhich dbop values?
17:55:05kugeland where do you print?
17:55:37domonokyjust dbop_din, so i know it works. and i do it just after i called read_dbop()
17:56:50 Quit webguest36 (Client Quit)
17:56:54 Join webguest36 [0] (n=ad167099@gateway/web/cgi-irc/
17:57:01webguest36Sorry if I already said something like this but I got diconnected and can't find where I was before. I am trying to upload my own theme for the sansa c200 since there are so little but I need write permission. My rockbox name is creativesansa333. I am Quinlan Moll on twiki.
17:58:49kugeldomonoky: well, it might work, but it may not be correct
17:58:55domonokywebguest36: done, promise not to spam !
17:59:14kugelit's rather "undefined" what DBOP_DIN is without the read data valid bit set
17:59:32kugelyou should rather print the return value
17:59:34domonokykugel: ?
17:59:48kugelaccording to the datasheet
17:59:49domonokyof course i do print the return value
17:59:58*SoapWork didn't get his check yet.
18:00:03kugeland why should that cause problems??
18:00:03 Quit Rich^ (No route to host)
18:00:07domonokybut it contains the dbop_din value, as i have just read it..
18:00:15kugelI thought you were saying you print dbop_din
18:00:22domonokyas i do a lcd_update in the ticktask ?
18:00:25kugelas in DBOP_DIN
18:00:46domonokywhich is the same "problem" as doing a one pixel write before the read..
18:00:50kugelI'm still unsure how this would cause problems
18:01:38kugelyou just copied DBOP_DIN, it's not the same space. and DBOP_DIN is valid when you copy it to ret
18:02:03kugelret and DBOP_DON are not in the same space in the memory, I mean
18:02:56kugeldomonoky: btw, you could also use the debug menu..
18:03:00domonokykugel: stop it, i know how to code, and i am reading and printing the value correctly. (ie not just DBOP_DIN, but the return value i get from my function,)
18:03:07 Join flydutch [0] (
18:03:35domonokythe only "problem" could be if the ticktask interrupts a lcd_transfer and we change the update window
18:04:28kugeldomonoky: yes, and that's just fine. So why should it cause problems to print that (maybe I just got "i also print out the dbop values in the button ticktask, and it doesnt cause any problems." wrong)
18:04:58kugelyes, that's the problem I am expecting (as I already said)
18:05:02domonokykugel: its one more lcd_update in the ticktask, which also changes the update window
18:05:43kugelah ok, sorry for the missunderstanding then
18:06:02kugeland, I know, that you know how to code :)
18:09:04 Join dan_ [0] (n=dan@dynamic-oit-lifenet-a-1.Princeton.EDU)
18:10:50 Quit webguest36 ("CGI:IRC (EOF)")
18:12:30dan_Selecting a .m3u text file causes rockbox to play the music files listed there. So they're a bit like a symlink. How hard would it be to add the ability to list either (a) directories or (b) other .m3u files in an m3u file? In the case of (a) rockbox would visit the directory and (b) would start playing the m3u file.
18:13:25 Join fdinel [0] (
18:13:48linuxstbdan_: Why are you asking "how hard"? That's something you want to work on?
18:15:59dan_Potentially, unless rockbox developers advise that it would be very difficult.
18:16:34 Join chrippa [0] (
18:19:10 Part toffe82
18:19:37 Join lymeca [0] (
18:19:55 Join jaykay [0] (
18:20:30linuxstbdan_: I don't know that code at all, but if I was you I would just download the source, look at it, and ask questions here.
18:20:47 Join toffe82 [0] (
18:23:29dan_OK will do. I just wanted to check that there wasn't something obviously stupid about the idea.
18:26:52SoapWorkI think maintaining compatibility with other users of .m3u files would be quite important. Perhaps there is already an extension in the standard to do this, or at worst drop the other features into comment fields?
18:27:20 Join hillshum [0] (
18:28:45 Join wemdowemd [0] (
18:29:24wemdowemdIn Vista, when plugging in & switching on my 'boxed H10, I get "Windows needs to install drivers for
18:29:26***Saving seen data "./dancer.seen"
18:29:43wemdowemd"Windows needs to install driver software for your rockbox media player"
18:30:03wemdowemdWhere can I find these drivers?
18:30:17hillshumthe Rockbox USB stack is incomplete
18:30:25SoapWorkDismiss the message.
18:31:28hillshumIt will only charge
18:31:50hillshumUse the OF to sync
18:32:52wemdowemdBy OF you mean 'UMD mode', as in mount it like a normal HDD?
18:33:21hillshumOriginal Firmware
18:33:49wemdowemdOh I see. Well it is easier just to mount it in hard drive mode anyway
18:33:50 Join bertrik [0] (
18:35:09wemdowemdOut of interest, are there any intentions to get Rockbox to work like that, mounting like an MTP device?
18:35:20 Join FOAD [0] (
18:35:49wemdowemdHuh. Well I'll look forward to it
18:35:56 Quit wemdowemd ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
18:36:03BigBambiRockbox has a USB stack, it just isn't enabled by default because there are some remaining issues
18:36:32BigBambier sorry, not MTP
18:36:45bertrik...gone already
18:37:44hillshumhow do I undo
18:42:05 Join idshark [0] (
18:42:24bertrikis anyone currently working on the ams sansa storage >1 GB problem?
18:44:31 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
18:44:36 Quit fdinel ("Miranda IM! Smaller, Faster, Easier.")
18:45:04 Join mrkiko_ [0] (
18:45:33kugeldomonoky: doesn't work for me
18:46:16domonokykugel: if you disable interrupts while doing the lcd_updates, it works fine.
18:47:00kugelthe normal ones? Will try
18:48:00domonokylcd_update and lcd_update_rect needs to be protected, so we dont mess up the lcd-configuration.. we need a better solution for this.
18:50:08domonokymaybe we should get the ticktask out of interrupt context for ams-sansas. How is the ticktask implemented on other targets ?
18:50:27 Part pyro_maniac ("Leaving.")
18:51:14 Join jordoex [0] (
18:53:33kugeldomonoky: as interrupt..
18:55:19 Quit Llorean (Read error: 104 (Connection reset by peer))
18:55:42 Join Llorean [0] (
18:56:38 Quit mrkiko (Read error: 110 (Connection timed out))
18:56:49hillshumhow do I get rid of the crosscompiler?
18:59:42kugelhillshum: delete what it installed
19:00:01hillshumfrom where?
19:00:04kugeldomonoky: got it working now
19:00:11kugelhillshum: look in your path
19:01:44kugeldomonoky: I think this will do for now. I have no idea how to do it else
19:01:49 Join robin0800_ [0] (
19:01:49 Quit robin0800_ (Connection reset by peer)
19:01:53bluebrotherLambdaCalculus37: are you still using a mac?
19:02:40 Join Thundercloud [0] (
19:02:45kugeldomonoky: we should look into unifying the button driver now, imo
19:03:29LambdaCalculus37bluebrother: Yep.
19:03:50bluebrothernice. Then you might be able answering a few questions regarding OS X :)
19:04:00domonokykugel: jup, it also needs a massive clean-up. (many static vars are not needed)
19:04:09LambdaCalculus37bluebrother: I'll try my best to. :)
19:04:34bluebrotherhow long does it take to remount the player (after ipodpatchers unmounting)?
19:04:54LambdaCalculus37bluebrother: I haven't timed it. Want me to?
19:05:09*LambdaCalculus37 runs to get an iPod
19:05:31bluebrotherand are you able building rbutil on mac? I made up a patch to wait until the system remounts the player after running ipodpatcher / sansapatcher, but have no way to test it
19:05:54LambdaCalculus37bluebrother: I've never tried building rbutil on OS X. I can do so right now, though.
19:06:43bluebrotherwould be cool if you could try the patch then :)
19:06:54jaykayjdgordon|zzz: should i report a statusbar bug in the plugins or do you know that it sometimes doesnt get redrawn in the plugins?
19:06:58LambdaCalculus37bluebrother: FS#?
19:07:09bluebrothercurrent version is here:
19:07:34hillshumkugel, thats only some of it
19:07:48kugeldomonoky: which one you mean?
19:07:49hillshumstill more somewhere
19:07:57bluebrothercurrently the delay until the remount wait times out is 15 seconds, but that might need adjustment
19:08:01kugelI thought I cleaned up the fuze's one
19:08:30LambdaCalculus37bluebrother: I guess I have to apply with patch -p1 then.
19:08:50bluebrotheryep, I'm using git-svn quite a bit these times
19:08:52domonokykugel: int_btn and alike are not needed, could be easily done via return values.
19:10:02kugeldomonoky: maybe, but only until we found a way to use an ISR for the wheel
19:10:11 Join xSlack [0] (
19:10:57LambdaCalculus37bluebrother: Okay, first I'm going to time ipodpatcher and see how long it takes to remount the player.
19:11:10*LambdaCalculus37 takes his watch up and hits enter on the terminal
19:12:20domonokyISR for wheel is not possible, there is no interrupt support for dbop reads. but only the wheel, (and maybe hold) needs to store the old value. All other buttons could be done via return values. which would make the code much clearer and more similar to other button-drivers.
19:14:21LambdaCalculus37Still ticking...
19:14:42LambdaCalculus37bluebrother: 2 minutes 43 seconds to unmount, write the bootloader, and remount.
19:14:48LambdaCalculus37On my iPod color.
19:15:33bluebrotherurgh. That's kinda ages.
19:15:51 Quit saratoga ("CGI:IRC (Ping timeout)")
19:16:02kugeldomonoky: are you sure it's not *possible*? I still hope we just didn't find a way yet
19:16:12LambdaCalculus37However, it appears that it hasn't remounted, but instead is still attempting to.
19:16:16domonokykugel: datasheet says so.
19:16:28LambdaCalculus37No wait, just showed up in my Finder.
19:16:45kugeldomonoky: dbop has several interrupt masks, nothing of use for us though
19:17:19domonokypage 85, says no interrupt support for reads.
19:17:55 Join piotrek2234 [0] (
19:17:56kugelhm, sad
19:17:58LambdaCalculus37bluebrother: Okay, how to compile rbutil? Just run qmake?
19:18:10bluebrotherrun qmake, then make
19:18:29bluebrotherat least that's the way on windows and linux, I assume it's the same on os x
19:19:49bluebrotheryou should increase the value of m_remountTries (currently 30) a bit −− from your timing I guess 600 should be a good value.
19:20:10bluebrotherthat's in bootloaderinstallbase:201 after patching
19:20:14LambdaCalculus37"qmake: Command not found"!? What the!?
19:20:53*LambdaCalculus37 knows he installed the Qt libraries
19:21:25bluebrothereither its not in the path or it has a different name −− on my linux box it's qmake-qt4 (to distinguish to qt3)
19:26:19LambdaCalculus37I found it.
19:26:28bluebrothernice :)
19:26:51LambdaCalculus37It was installed to /usr/local/Trolltech/Qt-4.4.0/bin here.
19:26:57*LambdaCalculus37 adds that to his $PATH
19:29:04LambdaCalculus37Now we're in business. :)
19:29:52LambdaCalculus37Oops... got this message:
19:30:13bluebrotheryou can safely ignore that.
19:30:38xSlackWhat do you use to manage audio on a rockbox ipod
19:30:40*LambdaCalculus37 makes
19:30:51bluebrotherbuilding for release tries to add the translations to the resources, but those are created later. Worst case would be missing translations
19:31:26LambdaCalculus37bluebrother: Meh... since we're just testing some stuff anyway, it's not really the end of the world to not have the translations.
19:31:49bluebrotherLambdaCalculus37: indeed ... I guess you're speaking at least some english ;-)
19:32:01hillshumxSlack, ??
19:32:17 Join roman_ [0] (
19:33:38LambdaCalculus37bluebrother: :)
19:34:13 Quit kadoban (Remote closed the connection)
19:34:55*LambdaCalculus37 goes to make a pot of tea while he waits for rbutil to compile
19:35:12kugeldomonoky: that would at least explain that the wheel doesn't feel too good in the OF either
19:35:23kugelbut they probably poll more often
19:36:09hillshumxSlack, do you mean what desktop software?
19:37:29xSlackhillshum: i just plugged my rockbox ipod in and nothing poped up
19:37:31bertrikxSlack, you can simply put the music anywhere on a rockboxed player and browse to it using the "files" menu, or use the database and select songs based on tags
19:38:23xSlackI understand that, but i was wondering how i work with the files on the ipod though my desktop
19:38:27xSlackas far as adding removing
19:38:42xSlackNormally i just access it like a mounted drive right?
19:38:47hillshumAre you using iTunes?
19:39:22xSlackim not using anything right now
19:39:45 Quit bertrik ("yay new kernel")
19:39:50xSlackBut doesnt a ipod usually show up like a usb drive
19:39:56kugeldomonoky: if it's really impossible to read without polling we should find a way to poll the wheel more often, to get at least a reasonable and reliable feeling
19:40:02xSlackand work the same way, as long as it has rockbox on it
19:40:57hillshumyeah, just drag and drop or sync with what ever will work
19:41:08hillshumwhat OS?
19:41:34domonokykugel: we need to somehow determine how fast we need to poll, to not miss a wheel-change. (current tick task runs every 10ms).
19:42:07 Join bertrik [0] (
19:42:39kugelI think the timer can be programmed faster, if not, we IIRC also have a second timer (I think that's unused)
19:43:18 Join n1s [0] (n=nils@rockbox/developer/n1s)
19:44:27LambdaCalculus37bluebrother: Another error:
19:45:05bluebrotherhmm −− you need libusb and the development headers.
19:46:53 Join kadoban [0] (
19:47:17bluebrotheryou can put libusb.a in the rbutil base folder (if you built libusb statically). At least on linux you can do that ...
19:48:48LambdaCalculus37I have libusb and libusb-shlibs installed.
19:49:05Unhelpfulerm, is there a libusb-dev that you need, perhaps?
19:49:39*LambdaCalculus37 looks in fink
19:50:23*LambdaCalculus37 kicks X-Chat
19:53:36LambdaCalculus37There's a libusb.a file located in /sw/lib; I'll copy it into /rockbox/rbutil then.
19:53:58 Quit kugel (Remote closed the connection)
19:54:23 Quit n1s ()
19:54:43 Join n1s [0] (n=nils@rockbox/developer/n1s)
19:55:32LambdaCalculus37bluebrother: Should libusb.a be in rockbox/rbutil, or in rockbox/rbutil/rbutilqt?
19:55:56bluebrotherthe latter, though if the file is in a standard linker path you don't need to copy at all
19:56:10bluebrothercurrently you're missing usb.h ;-)
19:59:12 Join fml [0] (n=4fd3cad8@gateway/web/cgi-irc/
19:59:27 Join kugel [0] (n=kugel@rockbox/developer/kugel)
19:59:38xSlackHow come when i copy my music off of my ipod to my hard drive all of the audio files come up named with four capitol letters like AHJF.mp3
19:59:49LambdaCalculus37bluebrother: And there are a whole bunch of files named usb.h all around my mac.
20:00:03n1sxSlack: you have itunes to thank for that
20:00:07kadobanxSlack: that's how iTunes stores it
20:00:27kugelThis contains the lcd function and the cleaned up fuze button driver
20:00:54bluebrotherwell, you alternatively could download libusb 0.1, extract that in the rbutilqt folder, build it and then add -Llibusb-0.1.12 to LIBS in
20:01:00kugelI'll commit that later, I need to go now. I'll read the logs in case you have objections or ideas. The Home button works with that now
20:01:22xSlackwow, so theirs no way i can pull the actual mp3 of the ipod?
20:01:28xSlackand keep the name
20:01:41bluebrothermaybe that's the easier route ... or check to get the right usb.h and as easy fix maybe copy it to the rbutilqt folder if it's not in the include path
20:01:56rasherxSlack: That *is* the name the mp3 has.
20:02:04n1sxSlack: itunes renames then when they are copied *to* the ipod
20:02:12linuxstbxSlack: There are many third-party apps to copy files of an ipod and rename it something recognisable. But that's off-topic here (ask google).
20:02:19domonokykugel: looks fine.
20:02:21hillshummp3tag and other software can read the tag and change the filename
20:02:30SoapWorkxSlack, IF your music is properly tagged any one of the numerous full-featured tagging programs out there can rename and organize your mp3s based on interior tags.
20:02:45xSlackknow any?
20:02:50 Quit kugel (Client Quit)
20:03:00hillshumthis is getting offtopic
20:03:01*linuxstb thought this was off-topic...
20:03:22*xSlack sincerely appolagizes
20:04:18 Nick mrkiko_ is now known as mrkiko (
20:04:40*bluebrother spots that libusb-1.0 is out
20:04:51*Bagder welcomes committer #70, BigBambi
20:05:03*LambdaCalculus37 hands BigBambi a beer
20:05:04BigBambiMany thanks all :)
20:05:23LambdaCalculus37BigBambi: Welcome to the Committers Club! :)
20:05:27moosdomonoky: please don't let your findings rot on one of your tree, please commit after :)
20:06:00moosBienvenue BigBambi !!!
20:06:19*n1s gets the cake out
20:06:21BigBambicheers :)
20:06:21jaykaybigbambi: congratulation :)
20:07:09*moos hightlight the half? of the channel: BEER BEER BEER
20:07:18LambdaCalculus37BEER BEER BEER! :D
20:08:08*BigBambi has no beer :(
20:08:18*LambdaCalculus37 hands BigBambi a case of beer
20:09:56LambdaCalculus37BigBambi: Now time for you to make your initiation commit. :)
20:10:17BigBambiyeah, I will do shortly - just have to work out how to spell my name...
20:11:42pixelmayou don't know that and want to contribute to the manual? ;) Congrats though :)
20:11:57 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
20:12:06BigBambipixelma: hehe, cheers :)
20:13:34 Quit {-phoenix-} (Remote closed the connection)
20:13:44 Join perrikwp|class [0] (i=9821476f@gateway/web/ajax/
20:14:12*moos is waiting for the new modified trunk/docs/COMMITTERS file from BigBambi :)
20:14:25BigBambimoos: I'm getting there :)
20:16:03 Quit bmbl ("Woah!")
20:17:03 Join fdinel [0] (
20:21:41gartral1shouldnt the bubbles game have a playback context? or is there simply not enough room?
20:23:49n1sBigBambi: you should get a dev cloak and a forum badge too :)
20:24:19domonokygartral1: its probably just because no-one added it :-)
20:25:30kadobanprobably quite a few of the plugins could use playback context...
20:26:00*gartral1 wonders how hard it is to port...
20:26:24linuxstb"playback context" ?
20:26:26*domonoky would think it wouldnt be too hard, plenty of examples around.
20:26:28kadobanto port playback context? it's like 3 lines of boilerplate if they use the normal menu
20:29:28***Saving seen data "./dancer.seen"
20:29:37fmlkugel: hello. I've seen your patch. Well done! The comment in gwps.h is not quite accurate though. It should read "non-negative if set explicitely" (not just "positive")
20:32:39Unhelpfulkadoban: "normal" menu? does that mean not oldmenuapi, or just not custom-drawn menu?
20:33:20kadobanUnhelpful: i think as long as they don't use a custom-drawn menu it should be easy, but then i haven't looked extensively
20:35:14 Join rakslice_work [0] (
20:35:33 Join pyro_maniac [0] (
20:35:39 Quit dan_ ("Leaving")
20:41:42LambdaCalculus37bluebrother: I'm still at a loss as to which usb.h file I need to use to get rbutilqt to build properly.
20:41:44hillshumshould the e200v2 build fine?
20:42:08bluebrotherLambdaCalculus37: I guess you have libusb-0.1.12?
20:42:40LambdaCalculus37I do.
20:43:21 Quit Llorean ("Leaving.")
20:44:22kadobanhillshum: seems to build for me
20:44:34bluebrotherwell, you could try using the one I have on my box and check if it's the same as one of yours. File is here:
20:44:50hillshummust me on my end then
20:45:54 Quit lymeca (Connection timed out)
20:46:27 Join miepchen^schlaf [0] (
20:47:08 Join dandin1 [0] (
20:47:31 Join MethoS [0] (
20:49:24LambdaCalculus37bluebrother: Third time's the charm. :)
20:49:41*LambdaCalculus37 tries once again
20:53:52*domonoky just commited the e200v2 button fix. Its still ugly because of the interrupt disable while lcd transfers. But i can finally hear music on my e200v2 :-)
21:02:10jaykaywhat is missing at the e200v2s to be called supported? or is there any (not outdated) wiki page for that?
21:02:48gevaertsjaykay: stable playback and support for >1GB storage
21:02:54hillshumThe page is SansaV2 and a manual
21:03:13domonokyjaykay: status on SansaV2 wiki page is still pretty much current.
21:03:51domonokyalso recording is still missing for all ams-sansas. but thats not strictly needed for a release.
21:04:34jaykaydomonoky: may i remove "cannot read scroll wheel" in SansaV2?
21:05:17domonokyjaykay: yes, while the button driver is still not ready, it works now fine.
21:07:33*domonoky just updated it :-)
21:07:42pyro_maniachi there, I#ve got
21:08:00linuxstbjaykay: The TargetStatus wiki page says (generally) what's needed for a device to be called "supported" -
21:08:11pyro_maniacI have got a question again. what happens on "make zip"
21:08:30bluebrotherpyro_maniac: the resulting files of a build get zipped up :)
21:08:57pyro_maniacbluebrother: where can I find the commands for that?
21:09:30linuxstbpyro_maniac: The "Makefile" calls
21:10:04 Quit bluebrother (Read error: 104 (Connection reset by peer))
21:10:07pyro_maniacok, because I got problems with unpacking with tar
21:11:10linuxstbpyro_maniac: ? "make zip" makes a zip...
21:12:22pyro_maniaclinuxstb: I thought tar can unpack even zip files. oh and ark got problems either
21:13:11 Join bluebrother [0] (n=dom@rockbox/developer/bluebrother)
21:14:08bluebrotherstupid network dropout
21:14:19bluebrotherpyro_maniac: tar can handle tar. Nothing more, nothing less.
21:14:29 Join Taylor [0] (
21:14:33jaykaycan someone look whether my edit in SansaV2 is ok?
21:14:54pyro_maniacbluebrother: ok sorry for that and thanks for help
21:15:05bluebrotherLambdaCalculus37: any news? I'm going to be afk for a couple of minutes now
21:15:37LambdaCalculus37bluebrother: Still building.
21:16:12Taylorwe've made some progress with the "vulnerability" in the new encrypted ipods - Linux4nano is looking for help from 5/5.5g firmware and RE experts- if you are interested head over #linux4nano-dev
21:16:39bluebrotherok, so there's some error in my patch. Well.
21:17:46bertrikyuk, do we really need a dummy read/write to the display to make the buttons work?
21:18:09bertrikfeels like a very hacky workaround to me
21:21:07bluebrotherLambdaCalculus37: updated patch:
21:22:04LambdaCalculus37I need to clean up the rbutilqt folder a little bit first.
21:22:41*bluebrother should've updated the timeout value the same time
21:22:55n1sTaylor: what progress have you made?
21:24:26 Quit Anges (Remote closed the connection)
21:24:42 Quit Nico_P (Remote closed the connection)
21:25:13 Join z35 [0] (
21:26:27 Quit perrikwp|class (" ajax IRC Client")
21:27:15 Quit robin0800 (Read error: 104 (Connection reset by peer))
21:27:15LambdaCalculus37bluebrother: Building once again.
21:27:46 Join Anges [0] (
21:28:07bluebrotherLambdaCalculus37: updated patch once more −− overlooked another thing (sic):
21:28:30bluebrotherI also increased the timeout time this time (wow, what an amount of time ;-)
21:29:08 Quit fml ("CGI:IRC")
21:29:20Unhelpfulbufalloc-like allocator is at FS #9916, if anybody wants to take a look. i'll work on integrating it into pictureflow as a bufalloc replacement next.
21:29:56*LambdaCalculus37 stops his build, removes the older patch, and adds bluebrother's new patch in instead, and starts up once again with building
21:32:33 Join robin0800 [0] (
21:35:27 Join {-phoenix-} [0] (
21:42:20pyro_maniaccan anybody help me with logf() ?
21:45:19 Join MethoS- [0] (
21:51:21 Quit MethoS (Read error: 60 (Operation timed out))
21:53:39 Join Chesteta [0] (
21:54:19gevaertspyro_maniac: what sort of help?
21:56:35 Quit robin0800 (Read error: 104 (Connection reset by peer))
21:56:51Chestetahello; I just updated to the latest svn and I have a feeling something is wrong... note: i just ran through a svn co svn:// rockbox command (checked out a totally new clean svn) see −−>
21:57:04Chestetaif anyone could help me I would appreciate it
21:57:26pyro_maniacgevaerts: I've put the "usb_serial_send("test\n",5);" into the usb_serial.c as you said. but after connecting with see you nothing happens
21:57:50n1sChesteta: "arm-elf-gcc: command not found" is a pretty strong clue...
21:58:14Chestetayeah; but since its a totally new svn checkout... should i reinstall cygwin?
21:58:52gevaertspyro_maniac: hm. That seems to mean that usb serial broke since I last checked. I'll look into it soon (probably not today though)
21:59:06n1sthe compiler has nothing to do with your checkout, make sure that you follower all the instructions for installing the cygwin environment
21:59:42Chestetaok; things had been compiling properly before... I will just reinstall cygwin :)
22:00:07gevaertsYou just need to set your $PATH correctly...
22:00:16 Quit CaptainSquid ("Miranda IM!")
22:00:42Chestetagevaerts; was that to me or to pyro_maniac?
22:00:53gevaertsChesteta: to you
22:01:27Chestetaok; how can I set my $path correctly? I am not very familiar with cygwin, all I know is patches and compiling rockbox...
22:01:32pyro_maniacgevaerts: could I test anything on the connection? debugmode said: cu: fconn_open: Opening port /dev/ttyUSB0 (default speed) cu: fsserial_open: Baud rate is 9600 cu: fconn_set: Changing setting to 0, 0, 2 Connected.
22:01:35 Join casainho [0] (n=chatzill@
22:02:10gevaertspyro_maniac: there's something wrong on the player side I think.
22:02:42Chestetaok; thank you :)
22:02:52 Join mcuelenaere [0] (n=mcuelena@rockbox/developer/mcuelenaere)
22:03:05pyro_maniacgevaerts: can you tell me more?
22:04:05casainhohello :-) −− I am working ion drivers for buttons, and I would like to know what values should be returned on "int button_read_device(void)"... like if I want to return a value for my target button_up, what value should I return?
22:04:22bluebrotherLambdaCalculus37: argh. You need a universal libusb. Now how to get that ...
22:04:25gevaertspyro_maniac: I'm trying to reproduce it now
22:04:55 Nick hillshum is now known as Hillshum (n=hillshum@unaffiliated/hillshum)
22:05:12bluebrotherLambdaCalculus37: can you try commenting out line 225 in That might make it only produce a binary for your cpu type
22:05:58 Join akur [0] (
22:07:41 Part akur
22:12:33mcuelenaerehow do I make the bitmaps compile first before any C files get compiled (in a bootloader build)?
22:12:41 Join LinusN [0] (n=linus@rockbox/developer/LinusN)
22:15:05 Quit midijunkie ("?(???~~)?")
22:15:06 Quit Horscht ("Verlassend")
22:15:15 Quit bertrik ("Leaving")
22:18:07 Quit miepchen^schlaf ()
22:18:23 Join miepchen^schlaf [0] (
22:19:19 Quit LambdaCalculus37 (Read error: 104 (Connection reset by peer))
22:22:22 Join LambdaCalculus37 [0] (n=rmenes@rockbox/staff/LambdaCalculus37)
22:24:24 Quit fdinel (Read error: 104 (Connection reset by peer))
22:24:49pyro_maniacgevaerts: can I help or should I wait?
22:25:57 Join matsl [0] (
22:26:11gevaertspyro_maniac: unless you feel like diving head first into the USB code, I don't think you can help much
22:27:25*gevaerts has an idea...
22:27:46*Hillshum listens up
22:29:30***Saving seen data "./dancer.seen"
22:30:30 Join jaykay_ [0] (
22:35:23 Quit jaykay_ (Client Quit)
22:36:21 Join bs66_ [0] (
22:36:49 Part LinusN
22:40:12gevaertspyro_maniac: nearly there...
22:42:27pyro_maniacgevaerts: I got some time to wait
22:42:32 Quit jaykay (Read error: 110 (Connection timed out))
22:43:42 Quit bs66_1 (Read error: 60 (Operation timed out))
22:43:48 Quit gregzx ("ChatZilla 0.9.84 [Firefox 3.0.6/2009011913]")
22:44:08 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
22:47:03gevaertspyro_maniac: r20023 should fix it. If you enable logf in firmware/usbstack/usb_serial.c, it has a logf() call that tells you the input it received back over the serial line.
22:47:06 Join kachna|lappy [0] (
22:49:08 Quit roman_ (Read error: 110 (Connection timed out))
22:49:56pyro_maniacgevaerts: I will test
22:53:58 Quit Chesteta ()
22:56:27 Quit Beta2K (No route to host)
22:56:46 Join gregzx [0] (
23:02:13 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
23:02:39CaptainKwelHi. ipod video 30gb. Is it normal for disk access to completely clobber the battery?
23:02:53CaptainKwelLike a single database init will drain the power all at once.
23:03:12LambdaCalculus37That doesn't seem normal.
23:03:14CaptainKwelOr switching songs maybe 5-6 times in succession.
23:03:22 Quit advcomp2019 (Nick collision from services.)
23:03:23 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
23:04:50 Join matsl_ [0] (
23:05:15 Quit matsl (Read error: 110 (Connection timed out))
23:05:29 Join Barahir_ [0] (
23:05:54SoapWorkCaptainKwel, was it your battery bench which appeared to elbow at 65%?
23:06:55CaptainKwelIt would just die altogether.
23:07:09pyro_maniacgevaerts: now I got feedback
23:07:38SoapWorkare we talking the same thing? FS #9890? logiccircut?
23:07:53SoapWorkTrevor beckerson?
23:08:09 Quit Horscht ("Verlassend")
23:08:44gevaertspyro_maniac: great! So now you can start actually using this :)
23:08:53CaptainKwelno and no. Just using any given svn build.
23:09:55CaptainKwelIt's been happening for as long as i've been running rockbox. Thought it was normal.
23:10:16pyro_maniacgevaerts: thanks for help
23:11:10CaptainKwelTrying to figure out which part needs replacing, the disk or the battery.
23:12:07SoapWorkCaptainKwel, as a battery ages the internal resistance of the cells increase. This can lead to a (relatively) healthy looking discharge curve during periods of low power consumption (because voltage drop is a factor of resistance and amperage), but a severe, sudden, drop in measured battery voltage during periods of high current draw.
23:12:21bluebrotherLambdaCalculus37: any more progress on your build attempts?
23:12:44 Join PaulJam [0] (
23:13:00LambdaCalculus37bluebrother: Nah, keeps crapping out.
23:13:23SoapWorkSince there is no way to look inside a battery and say "hmm, 75% capacity left", capacity is infered by looking up voltage against a standard discharge curve.
23:14:07SoapWorkIF your battery is failing, your battery voltage will dip during periods of high current consumption so low that it may trigger the low-battery shutdown.
23:14:11bluebrotherthat's with the line in removed?
23:15:02CaptainKwelI have noticed that happening too.. it would dive and sometimes shut down during periods of long disk access, but then still appear to have a healthy percentage left after rebooting.
23:15:03domonokyLambdaCalculus37: add a -L. to the LIBS in line 226
23:15:32LambdaCalculus37domonoky: Thanks.
23:15:56SoapWorkthere are other possibilities, CaptainKwel, but I think the one I outlined is the most likely.
23:16:23CaptainKwelI picked it up as a refurb so it's probably got more years on it than it looks like it does.
23:16:47SoapWorkPerhaps your HDD is pulling too much juice for some reason. Perhaps your hardware is incorrectly reporting voltage. Perhaps the imps in your iPod are just mucking with you.
23:16:54LambdaCalculus37domonoky: My current line 226 in is this: LIBS += -L/usr/local/lib -framework IOKit
23:17:01CaptainKwelipod gremlins.
23:17:51 Quit Barahir (Read error: 110 (Connection timed out))
23:17:55CaptainKwelhmm. What is the BCM?
23:18:07domonokyLambdaCalculus37: yes, because i think libusb is normally at usr/local/lib in macosx, just add a -L. if you have the lib in rbutils directory
23:19:28 Join Horscht [0] (n=Horscht@xbmc/user/horscht)
23:19:42 Part Anges
23:21:49 Quit domonoky (Read error: 104 (Connection reset by peer))
23:23:07n1sCaptainKwel: a video decoding coprocessor thing
23:24:39 Quit Zoxc ()
23:26:36 Quit mrkiko ("leaving")
23:33:08 Quit jhMikeS (Nick collision from services.)
23:33:14 Join jhMikeS [50] (n=jethead7@rockbox/developer/jhMikeS)
23:39:56LambdaCalculus37I was wondering... shouldn't the images in manual/appendix/images be updated? They're still the old Archos/Rockbox default icons, and that's already very out of date considering that we have the Tango icons by default now.
23:41:17BigBambiThe file format icons?
23:42:11 Quit ender` (" Life is what happens to you when you had other plans. -- John Lennon")
23:45:38LambdaCalculus37We can keep the old icons for the Archos manuals, but we really should use the new Tango icons for newer targets.
23:48:17n1sBigBambi: for future commits it's customary to reference the FS# of a patch/bug in the commit message, and great work on the manual! :)
23:48:51 Join matsl__ [0] (
23:49:25BigBambin1s: I know in general, but as I wrote the patch and the task didn't have any discussion, I didn't think it was necessary - however I'll put it in next time :) and thanks :)
23:50:38 Quit matsl_ (Read error: 110 (Connection timed out))
23:50:57pyro_maniacgevaerts: after the connection is established the buffer should be red out right? I am asking because I only see the test I have written into the transfer_complete
23:52:10gevaertsIt doesn't write out the entire logf buffer
23:53:15pyro_maniacI have to force it by myself?
23:53:38 Quit matsl__ (Client Quit)
23:53:51 Join matsl__ [0] (
23:55:30pyro_maniacI have written some logf into the init functions of the hardware and I want to read that after I have made the connection
23:55:45bluebrotherLambdaCalculus37: if you still can't get the libusb thing to work there's always the option to simply remove all libusb related stuff. I've put up a patch at
23:55:48 Quit matsl__ (Client Quit)
23:56:40gevaertspyro_maniac: you'd have to dump the logf buffer exsplicitely I think
23:56:45LambdaCalculus37bluebrother: I'll try it in a minute. I'm doing some manual work.
23:56:55bluebrothernoticed that :)
23:57:34LambdaCalculus37On the Rockboy manual page, shouldn't the Screen Rotate option be described as "Rotate the screen by 90 degrees" instead of "Rotate the screen by 90 percent"?
23:57:45pyro_maniacgevaerts: can you say where?
23:57:47bluebrotherwon't be around for too long anymore (getting late here) but I'll be around tomorrow. Would be nice if you could confirm my code to work correctly ;-) so we can have that in the next rbutil release
23:57:54n1sLambdaCalculus37: probably :)
23:58:08bluebrotherLambdaCalculus37: definitely degrees. Percent is ... off
23:58:09LambdaCalculus37bluebrother: I'll try it and I'll announce it on the logs.
23:58:14gevaertspyro_maniac: not really
23:58:15LambdaCalculus37Very off.
23:58:36bluebrotherok, great. Thanks a bunch for your work (and sorry for the trouble with it ...)

Previous day | Next day