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 2007-11-07

00:00:18eigmaunless it's Tuesday, then it's three decades. least, that should reset it.
00:00:51eigmait's funny though, because you could easily create a program that requires you to hold a button for two decades
00:00:56hcsNo one has completed the procedure yet...
00:01:04eigmaexcept Chuck Norris
00:01:29 Join JdGordon [0] (
00:01:32hcsI wouldn't consider Chuck Norris "one".
00:02:27*eigma laughs out loud
00:03:46eigmahey JdGordon
00:05:34 Join spiorf [0] (
00:06:43 Quit FOAD (Read error: 110 (Connection timed out))
00:06:43 Nick FOAD_ is now known as FOAD (
00:06:50lemurchuck norris is many
00:07:14hcsFor official Chuck Norris discussion, please join #rockbox-community
00:07:23 Join DarkDaemon [0] (
00:07:42DarkDaemoni was wondering if you guys could help me with a little situation i'm having with itunes?
00:08:02scorche|wDarkDaemon: does it have something to do with rockbox?
00:08:14DarkDaemonno :(
00:08:35scorche|wplease read the guidelines linked in the topic :)
00:09:01DarkDaemonok sorry
00:09:05 Part DarkDaemon
00:09:37 Nick eigma is now known as ChuckNorris (n=cat@
00:10:00 Nick ChuckNorris is now known as eigma (n=cat@
00:10:16 Quit MajorC ()
00:11:10 Quit jgarvey ("Leaving")
00:11:22 Quit linuxstb (Read error: 113 (No route to host))
00:14:14 Nick guyzmo_ is now known as guyzmo (
00:15:07lemurtopic topic
00:20:21 Quit ompaul (Client Quit)
00:22:51amiconn"The one who is many"
00:27:35 Join JRoT|Stage [0] (
00:28:44eigmaamiconn: can I bother you with some LCD initialization sequence questions?
00:31:59 Join Jakeworld [0] (
00:32:35jpt9in a WPS file, is there any way to display the volume as a percent?
00:33:06Jakeworldhey, I tried updating my build today and encountered a major problem with playback, the files endlessly skip forward
00:33:09amiconnThere is no such concept as volume in percent. Volume is logarithic
00:33:15amiconneigma: Just ask
00:33:36JakeworldI tried using the most recent build, and with a clean build, but it was still happening
00:34:12eigmaamiconn: it's a bit of a long story.. have you ever heard of persistent storage (like an EEPROM) in a LCD controller?
00:34:29Jakeworldactually I suppose not the *most* recent, 15498, but I'll give it a shot quickly
00:35:16 Quit Arathis ("Bye, bye")
00:35:50pixelmaJakeworld: what player do you have, what files are these and did they play fine before?
00:36:08amiconnThe Samsung S6B33 has an OTP eprom
00:36:16JakeworldGigabeat F40, musepack sv7 q8 files
00:36:29JakeworldI tried 15470 and it works just fine
00:36:30 Quit Soap (Read error: 110 (Connection timed out))
00:36:53Jakeworldyes, 15500 does it too
00:36:54eigmaall right, here's the entire story
00:36:58Jakeworldthen it data aborts
00:37:31eigmaI was trying to figure out how to control the backlight brightness, and by looking at the OF, I found that whatever controls it hangs off the SPI bus; I also figured out which GPIO was the chip select.
00:37:35JakeworldI'd say what the data abort is, but the LCD is off
00:38:02 Join dandin1 [0] (
00:38:10eigmathe commands are something like this: A4h 00h XXh BBh, where XX is between 00h and 80h and sets the brightness of the backlight.
00:39:10pixelmaJakeworld: hmm... can't help you there (I neither have a gigabeat nor musepack files). I guess though that it's still some resulting bugs from the Metadata on buffer changes and followup changes. Maybe Nico_P could look into it, he's involved in this and has a gigabeat
00:39:12eigmait turns out that BBh also controls something about the screen; I think it's something like contrast.. when too high or too low, the screen looks washed out, almost a negative image and in shades of gray
00:39:51eigmaI tried sending other Axh commands, I scanned from A0h and AFh, and I somehow got it stuck on the 00h contrast setting.. it wouldn't go back to normal even after disconnecting the battery
00:39:56JakeworldI'll try and pass it along to him then, I'll give some other codecs a run and see if they do it as well
00:40:09 Join Angryman [0] (i=50e0a118@gateway/web/cgi-irc/
00:40:26 Quit bertrik ("bye")
00:41:03eigmaI also found that the OF sends a command that returns data (the only one that I have seen to far that is a "read"), which is A8h 00h.
00:41:23eigmawhen the screen was borked, it was returning a value that was causing the OF to skip some initialization commands
00:41:51 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
00:41:55eigmaI tried sending those commands myself manually and the screen was fixed, for good −− and from then on, A8h returned a value that caused the OF to send those initialization commands on every subsequent boot
00:42:16eigmais this making any sense so far?
00:44:07 Quit Angryman (Client Quit)
00:44:45amiconnNot really. I haven't seen an lcd controller with an eeprom
00:45:11jpt9oh yeah... is there any way you guys could get a Z-machine running under rockbox? :-)
00:45:16amiconnBut then the lcd controllers I know are for lower resolutions than the m:robe
00:45:54 Quit ender` (" Doesn't it strike you as mildly ironic that most of the people who are against abortion are people you wouldn't want to fuc")
00:47:08eigmahrm :\
00:50:16Jakeworldinteresting, it would appear my seeking bug is exclusive to musepack, at least in the codecs I've tried
00:51:21 Join Soap [0] (n=Soap@rockbox/staff/soap)
00:51:53 Quit JRoT (Read error: 110 (Connection timed out))
00:53:17 Quit XavierGr (Nick collision from services.)
00:53:18 Join XavierGr_ [0] (
00:54:35 Join atsea- [0] (i=atsea-@gateway/tor/x-74ed19e71ae2c30b)
00:56:11 Quit japc (Remote closed the connection)
00:56:49 Quit linuxstb_ (Read error: 113 (No route to host))
00:58:32 Join spiorf_ [0] (
01:00:01 Quit cooz ()
01:00:46Nico_PJakeworld: I'll test
01:01:02Jakeworldoh, I was just about to file a bug report
01:01:02 Quit scorche|w ("CGI:IRC (Ping timeout)")
01:01:21Jakeworldit seems it only occurs with musepack files
01:01:39Jakeworldthough I only tested ogg, musepack, and LAME mp3
01:01:50Nico_Phmm yeah I have it too
01:02:03 Quit Genre9mp3 ("I don't suffer from Rockbox psychosis. I enjoy every minute of it.")
01:02:05Nico_PI remember having it at an early point of MoB development
01:02:43JakeworldI'm not totally sure which revision is responsible for starting the behavior, but after 15470, and before 15496
01:02:46lostlogicNico_P: random question −− I notice some commented out code WRT cue sheets. What's the deal with that?
01:03:04Nico_Plostlogic: it's preloading which I disabled
01:03:11***Saving seen data "./dancer.seen"
01:05:16 Quit spiorf_ (Remote closed the connection)
01:05:16lostlogichmm, I don't get it.
01:06:19Nico_PJakeworld: no need to file a bug report, I intend to fix the bug ASAP
01:06:29Nico_Plostlogic: what don't you get?
01:06:30Jakeworldsure thing, I closed it out
01:07:12hcslostlogic: yo, tracking down issues, want to play along>
01:07:36lostlogicNico_P: I guess I don't get what cue sheet support is or why that bit of code was there in the first place, so I therefor don't see why it's disabled. I'll ignroe it for now.
01:07:43lostlogichcs: nope, only chatting ATM.
01:08:00lostlogicNico_P: you're fixing rebuffer_handle? Because that would make you my hero... again.
01:08:08lostlogichcs: what issue you tracking though, out of curiosity
01:08:14Nico_Plostlogic: not right now... it seems bufread is broken
01:09:09hcslostlogic: bugs with ADX. seem to involve backwards seeking not working
01:09:10Nico_Pwow... segfault in the sim
01:09:18Nico_Pand yeah, bufread is the culprit
01:09:24hcsNico_P: I've been getting those all the time now.
01:09:51Nico_Phcs: the ADX bugs? I thought lostlogic fixed them
01:09:57 Quit spiorf (Read error: 110 (Connection timed out))
01:10:36hcsNico_P: Well, I rewrote it as it should have been. And I meant I've been getting segfaults in the sim all the time.
01:10:57Nico_Pah. on what codecs?
01:11:04hcsany, it seems
01:11:30 Join barrywardell [0] (n=barrywar@
01:11:35Nico_Peven MP3?
01:12:23hcsmust be something wrong with my libraries... seems to happen within ALSA
01:14:41 Join billenium [0] (
01:15:10Nico_Pdid you update the codecs? (make && make install)
01:15:29hcsNico_P: yeah, from scratch
01:16:08 Part pixelma
01:20:01 Join Rob222241 [0] (
01:20:02 Quit Rob2222 (Read error: 104 (Connection reset by peer))
01:20:16 Quit Dark_Apostrophe (Connection timed out)
01:20:17hcsIf I attempt to seek backwards in a file when that part of the buffer has been overwritten, what should happen? It looks like I just get an error.
01:21:45lostlogichcs: that's broken.
01:21:53lostlogichcs: how broken varies on other factors
01:22:22hcsThus "this is broken" in the comment, I suppose.
01:22:36hcsGuess it's back to the old ADX, eh?
01:23:14lostlogicNico_P: hmm, I'm slightly confused.
01:23:22hcsso much for playing nice...
01:23:50lostlogichcs: why are you changing back?
01:23:57Nico_Plostlogic: by what?
01:23:59lostlogicall codecs have the problem of seeking backwards off of the buffer
01:24:10lostlogicNico_P: that diff doesn't apply to current SVN afaics
01:24:20hcslostlogic: because ADXs almost always do this seek, so I run into it an awful lot
01:24:38Nico_Plostlogic: no it doesn't... it's from MoB prehistory
01:24:39lostlogichcs: they seek back to the beginning of the file later during playback!?
01:25:01Nico_Plostlogic: MPC relies on its bufread calls being satisfied
01:25:02hcslostlogic: not necessarily clear to the beginning, but back, yeah
01:25:33lostlogichcs: :( that... that's potentially problematic.
01:25:33 Quit JRoT|Stage (Read error: 104 (Connection reset by peer))
01:25:50hcslostlogic: That's why I did those huge requests, to keep the beginning of the file from being overwritten. Why did you think I was doing it?
01:25:52lostlogichcs: if the buffer's been shrunk since it last seeked there it'll be off buffer regardless of old vs. new buffering code I think
01:26:47lostlogicNico_P: hmm, by definition a bufread call can be checked &| looped over to get the full amount of data
01:26:55lostlogicit's only bufgetdata that is pretty much past-fail
01:27:08eigmaamiconn: know anything about the GCP signal in EPSON LCDs?
01:27:17lostlogicbecause bufgetdata may have a permanent unsatisfiable request for contiguous data
01:27:27lostlogicbut bufread can always fill furthre into the client's poitner
01:27:34lostlogicwhat's your proposed solution?
01:27:50Nico_Plostlogic: yeah but the fact that the chunksize check is common to both breaks bufread... I'll move it to bufgetdata
01:28:01lostlogicNico_P: no.
01:28:31lostlogicNico_P: neither one is allowed to request data that would require a buffer_handle request to read more than a single filechunk in order to satisfy it
01:28:35Nico_Plostlogic: why? also in it, size = MAX(avail,BUFFERING_DEFAULT_FILECHUNK); looks wrong to me... shouldn't it be MIN?
01:28:41lostlogicintroduces potential (likely) long waits
01:28:47Nico_Plostlogic: I don't agree
01:29:03lostlogicno, it's _ok_ to satisfy more than a filechunk if it's already available
01:29:24lostlogicNico_P: we cannot offer any guarantee of _when_ any more than _one_ chunk of data will be read by a buf_request_buffer_handle call.
01:29:33Nico_Pyeah but then you return more than originally asked for
01:29:46lostlogicso unless it loops the buf_request_buffer_handle call it's absolutely illegal to request more than filechunk mor than avail
01:31:02Nico_Pyeah I see what you mean
01:31:03lostlogicNico_P: could do a MIN(size,MAX(avail,filechunk)).
01:31:31Nico_Pyeah that's what I was thinking
01:31:49lostlogicI think that's right, I had misthought that size was already > avail at that point.
01:32:19 Quit eigma ("homework")
01:32:45Nico_Pin my case it wasn't I think... size was less than avail but then size became avail
01:33:01Nico_Pand trying to copy that much data failed
01:33:02lostlogicNico_P: yeah, definitely possible.
01:33:08lostlogicI were wrong.
01:33:16Nico_Psize = MIN(size, MAX(avail, BUFFERING_DEFAULT_FILECHUNK)); fixes MPC
01:33:20Nico_PI'll commit
01:33:21lostlogicgreat :)
01:33:40Nico_Pthis is a very delicate alchemy :p
01:34:12lostlogicNico_P: yeah, I'm really glad that we have two minds actively engaged in it because we seem to have different perspectives that end up with a better final result.
01:34:20 Quit Rob222241 (Read error: 104 (Connection reset by peer))
01:34:39hcsSo help me out here, minds.
01:34:55Nico_Pvery true... I'm also very thankful that you helped out that much
01:34:58hcsDo I give up and revert?
01:35:07lostlogichcs: I don't think revert will fix
01:35:13lostlogicI have a theory though
01:35:15Nico_Phcs: wait a minute
01:35:27 Join Rob2222 [0] (
01:35:34 Join JRoT|Stage [0] (
01:35:47lostlogic(I'll speak my theory before I hafta run here and then Nico can get to you after his minute)
01:36:36lostlogicwhat kind of backward seeks are likely in adx?
01:37:09lostlogicmy idea is that you can avoid advancing the buffer after reads until you won't need the earlier data any more but that might depend on where buffer wraps fall :(
01:37:27hcslostlogic: that was the logic on which the old version worked
01:37:40scorchehcs: but it got lost?
01:38:07hcsit seemed to work pretty well until I got to the Sansa
01:38:28hcsand all these changes being made to buffering recently have destabilized things
01:38:38lostlogichcs: hmm, so what happens if the distance between playing and earliest seek needed is wrapping the buffer end?
01:39:11lostlogichcs: I think you're right that with the latest patches to buffering today the old code might be more stable than the new.
01:39:31hcsI can accept a smaller buffer than the whole file, then when it hits the end it requests the whole thing again.
01:39:31lostlogic(even though it might cause some logf warnings from the buffering code)
01:39:48lostlogichcs: but what if the whole thing will never be satisfied because it wraps the buffere nd?
01:39:52lostlogicthat's what I'm concerned with.
01:40:03hcsyeah, it's a bad strategy, don't know why it worked so long
01:40:23Nico_Phcs: I ended up not trying your ADX files because I thought the prob was fixed, but I'll have a look
01:41:03lostlogicNico_P: "which would be bad" <−− :)
01:41:18hcsNico_P: please do, those three are large enough that the first two fill the buffer, then when the third is loaded near the end of the second (but before the second is entirely done) things go awry
01:41:28Nico_PJakeworld: the problem is fixed... thanks for reporting
01:41:43 Join iamben [0] (
01:41:49Jakeworldno problem, thanks for the awesome work you do :D
01:42:27lostlogicok, I gotta run, sorry I don't have a good answer for your complex codec there hcs :(
01:42:39hcslostlogic: thanks for your attempts
01:42:59Nico_Phcs: I probably won't come up with one this evening either... it's already late here
01:43:16hcsglad to have more minds on the problem
01:55:22 Quit einhirn (Read error: 104 (Connection reset by peer))
01:55:43 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
01:55:51 Quit advcomp2019 (Nick collision from services.)
01:55:55 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
01:56:38 Quit Soap (Read error: 110 (Connection timed out))
01:57:10 Quit billenium ("Ex-Chat")
01:59:24 Quit Rob2222 (Read error: 104 (Connection reset by peer))
01:59:58 Join Rob2222 [0] (
02:01:03 Join Soap [0] (n=Soap@rockbox/staff/soap)
02:02:04 Join donutman25 [0] (n=chatzill@
02:02:26 Quit Jakeworld ("ChatZilla [Firefox]")
02:03:36Nico_Pdionoea: I comitted a fix to the problem you reported earlier
02:06:48Nico_Phcs: sorry, I'm really too tired to investigate the ADX problems today... bed time
02:07:01Nico_P...or after I fix my warning
02:07:05*Nico_P slaps forehead
02:15:56 Join billenium [0] (
02:16:48 Join Thundercloud [0] (
02:16:59 Nick fxb is now known as fxb__ (
02:17:43 Quit Nico_P (Remote closed the connection)
02:20:31 Join micols [0] (
02:22:06 Quit micols_ (Read error: 104 (Connection reset by peer))
02:22:20 Quit JRoT|Stage (Read error: 110 (Connection timed out))
02:24:50 Join JRoT|Stage [0] (
02:26:31 Quit petur ("Zzzzz")
02:27:27 Quit Thundercloud (Remote closed the connection)
02:27:31 Quit dandin1 (Read error: 104 (Connection reset by peer))
02:29:46 Join dandin1 [0] (
02:36:34 Join sdoyon [0] (
02:37:30 Quit jhMikeS (Read error: 104 (Connection reset by peer))
02:39:24 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
02:43:36 Quit barrywardell ()
02:45:47lostlogicNico_P: (If you read logs) Nice fix for something I missed with my id3v2 dropping
03:03:12***Saving seen data "./dancer.seen"
03:05:05 Quit JRoT|Stage (Read error: 110 (Connection timed out))
03:06:50 Join JRoT|Stage [0] (
03:08:28 Join Brian_ [0] (
03:13:17 Join Dark_Apostrophe [0] (
03:15:24 Join allele [0] (
03:16:23Brian_is there rockbox planed for sansa clip
03:17:34safetydanBrian_: new ports aren't really planned they just happen when someone is interested enough
03:18:00safetydanBrian_: see
03:18:03SoapI don't want to quote the party line, because it always sounds...well..."jerky" - but it is true. Rockbox is never "planned" in such a manner. People port it to the players they are willing and able to port it to. There is no central command issuing orders like "port to player X" or "make more games".
03:18:28alleleI installed the rockbox toolchain using 'sudo sh', but when I try compiling rockbox I get "arm-elf-gcc: command not found".
03:19:44Soapallele, that is a failure to add the newly compiled (via the shell script) cross-compilers to your PATH. This step is covered in the wiki. Give me a sec and I'll pull up the page if you are not currently looking at it.
03:23:01Soaplooks like that step isn't mentioned explicitly in after all, allele. See the very last part of the Manual Build section of that page for a mention of adding the compilers to your PATH.
03:26:22 Join barrywardell [0] (n=barrywar@
03:27:25 Quit Brian_ ("Leaving")
03:27:45TTThomasok. thats scary. I'm not going to be trying on my old rio carbon anytime soon like i was pondering.
03:28:11Lloreansdoyon: If it's supposed to be _spelling_ the filename, why should it speak the filetype rather than spell it at the end?
03:29:30sdoyonLlorean: The idea is to cut down on the painful spelling you have to listen to. You can quickly skip over files of the wrong type.
03:30:17LloreanPerhaps there should be an option "speak type first" that speaks the type of what you have highlighted first (whether spelling or using .talk clips)?
03:31:14LloreanIt seems to me even with .talk clips you might want to know the type first, meanwhile just arbitrarily speaking the type first when people expect spelling is forcing your choice on users who may not want it (it would actually slow me down a lot since I don't have any repetition in my lists, so I just use the first letter or two to identify what's highlighted)
03:31:53LloreanHearing "Folder" first before hearing the beginning of the folder name would slow me down drastically the few times I use voice.
03:32:00 Join RaRe` [0] (
03:32:02sdoyonLlorean: I had a setting but I took it out, expecting people wouldn't want me to add a new one, and I had had positive feedback...
03:32:20LloreanIt sounds like you're creating more repetition, rather than less.
03:32:49 Quit Rob2222 (Read error: 104 (Connection reset by peer))
03:33:01LloreanHearing "Folder" before every folder and having to wait longer before you can hear the first few letters of its name, hearing "MP3" before the letters to identify the track name...
03:33:22 Join Rob2222 [0] (
03:33:55LloreanIt only really seems useful if A) You're very uncertain of the filenames on your player, and B) You've got a mix of filetypes in each folder. Most times you only have Folders, and Files, and I think a lot of times you don't have files and folders alongside each other, since the first entries are always folders, so as soon as you get to a letter *before* the previous letter, you know you're past the folders.
03:34:08sdoyonLlorean: well, in use I found it helped. It's true in some cases it'll introduce repetition.
03:34:25LloreanIn what situation does it reduce repetition?
03:34:32sdoyonLlorean: Try looking for a bookmark file...
03:35:10LloreanSo, it reduces repetition for bookmarks while increasing it in "Scrolling through folders" and "Scrolling through MP3 files"
03:35:15LloreanWhich ones are more commonly used features?
03:35:59sdoyonLlorean: Well, I guess I should have brought this up here then. I had kind of assumed it was OK as it had been languishing on the tracker for a long time. I'm not blaming you mind you, hard to follow everything on there.
03:36:13 Quit hardeep ("leaving")
03:36:40LloreanWell, I didn't expect it to be committed one-sided, since it's very definitely a case of something that works contrary to the setting name, since it's not spelling any more.
03:37:30 Join lemur_ [0] (
03:37:39sdoyonLlorean: uh... there's still a lot of spelling going on :-). You know it's not a big deal, I'll just revert it while we discuss a better alternative then.
03:37:54LloreanI think an option "Speak extension first" is all you need.
03:38:07LloreanAnd have it speak it for either .talk or spelling if that's enabled.
03:38:08 Quit barrywardell ()
03:38:16LloreanSince I *can* see it being useful, but think it is even for .talk when it is.
03:39:52sdoyonLlorean: .talk is another case IMHO. You hear the whole name in under 1.5s most of the time. I think it's best to include the extension in the .talk clip (although the current .talk generation script doesn't do that I believe). The file types that are defined are not so specific: it's audio, not mp3 for example.
03:40:03qwmbig yawn.
03:40:34Lloreansdoyon: Still, folder names wouldn't have extensions. Files don't necessarily need extensions either, if they're not audio files.
03:41:01Lloreansdoyon: Even if you do have .talk clips generating extension, and even if you have it generate .talk clips for *all* files and folders, it's possible that you might want to hear them first.
03:41:24LloreanIt seems at least as useful to me as speaking them for spelling.
03:41:51sdoyonLlorean: this commit has it say "folder" for .talk clips as well, only AFTER, as most of the time you know it's a folder already. And when there's no extension, there's no type...?
03:42:27sdoyonLlorean: "hear them first"?
03:42:40LloreanThe type.
03:42:46LloreanI still don't understand the difficulty with spelling.
03:42:55LloreanWhy is it difficult to find a bookmark file?
03:42:59 Join nanok [0] (n=nanok@
03:42:59LloreanIt's named the same as the audio file, right?
03:43:16LloreanJust a different extension, and since sorting is alphabetical, you'll know whether it'll be before or after.
03:43:21LloreanOr same as the playlist.
03:43:44 Quit allele ("Leaving")
03:45:47 Nick XavierGr_ is now known as XavierGr (
03:45:54 Quit RaRe (Read error: 110 (Connection timed out))
03:46:42LloreanEither way, I think forcing people to listen to the type first is a bad choice to make, especially if they willingly choose to have it spelled because they want it spelled
03:49:14sdoyonLlorean: Well, at the risk of sounding stupid... I admit this one was kind of in the backlog of patches, and I have been using .talk clips a lot more than spelling for a while. You make good points. Hearing folder all the time could often be annoying in particular.
03:51:16sdoyonLlorean: I remember liking this much when I had a mix of file types, configs and firmwares and such. And the point about the bookmark: it's still faster in some cases to listen just for folder or bookmark with little or no spelling...
03:52:18Lloreansdoyon: I can see that as being true only if you don't know what file you've saved the bookmark for, or if you can't remember.
03:52:26LloreanBut either way, I'm not saying it's a bad option
03:52:30LloreanJust a bad thing to force on users. :)
03:53:19 Quit lemur (Read error: 110 (Connection timed out))
03:54:20Langlywow this mp3 that wont play in rockbox is one weird sucker
03:54:37Langlyplays in winamp, but open it with something like goldwave and it shows as having no sound
03:55:27LanglySilent in windows media player
03:55:27sdoyonLlorean: well I can just add the setting I used to have. But there's probably ways to make it more repetitive.
03:55:46sdoyonLlorean: ... uh... less repetitive?!
03:55:59LloreanWell, I'm not sure of a good way to make it less repetitive.
03:56:07LloreanBut you can think about it, and I will.
03:56:16LloreanNo need to rush something, as long as we don't fix about it
03:56:47sdoyonLlorean: well do I add the setting or revert?
03:57:05LloreanI'd say add the setting for now. It sounds useful enough that it should be made available for users.
04:00:16JdGordonwe havnt got enough settings already?
04:00:31Langlywow this file is weird
04:00:42sdoyonNow that's what I expected to hear...! :-)
04:02:41Langlywell thats even more bizzare
04:03:13Langlyrecoded the file using VLC as no editor could open it, file now looks like solid volume all the way in an editor, but plays perfectly
04:04:39 Quit XavierGr (Remote closed the connection)
04:04:40 Quit Soap (Read error: 110 (Connection timed out))
04:04:47 Part Llorean
04:04:52 Join XavierGr [0] (
04:05:08 Quit dandin1 ()
04:05:29 Join Mouser_X [0] (i=cf9bb003@gateway/web/cgi-irc/
04:05:35 Join Soap [0] (n=Soap@rockbox/staff/soap)
04:07:32 Join barrywardell [0] (n=barrywar@
04:20:13nanokas i thought: the steapness of the curve increases dramatically after a certain voltage
04:20:45nanokat about 3.6 volts it starts to be obvious, it takes about 20 minutes form there to shutdown
04:20:54nanok(sansa e200)
04:21:23nanokup to that point, it seems to stay fairly linear
04:21:38nanokbut i would have to do more tests..
04:21:59sdoyonnanok: Interesting. I recently added spontaneeous voicing of battery level, but didn't know when a final warning would be appropriate...
04:28:00nanokcurrently it seems to shutdown at 3277 mV, but i need more samples
04:28:36nanokfrom what i have now, it seems a smarter threshold would be 3500 (would loose only about 10 minutes of play time)
04:28:55nanokor at least 3450, 3400
04:29:13sdoyonnanok: is there a notion of unsafe level to do writes to flash?
04:29:18nanokin that range the curve is getting quite steap, so the battery is not very happy
04:29:43nanoksdoyon: interesting point. that i think should be considered also ;)
04:30:08nanoksdoyon: right now i'm just playng with the bat bench and ploting with gnuplot
04:30:43nanokthe main reason is that the battery indicator is currently fairly.. dumb. and i think it can be made _very_ smart very easyly
04:31:07 Quit jmworx (Read error: 104 (Connection reset by peer))
04:32:08nanoki thought about a safer level of shutdown, so as not to stress the battery too much, but now that you mention it, writes to "disk" would be another thing to consider (but i think the lowest of the two should just be set as shutdown threshold)
04:33:26nanokso, i take it we still haven't got the new usb code.. :(
04:33:57*nanok hates having to use the OF to charge the battery, not to mention file transfers
04:34:12sdoyonDigressing a bit, but why is the estimated battery life so low on my X5 (X5L actually), like 6hours, while it used to be more like 15ish (and more in reality)?
04:34:28 Join Brian_ [0] (
04:34:47nanoksdoyon: i think it has a lot to do with the theme settings also
04:35:03Brian_When will rockbox for sansa clip be released
04:35:19sdoyonnanok: I mean just the estimate, in info screen...?
04:35:30nanokcurrently, as far as i understand, the voltage is read and, based on that, a "blind" linear estimate is made
04:35:58nanoksdoyon: oh, okay, that one i think is theme independent
04:36:21sdoyonnanok: must have changed at some point in the last months, it used to be closer to reality...
04:36:28advcomp2019Brian_, do not know because there is no timeline
04:37:13nanokso basically you have 4.2 i think which is 100% and the shutdown threshold which is 0%, between them, the rb estimates linearly where you are, and gives you a percentage
04:37:14advcomp2019Brian_, plus the people here does not have much info of what is inside the clip
04:37:22nanoknow, for the estimated time, i am not so sure
04:39:50nanoki am not certain if it estimates the consumption somehow or it is also just based on some figure (like, 100% on target X is y hours, so we have this many hours left)
04:40:37nanoksdoyon: and i am not smart enough to read the code and tell ;)
04:43:32 Quit Brian_ ("Leaving")
04:52:17nanokthe idea would be to set the thresholds to something more suitable, and maybe to have a simple algorithm to estimate the time remaining and percentage more realitcially, like estimating the steapness of the curve locally, and predicting the shutdown time based on that (this should mean one would also have a better idea of how the current "avergae" usage pattern influences the lifetime)
04:52:33nanokand it should be easy enough to do
04:52:45 Join KWhat_Mobile [0] (
04:53:12nanoki think by the end of the week i should have enough data to be able to make some acceptable estimations
04:53:33KWhat_MobileIs the problem with the X5 nightly as of last night preventing it from playing mp3 files?
04:54:41nanokKWhat_Mobile: can you find a bugreport about that?
04:55:02KWhat_Mobileyou have ~bugzilla somewhere?
04:55:12KWhat_Mobile ?
04:55:21nanokdamn, i nedd to go to sleep
04:55:22sdoyonKWhat_Mobile: err well FWIW mine seems to play...
04:55:35nanokKWhat_Mobile: i think that's it, yes
04:55:57nanokKWhat_Mobile: you have a search box on the main page labeled "flyspray"
04:56:25KWhat_Mobileyah got it
04:56:32KWhat_Mobiledoes this thing show the build id in a menu somewhere
04:56:53 Quit hcs (Read error: 104 (Connection reset by peer))
04:56:58nanokKWhat_Mobile: on the player you mean? system, rockbox info
04:57:03KWhat_Mobilewooo i got the future build
04:57:13 Quit Rob2222 (Read error: 104 (Connection reset by peer))
04:57:27KWhat_Mobiler15493 dec 4 07
04:57:52 Join Rob2222 [0] (
04:58:15 Quit homielowe (Remote closed the connection)
04:58:21nanokKWhat_Mobile: 15507 is the latest i have, got it 1/2h ago or so
04:58:46nanokso yours is maybe just labeled wrong (timestamp)
04:59:06 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
04:59:17nanokit is not really from the future (fortunately, that would have been something to keep me up at night :-P )
05:00:15KWhat_Mobilelol, its my local time, need ntp for this thing
05:00:24nanokKWhat_Mobile: ;)
05:00:52 Quit JRoT|Stage (Read error: 110 (Connection timed out))
05:01:34 Join JRoT|Stage [0] (
05:01:37KWhat_Mobilestrange fixxxed the date and now the booger players =)
05:01:59KWhat_Mobilegood job on a sweet ass app guys
05:02:23KWhat_Mobilei bought this player specifically for rockbox
05:03:15***Saving seen data "./dancer.seen"
05:03:49sdoyonLlorean: still around?
05:07:08 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
05:07:13 Join miepchen^schlaf [0] (
05:09:51jpt9is there any way to have Rockbox as an application on a Nokia 770?
05:09:54jpt9it's be really cool.
05:10:09jpt9or on pretty much any other pda.
05:10:24jpt9it could totally kick pocket tunes' ass.
05:12:04krazykitjpt9, if the 770 has SDL, it can run the sim as an app.
05:12:25jpt9it might...
05:12:28jpt9not sure.
05:13:22jpt9there is sdl.
05:13:30jpt9they've done dosbox:
05:13:56Mouser_Xkrazykit, jpt9: Running the Rockbox SIM, while possible, won't be idea...
05:14:10jpt9yeah... i know...
05:14:10Mouser_XThe SIM has problems.
05:14:22jpt9i mean, you wouldn't want it as *firmware* per say...
05:14:26jpt9the 770 already runs linux.
05:14:34Mouser_XAh, sorry, I didn't recognize your name (I don't come around here frequently.)
05:14:41jpt9but you wouldn't want it to simulate an existing device.
05:15:04Mouser_XWhat I mean, is that audio output in the SIMs is lacking.
05:15:13jpt9i noticed.
05:15:23 Join bb [0] (n=bb@unaffiliated/bb)
05:15:40jpt9it's a bit crackly (quiet artifacts occasionally), and a bunch of the audio settings don't seem to work.
05:16:36jpt9otherwise, it makes a great mp3 player.
05:16:41jpt9just needs a real scroll wheel :-)
05:17:43jpt9you guys should rig up the mouse wheel and button to scroll up/down and enter...
05:18:55Mouser_Xjpt9: Check the Wiki. Someone is working on adding touchpad/mouse support.
05:19:04Mouser_X(I don't remember the article's name.)
05:19:19jpt9i just mean in the simulator...
05:19:29jpt9basically add a couple of keybindings.
05:19:34Mouser_XI know. It'd be added to the SIMs as well.
05:19:43jpt9what devices would there be that would use it?
05:19:56jpt9afaik, there aren't any touchscreen mp3 players with rockbox ports.
05:20:02Mouser_XThe m:ROBE is entirely touchscreen based.
05:20:14Mouser_XIt has 1 button, and that's the power button.
05:20:38jpt9are there any skins for it? it's VGA!
05:20:53Mouser_XNot yet. The port is still in the works.
05:21:10Mouser_X(Though, that would be sweet!)
05:27:46 Quit bb_ (Read error: 110 (Connection timed out))
05:28:23 Quit barrywardell ()
05:28:45*jpt9 is playing minizork in the simulator while listening to music.
05:28:45 Quit sdoyon ("ircII EPIC4-2.4 -- Are we there yet?")
05:28:59jpt9thank god for the simulator... my Sansa is being shipped via SmartPort :-P
05:29:48jpt9you guys should add the ability to change the font in the text viewer.
05:29:52jpt9(and possibly the editor)
05:34:38 Join Llorean [0] (n=llorean@
05:35:00 Join kkurbjun [0] (
05:35:13Mouser_XMinizork on the simulator?
05:35:19Mouser_XWhat's minizork?
05:35:20jpt9the gameboy game.
05:35:30jpt9it's a miniature (scaled-down) version of Zork.
05:35:33jpt9the text adventure.
05:35:33Mouser_XOh! I didn't know there was a GB of that.
05:35:47*Mouser_X will load it onto his Gigabeat later.
05:36:07Mouser_X(I've already got the GoodGB/C
05:36:18jpt9how fast/slow is the GB emu on a real device?
05:36:38Mouser_XOn the Gigabeat, it's runs fine (though, the Gigabeat has a 300 mhz CPU.)
05:36:50Mouser_XAs for other targets, I've heard it struggles on the iPod.
05:36:59jpt9what about the e250
05:37:12Mouser_X(Most people have audio disabled on the iPod.)
05:37:29Mouser_XNo idea, but I'd guess it's about the same as the iPod, since the CPU types are similar.
05:37:31jpt9it's laggy with audio on the simulator on a 2GHz dual-core laptop :-)
05:37:44jpt9is it still laggy on the ipod without the audio?
05:37:56Mouser_XI don't know. Never played it on an iPod.
05:38:00jpt9amd how laggy are we talking about?
05:38:03jpt9like unplayable?
05:38:09Mouser_XNo, it's playable.
05:38:22Mouser_X(Though, I don't know if that's with, or without, audio.)
05:38:24jpt9a zmachine that used normal save formats would make a great plugin.
05:38:52jpt9(it'd be nice if there were other input methods besides the fullscreen keyboard... perhaps something with a dictionary or at least pairs/triplets of common letters)
05:40:10 Quit hcs (Read error: 104 (Connection reset by peer))
05:40:13Mouser_XHonestly, I think something like a GB game's "Input name" screen keyboard would be nice...
05:40:35jpt9maybe with a few (5 or 10) user customizable words.
05:40:45jpt9and the ability to pick words off the screen (for the text adventure)
05:41:00jpt9that's pretty much what most palm ones provide (besides graffiti), and it works quite well.
05:41:05Mouser_XMulti-screen switching, instead of having to scroll through everything to access stuff.
05:41:33jpt9some sort of input mode (like T9/abc/123/symbols on phones).
05:41:50jpt9you guys should check out the insane number of input methods iPodLinux has, and possibly steal some from there :-)
05:41:58jpt9then again, most of those are designed for a wheel.
05:42:02 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
05:42:04Mouser_XHCS was working on input methods. But not for this purpose.
05:42:46jpt9a binary search/tree thing −− press Up for A-M, Down for N-Z, etc. might be interesting. I wrote something like that for my palm; it was (I think) 5 presses/character, but I actually got pretty quick at using it.
05:48:02jpt9how about user-defined keyboard layouts?
05:49:10Mouser_XI think the problem you run into with changing the keyboard, is in making it too complex.
05:49:28jpt9user-defined layouts would be nice...
05:49:35Mouser_XI don't think a user-defined layout would be a good idea, because of that.
05:49:35jpt9and if you loaded them from a text file...
05:49:44jpt9you could edit them onboard the device :-)
05:54:34 Join qweru [0] (
05:54:53jpt9why do i have a feeling that i'm going to spend a month or so doing nothing but grabbing the latest rockbox updates every 10 minutes and playing with them, and not actually using the 2GB on my Sansa to play any music? :-D
05:54:57jpt9you guys rock!
05:56:12 Quit Rob2222 (Read error: 104 (Connection reset by peer))
05:56:40*Mouser_X is hoping that all the nasty bugs will be fixed soon...
05:56:45jpt9i'm searching for songs, and since it's case insensitive, i'm just jumping between upper/lower case depending on which one's closer.
05:56:50jpt9what nasty bugs?!
05:57:07Mouser_XA lot of stuff showed up from the MoB commit.
05:57:17 Join Rob2222 [0] (
05:57:21Mouser_XMuch of it has been fixed, but things are still showing up occasionally.
05:57:31Mouser_XMetadata on Buffer.
05:57:43*jpt9 loves Database > Same as current...
05:57:47jpt9it's awesome.
05:58:04Mouser_XI don't use the database.
05:58:15 Quit Bjoern-Erik (Connection reset by peer)
05:58:19jpt9i just like that Rockbox gives you the option to browse or use the database.
05:58:20jpt9or use both.
05:58:26jpt9it's the best of both worlds.
05:58:33Mouser_XI have over 40,000 untagged files on my Gigabeat. It'd make a mess in the database...
05:58:48jpt9can you guys get the wheel on the Sansa to pulsate to the beat? :-) that'd be fun.
06:00:08Mouser_XCheck the tracker. Maybe there's already a patch for it.
06:01:02alienbiker99i think there is.
06:01:35 Quit Rob2222 (Read error: 104 (Connection reset by peer))
06:02:11 Join Rob2222 [0] (
06:05:03 Join Calcipher [0] (
06:05:58 Quit JRoT|Stage (Read error: 110 (Connection timed out))
06:07:48 Join psycho_maniac [0] (
06:08:48jpt9for what?
06:10:36Mouser_Xjpt9: I already *have* the High Voltage SID collection on my Gigabeat. (Remember those 40,000 untagged files? 32,000 of them are SIDs)
06:10:42*Mouser_X is reading logs.
06:10:56jpt9what was the patch that you were talking about for?
06:11:32Mouser_XRead what you said, *immediatly* before what I said.
06:11:45jpt9lemme see...
06:12:48 Join homielowe [0] (
06:12:59 Quit Rob2222 (Read error: 104 (Connection reset by peer))
06:13:11jpt9can't seem to find it...
06:13:23 Join Rob2222 [0] (
06:19:12 Quit Rob2222 (Read error: 104 (Connection reset by peer))
06:19:30 Join dominx [0] (
06:19:49 Join Rob2222 [0] (
06:20:50 Join Bjoern-Erik [0] (
06:21:46 Quit psycho_maniac (" HydraIRC -> <- \o/")
06:25:25 Quit dominx ()
06:25:49 Join ijuz_ [0] (
06:27:03 Join SirFunk [0] (
06:36:30 Join Rob222241 [0] (
06:37:36 Join psycho_maniac [0] (
06:39:42 Quit atsea- (Read error: 104 (Connection reset by peer))
06:41:18 Quit KWhat_Mobile ("Download Gaim:")
06:42:08jpt9i love the little menu thingy that pops up when you hold Menu at the WPS on the Sansa. it's quite spiffy.
06:42:26DogBoysay what
06:42:42 Quit ijuz (Read error: 110 (Connection timed out))
06:43:23psycho_maniacyou mean quick menu?
06:43:32Mouser_XI think he's talking about the quick menu that shows up in the WPS, and allows you to view track info, change track ratings, and that stuff.
06:44:23psycho_maniacoh sorry got them mixed up.
06:44:42psycho_maniacthe "context menu"
06:45:49Mouser_XI haven't read the manual in awhile. I forget what things are called, but I generally know what they *do.*
06:45:53 Join hannesd_ [0] (
06:46:19psycho_maniacactually i figured thats what it was called, but i made the menu come up on my player and it said that on top.
06:46:50 Quit lazka (Remote closed the connection)
06:48:02 Quit Mouser_X ("CGI:IRC (EOF)")
06:50:29 Join Mouser_X [0] (i=cf9bb003@gateway/web/cgi-irc/
06:52:36 Quit Rob2222 (Read error: 110 (Connection timed out))
06:53:05 Quit TMM ("Ex-Chat")
07:00:45 Quit hannesd (Connection timed out)
07:00:45 Nick hannesd_ is now known as hannesd (
07:03:18***Saving seen data "./dancer.seen"
07:07:46 Join J3TC- [0] (
07:12:20 Join Siku [0] (
07:12:26 Join RaRe [0] (
07:14:20 Join lemurgrl [0] (n=chatzill@
07:14:27lemurgrlHi. :D
07:17:14lemurgrlI had a question about where I would find out about 6th gen iPod classic support?
07:17:32psycho_maniacnot from rockbox.
07:18:00lemurgrlI know Apple took measures against installing user created firmware on the 6th gen iPod.
07:18:41lemurgrlI'm coming down from an iAudio X5L which crapped out on me. To my dismay Apple is pretty much the only company still making HDD based DAPS (which aren't PMPS)
07:19:06lemurgrlIs there a comprehensive source where I can read up on the progress of Rockbox on the classic.
07:19:21lemurgrlIf the prospect is hopeful or grim?
07:19:54psycho_maniacthere is no progress.
07:19:55lostlogiclemurgrl: nobody's started porting to those devices
07:19:56krazykitlemurgrl, grim. here's what's happened so far: pretty much nothing.
07:20:09lostlogicaren't we a bunch of gloom and doomers
07:20:26lostlogicit's not remotely impossible though, you should start the port
07:20:29lemurgrlMakes me wish for a follow up to the X5.
07:20:50lemurgrlI would if I wasn't an idiot when it comes to programming.
07:21:00lemurgrlI'd like to.
07:21:06lostlogiclemurgrl: :)
07:21:06 Join B|utEngel [0] (
07:21:16Mouser_Xlemurgrl: You could buy a Gigabeat. It's a HDD based player.
07:21:34lemurgrlI dunno. :/
07:21:36Mouser_XI've heard the H100 is an excellent choice as well.
07:21:39lemurgrlHow's the battery life?
07:21:43lemurgrlI owned an H300.
07:21:51lostlogicI miss mjy H300 :(
07:21:54lemurgrlMost of the best players are out of production.
07:21:57lostlogic(I broke two of them before getting my ipv)
07:22:00Mouser_XThe battery life on the Gigabeat is fine. Easily 12+ hours.
07:22:08Mouser_XProbably 15+.
07:22:12B|utEngelive been reading the manual to rockbox it dont say how to load music on the ipod after rock box has taken over
07:22:16lemurgrlIt seems most companies have totally given up on the prospect of HDD based players to compete with iPod.
07:22:21Mouser_X(I've never timed it myself.)
07:22:25lemurgrlWhich sucks, especially since the iPod Classic is a buggy POS.
07:22:33scorcheB|utEngel: anyway you wish, really
07:22:39krazykitB|utEngel, because you have a choice. use itunes and be able to play music in the OF too or drag and drop
07:22:45scorcheand chris pohl is a gothic god :)
07:22:58lemurgrlThere's no reason to upgrade to the Gigabeat from my Zen Vision M really if there isn't a bump in capacity or battery life.
07:23:04 Quit Mouser_X ("CGI:IRC (EOF)")
07:23:26B|utEngelumm the ipod gose into disk mode when its plugged into USB cord
07:23:32lemurgrlI can only hope the ZVM 60gb will fetch a great price on ebay once it's been out of production a little while longer.
07:23:44scorcheB|utEngel: as it should
07:23:46B|utEngelthe Apple OS for the ipod is gone
07:23:48lemurgrlUMS and hdd daps are going the way of the dodo :[
07:23:51scorcheno it isnt
07:24:08scorchedisk mode is part of the apple OS too
07:24:12lemurgrlAnd I refuse to pay archos $20 for codec support.
07:24:19 Join Mouser_X [0] (i=cf9bb003@gateway/web/cgi-irc/
07:24:50B|utEngelhow can i get the ipod to not boot rock box but boot the apple Firm ware ?
07:25:03scorchethat is described in the manual
07:25:05psycho_maniacits in the manual.
07:26:02 Quit Mouser_X (Client Quit)
07:26:35lemurgrlWould it be possible to shove the 80gb iPod hdd into the X5?
07:26:45 Join Mouser_X [0] (i=cf9bb003@gateway/web/cgi-irc/
07:26:45B|utEngelwhat about this RECORD option on the main menu in rockbox ?
07:26:56scorcheB|utEngel: again, the manual ;)
07:27:05scorchelemurgrl: check out the wiki page HardDriveReplacement
07:27:10psycho_maniaclemurgrl: look up harddrive replacement on the wiki
07:27:29scorcheme gently pats psycho_maniac :)
07:27:30lemurgrlHow about a comprehensive list of improvements that rockbox makes in the 5th gen iPod?
07:27:37lemurgrlI've always found that info to be really hard to find :[
07:27:45scorchelemurgrl: improvements over what?
07:27:54psycho_maniacits very easy with the SEARCH button
07:27:55lemurgrlThe original firmware.
07:27:59lemurgrlAdded features.
07:28:12scorchewell, we typically dont look at the original firmware much
07:28:15lostlogiclemurgrl: also has some info on the ipod 6 gen, doesn't look good
07:28:17psycho_maniacwho cares about the OF. thats not rockbox.
07:28:20lostlogicsays something about encrypted firmware
07:28:20scorchebut you can check out the wiki page WhyRockbox
07:28:52lemurgrlI wonder why Rockbox wasn't ported to the Zen Vision M
07:29:00lemurgrlDoes the ZVM have the same firmware encryption?
07:29:05psycho_maniacill decrypt it with my sledge hammer.
07:29:22scorchebecause no one has stepped forward with the time, player, and motivation to make it happen fully yet
07:30:21Mouser_XA sledge hammer can decrypt and/or disassemble anything!
07:30:22lemurgrlI figure it it's gonna happen for a player it's gonna be the classic.
07:30:33scorcheit doesnt just happen on its own...
07:30:34lemurgrlIt's so popular... why is beyond me.
07:30:51lemurgrlscorche. Do you go to #ga?
07:31:01scorchei dont know what that is
07:31:03lemurgrlYour nick looks really familiar.
07:31:08lemurgrlOkay nvm.
07:31:11 Quit RaRe` (Read error: 110 (Connection timed out))
07:31:13 Part safetydan
07:32:16 Join ramon8 [0] (
07:32:50lemurgrlThe X5L is going for like $400 on ebay.
07:32:55JdGordonkkurbjun: was your commit to doom/i_system.c a mistake?
07:36:35 Quit psycho_maniac (" HydraIRC -> <- The alternative IRC client")
07:38:08 Join JR0T [0] (
07:38:37 Quit lemurgrl ("Chatzilla 0.9.74 [Firefox]")
07:40:26 Join LinusN [0] (i=linus@gateway/web/cgi-irc/
07:43:28 Quit qweru ("moo")
07:44:06B|utEngelpodzilla Sucks compared to RockBox :D
07:44:24Mouser_XI'd be surprised if it was otherwise...
07:44:51B|utEngeli couldnt do nuthin with Busybox/podzilla
07:45:23B|utEngeli spent like 48 hours trying to get idoom on podzilla
07:45:36 Quit amiconn (Nick collision from services.)
07:45:45 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
07:47:31 Join RaRe` [0] (
07:51:19 Part Mouser_X
07:51:58lostlogiczvm will probably happen... isn't it similar to what linuxstb has been working on?
07:56:18 Quit jhulst ("Konversation terminated!")
07:58:52B|utEngeldame 10100 Songs in my itunes :D
07:59:28 Part Llorean
08:01:18amiconnred deltas everywhere, and warnings added but not fixed by sdoyon :(
08:02:37JdGordoneveryone knew we woul get red delats if more voicing went in....
08:04:46amiconnThe fm radio voicing contains redundant code
08:05:05B|utEngelcant that new nano Video Firmware be used on the 5g ?
08:05:20scorchenano video?
08:05:37B|utEngeli want the firmware on the 5g
08:05:49scorchei have no idea what "nano video" is supposed to mean
08:05:59*JdGordon thinks B|utEngel might be lost...
08:06:14 Quit RaRe (Read error: 110 (Connection timed out))
08:06:37B|utEngelnano Video is the new nano it is simply a 5g Ipod compact ....
08:06:51JdGordonoh is it?
08:07:09scorcheare you referring to the 3rd gen nano commonly known as "fatty"?
08:07:09krazykitnot even remotely.
08:07:12 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
08:07:29B|utEngelok what dose it do the 5g dont ?
08:07:41krazykityour question has nothing to do with rockbox.
08:08:14*pondlife notes another dose of red deltas :/
08:08:58*JdGordon confused by the new warning
08:09:12pondlifenew warning?
08:09:36pondlifeAh, unused var
08:09:54JdGordontalk_ids() is defined to blank for hwcodec
08:09:57lostlogic:( red deltas bad. I make pretty green deltas.
08:10:15LinusNJdGordon: i wonder why
08:10:16pondlifeWhat timezone is sdoyon in?
08:11:18JdGordonLinusN: ditto.. no idea
08:11:56pondlifehehe, time for the Archoses to have some red delta ?
08:12:08JdGordonlots of nasy #defines in talk.h
08:12:56pondlifeA lot of them could be made into a common routine to check if talking is possible, IIRC.
08:13:25JdGordontalk_idarray() is compiled for hwcodec.. so i dont kjnow why its not being used?
08:13:48pondlifeWhy are those #defines (talk_ids etc.) in lower case
08:13:58pondlifeMacros are meant to be upper-case, no?
08:14:12pondlifeOr are they functions normally?
08:14:14*JdGordon very confused
08:14:36LinusNi guess they are meant to look like functions for some reason
08:14:51JdGordonOH! right i get it
08:14:52pondlifeWhich is why this is confusing (well, one reason)
08:15:11JdGordonits the hwcodec sims which cant talk...
08:15:18pondlifeWe should build talk.c into the HWCODEC sims regardless
08:15:20amiconnThe radio screen voicing doesn't work on hwcodec atm
08:15:31pondlifeJust for sim-plicity :p
08:15:39JdGordonhaw haw haw!
08:17:00amiconnpondlife: #define'ing functions for targets where they're empty is used in quite a number of places
08:17:04 Quit jpt9 (Read error: 110 (Connection timed out))
08:17:13amiconnIt's better than using an empty function
08:17:28LinusNamiconn: sure, but talk_ids() is a macro on all targets
08:17:52amiconnOh, then it should not look like a function of course
08:18:00 Join OlivierBorowski [0] (
08:18:12LinusNtalk.h contains plenty of fake functions
08:19:04LinusNnot sure why
08:20:15LinusNwe should either convert them into functions, or convert to upper case
08:20:27 Join n17ikh [0] (
08:21:03amiconnThe last 3 are nasty in that they waste binsize. The check for global_settings.talk_menu is compiled in everywhere those "functions" are used"
08:22:16 Join nick_ [0] (
08:22:56 Quit nick_ (Client Quit)
08:23:16pondlifeFunctions would be better than macros, surely.
08:23:48 Quit n17ikh|Lappy (Connection timed out)
08:24:09 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
08:24:44amiconnI think that it would be a good idea to go through apps/, and give all talk_* functions which are called depending on global_settings.talk_menu from more than one place a counterpart that does the check itself
08:25:19amiconntalk.c is apps code and hence allowed to access global_settings
08:26:07JdGordonwarning "fixed"
08:26:14 Join aliask [0] (
08:28:15JdGordonthose undefined talk_ functions should be compiled as stubs on hwcodec sim...
08:28:39pondlifeWhy can't they use the same code as HWCODEC devices?
08:28:43pondlifeIt should come out in the wash
08:28:45 Join GodEater [0] (n=bryan@rockbox/staff/GodEater)
08:29:47*amiconn also wonders what timezone sdoyon is in
08:30:11amiconnIt's always a bad surprise in the morning for me :|
08:31:04JdGordonhe isnt on IrcNicks!
08:31:11JdGordonoh well :p
08:32:07 Join Rob2222 [0] (
08:32:40 Quit nanok (Read error: 113 (No route to host))
08:33:48JdGordonthere are 4 functions which need stubbing (mas related) for the sim to enable the hwcodec sim to at least compile the talk code
08:34:07lostlogicdidn't know timezones were on ircnicks *adds*
08:34:40B|utEngelany one willing to download then DCC me a .bin firmware ? :\
08:34:49B|utEngelrapidshare sucks
08:35:07scorcheB|utEngel: what does this have to do with rockbox?
08:35:23B|utEngelnothing . i just need a favor
08:35:39B|utEngelit hassed to do with ipods tho
08:35:58scorchethis isnt #ipod, this is #rockbox
08:36:06scorcheplease read the channel guidelines ;)
08:36:51B|utEngellol Ok chill ill stfu
08:38:20LinusNStphane Doyon seems to be from canada
08:39:07JdGordonanyone know what a safe value for mp3_get_pos() to always return for the sim would be?
08:39:08 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
08:39:21 Quit ompaul (Read error: 113 (No route to host))
08:45:59JdGordonLinusN: amiconn: <- get hwcodec sim compiling talk (doesnt actually talk though...) commit?
08:47:21JdGordondinner.. back in 20
08:47:25pixelmaJdGordon: you messed up Zagor's name again
08:48:10pondlifePoor Zagor
08:48:27 Join ender` [0] (
08:49:23pondlifeJdGordon: Can you not let most of talk.c be the same on sim, and only #ifdef at a lower level?
08:49:32 Quit BigBambi (Read error: 113 (No route to host))
08:51:11 Quit Rob222241 (Read error: 110 (Connection timed out))
08:51:51 Quit Toki (Read error: 104 (Connection reset by peer))
08:54:40 Join annulus__ [0] (
08:57:01amiconnI think I know now how to add backlight brightness to G5 and keep software pwm fade-in + fade-out
08:58:49krazykitamiconn, a better solution than the patch that some use, where it flips the backlight on and off really fast.
08:58:52amiconnIt's a bit tricky because backlight-nano_video.c needs to track state of 2 pins, and pulse counts on a 3rd
08:59:17amiconnkrazykit: That's a hack. I know now how to control the hardware circuit for backlight brightness
08:59:46amiconnWe could also fade in/out using that circuit, but that wouldn't look smooth, especially at low brightness levels
08:59:49LinusNamiconn: but the resolution is too low for nice fading?
08:59:54amiconn32 levels
08:59:54krazykiti know it's a hack, that's why your solution is better.
08:59:56LinusNah :-)
09:00:18amiconn32 true different levels that is. The lowest level is *not* off
09:00:34amiconnIt's the level that the apple loader uses
09:01:22amiconnI think our bootloader should use that level too, in order to avoid flickering backlight at boot (minimum in apple loader -> maximum in rockbox loader -> user level in rockbox)
09:02:37amiconnLinusN: The (well, your) G5.5 already has experimental backlight adjustment on it (adjusted from 'View HW Info', and the level not surviving reboots)
09:03:22***Saving seen data "./dancer.seen"
09:03:44amiconnAnd it turned out that the Nano has the exact same circuit - but apple doesn't offer brightness adjustment in the Nano OF (!) They only use it for fading
09:03:58 Quit annulus_ (Read error: 110 (Connection timed out))
09:05:46 Quit Siku ()
09:07:07 Join petur [0] (n=petur@rockbox/developer/petur)
09:07:09LinusNamiconn: interesting...
09:09:03 Join Toki [0] (
09:09:38 Join ddalton [0] (
09:09:46ddaltonpondlife, around?
09:11:06ddaltonI read your stuff on 8083 and not sure if it would be accepted.
09:11:14pondlifeWhich bit?
09:11:43ddaltona function talking all this stuff and not shutting up isn't a good idea in my opinion. There has to be a way to shut it up.
09:12:08pondlifeWell if you want to read the WPS, you can choose when to trigger a read
09:12:10ddaltonpondlife: the menu option part...
09:12:20pondlifeThat's just a way to access it.
09:12:35ddaltonyes but I prefer "track info"
09:12:41 Join JRoT|Stage [0] (
09:13:11ddaltonyou want it to just voice all the stuff if the user activates that option? If I have heard enough what do I do?
09:13:11pondlifeBasically, you'd press LONG SELECT to get the context menu, then it would say "Read WPS". You'd press SELECT and it would return to the WPS whilst speaking all of the WPS tags
09:13:29pondlifeHave a simple WPS with just the basics on
09:13:47pondlifeAlternatively, we agree. Just need to get the track info spoken.
09:13:48ddaltonpondlife: that wouldn't shut up for ages. Remember spelling as well. How much of your aduio do you want to be interupted.
09:13:51 Join Zagor [0] (
09:14:07pondlifePersonally, I'd only use the track info option
09:14:17ddaltonpondlife: then commit the id3 patch from fs#6323.
09:14:25ddaltonmaybe it not haven't looked at the code...
09:14:34ddaltonmaybe not
09:14:48pondlifeWould talking support in Track Info do this for you?
09:14:58pondlifeIf so, I'll close 8083
09:15:01ddaltonok ill right a comment there saving me having to write the same thing twice then tell me your thoughts.
09:15:16ddaltonpondlife: don't close it!
09:15:26JdGordonpondlife: ifdeffing it at the lower level is going to be annoying.. so untill somone actually tries getting sound on hwcodec sim i tinhk that patch is good enough
09:15:30ddaltonthere is another part to it which hasn't really been addressed...
09:16:31ddaltonpondlife: give me a minute. Ill try and write something up
09:17:52ddaltonpondlife: ill do the replaying part and if that doesn't get accepted then oh well. And tell them to use track info like we agreed on. Happy with that?
09:19:17pondlifeReplaying part?
09:19:19*ddalton Life will be easier with the commit of p6331
09:19:35pondlifeYou mean, like read the WPS tags?
09:19:39ddaltonpondlife: the repeating last spoken thing part...
09:19:49ddaltonread it
09:19:54ddaltonthe task
09:20:05ddaltonpondlife: no
09:20:11ddaltonpondlife: yes
09:20:29ddaltonlet me see if im missing something
09:20:44pondlifeI thought we agreed that voiced track info would cover this.
09:21:01ddaltonpondlife: yes;
09:21:26ddalton"FS #8083: Voice feature for Accessibility : Speak current item, or WPS infornation on button press"
09:21:34ddalton"speak current item"
09:21:39pondlifeAh, so respeak
09:21:44pondlifethat's a different matter
09:22:00ddaltonpondlife: yes.
09:22:08ddaltonill close it and put it in a new patch then?
09:22:17pondlifeNo, keep that going
09:22:22ddaltonoh ok.
09:22:30pondlifeBut I doubt anything using a button will make it into SVN
09:22:37pondlifeAlthough I quite like the idea
09:22:51pondlifeYou'd need a dedicated "SPEAK" button really
09:22:55pondlifeGlobally available
09:23:13ddaltonpondlife, yes not going to be accepted though
09:23:24pondlifeOK, I'll leave it with you
09:23:30ddaltonwhats the sintax for play +rec?
09:23:35ddaltonin the keymap file
09:23:45pondlifeNo idea
09:23:56*ddalton Looks at 5555
09:26:03 Join barrywardell [0] (n=barrywar@
09:26:12pondlifeRather than trying to update every screen, see if any "repeat talk" function could be built into talk.c. It would need to remember the last spoken ID array (or string or whatever).
09:26:35pondlifeBut we need to avoid bloating the code much more, or all this talking is going to start getting removed ;)
09:27:21ddaltonpondlife: good idea. Got any suggestions on how to do that?
09:27:40pondlifeNo, but bear in mind that Stéphane's red deltas aren't very popular :)
09:30:32 Quit advcomp2019 ("Leaving")
09:30:38 Quit JRoT|Stage (Read error: 104 (Connection reset by peer))
09:35:33barrywardellkkurbjun: you got jtag working on the gigabeat, right?
09:36:05 Join CaptainSquid [0] (
09:36:24JdGordonbarrywardell: eigima has jtag cnnected and working
09:37:43barrywardellJdGordon: on which target?
09:39:59 Join midkay_ [0] (
09:41:25Jpondlife: if, to pick a button, the menu button could be changed (based on a setting) to, on first press speak on second (within [time] seconds) do functionality. Would that be acceptable? (WRT ddaltons patch)
09:41:43J(hope that made sense!)
09:42:05pixelmapondlife: it's not only because of the red deltas for me. It's also because there's is the possiblilty that voice starts getting annoying if it talks too much. I want to hear the music, and by the way, I already enjoy voice more on hwcodec because it talks less...
09:42:06pondlifeProbably not. Both short and long menu are already used
09:42:18pondlifepixelma: I agree
09:42:21Jdouble click?
09:42:33pondlifeUrgh, please no :)
09:42:40J:) just a suggestion :)
09:43:03JdGordonbarrywardell: ha.. sorry... for some reason i read gigabeat as mrobe :p
09:43:30 Quit JR0T (Read error: 110 (Connection timed out))
09:43:37barrywardellI'm actually mostly just interested in jtag in generel, neither for gigabeat nor mrobe
09:44:07pixelmaJdGordon: already fixed Zagor's name? ;)
09:44:07barrywardellI have jtag on my H10 now :)
09:44:08 Join gtkspert [0] (
09:44:32JdGordonpixelma: realaaaaaxx.. you can see its not commited yet
09:44:37*JdGordon goned
09:45:03pixelmaI'm referring to your last commit
09:45:15pondlifeAnd he's referring to his next one I guess :)
09:45:41*pondlife enjoys a scotch egg and goes to work
09:45:58Bagdereditors that changes unrelated lines like that should be taken out and get shot
09:46:09LinusNrepeatedly :-)
09:46:16pondlifeJdGordon: Which editor do you use?
09:46:23*LinusN guesses Kate
09:46:31*Bagder makes the drum roll
09:46:42*LinusN loads his gun
09:46:46pondlifeBagder: You're going to have tired arms
09:46:50ddaltonpondlife: Ill have to look at it later. Got to try and get this linux stuff installed
09:46:56markundoes "goned" mean he's gone?
09:47:06pondlifeIn JdGordonian
09:47:08*Bagder slows down the roll...
09:47:08 Join JR0T [0] (
09:47:27pondlifeThe gun will remain loaded though
09:48:42*Bagder gets tired and stops drumming
09:48:59pondlifePeace at last
09:50:35pondlifeSoap_: Good idea locking the Nano thread. Maybe there should be a sticky summary to hopefully stop a new one springing up?
09:51:05Bagderis there a wiki page summarizing the nano situation?
09:53:58 Quit daurnimator ("Cyas later...")
09:54:46pondlifeHmm, could it be that the Nano is poorly designed, so whilst the CPU runs fine, it produces enough heat to upset the ATA controller or other components? Maybe the OF sleeps more than Rockbox, so no build up occurs?
09:54:56*pondlife waffles in an uninformed manner..
09:54:59Bagdercould be yes
09:55:15Bagderwe do use more power so we could be making it hotter
09:55:18pondlifeIt's quite a tiny thing
09:55:27 Quit midkay (Read error: 110 (Connection timed out))
09:56:33pixelmait's not in Soap's summary but the theory came up yesterday that the Rockbox's power consumtion problem on PP "supports" this problem
09:56:41LinusNi haven't followed the nano situation very closely, but to me it seems to boil down to bad ata timing at high clock frequencies
09:59:16pondlifeHas anyone done the old IR thermometer trick on a Nano?
09:59:26pondlifeComparing OF and Rockbox
10:02:24 Join JRoT|Stage [0] (
10:07:55JdGordonpondlife: yeah, kate
10:08:05JdGordonfor some reason it likes converting files to utf8
10:08:27JdGordoni havnt foudn a better one yet
10:09:21pondlifeJdGordon: Did you see my comment about talk and HWCODEC sims?
10:09:30pondlifei.e. only #ifdef at a lower level
10:10:20JdGordonyeahm you missed my reply i guess...
10:10:37JdGordonthere isnt any point ifdeffing it much lower unless someone tries getting sound in hwcodec sims
10:11:41pondlifeI find #ifdef SIMULATOR annoying if it's not in a hardware code..
10:12:40*JdGordon agress
10:13:13pixelmatry again? :P
10:13:33markunJdGordon: maybe you could file a bug report to kate?
10:13:45 Join cowslayer [0] (
10:14:00JdGordonwhy arnt we using utf8 files anyway though?
10:14:26markunJdGordon: be glad that we don't, otherwise you wouldn't have found this bug to report :)
10:14:49 Part cowslayer
10:14:55pixelmaJdGordon: I wouldn't be surprised if kate would change the encoding to something else then...
10:15:08markunnot sure why we don't use UTF-8, as we need a UTF-8 editor for the language files anyway
10:16:09peturKdevelop has a setting for this, maybe kate too?
10:17:04JdGordonI can manually change the encoding.. but i havnt found a way to autodetect though
10:19:28 Join n1s [0] (
10:23:49 Part pixelma
10:24:06 Quit barrywardell ()
10:27:04*petur kicks memory - KDevelop also just has a setting for the encoding and doesn't autodetect (iirc)
10:27:11 Join DerPapst [0] (
10:27:25DerPapstmorning all :)
10:28:03 Quit JR0T (Read error: 110 (Connection timed out))
10:28:11*GodEater also thinks we should be using UTF-8
10:29:51markunthere is one thing I'm afraid of: that editors start adding and removing the BOM at the start
10:30:17GodEaterBOM ?
10:30:50GodEaterhaha - I just found it before you posted ;)
10:31:32GodEateraccording to that article though, UTF-8 shouldn't need one
10:31:36GodEateronly UTF-16
10:32:05Bagderbom for utf8 doesn't make any sense
10:32:05markunstill some editors add it (for autodetection I guess)
10:32:26Bagderthis byte is in what order? ;-)
10:32:27GodEateroh yes - it says notepad.exe adds it
10:32:33GodEaterhow stupid
10:33:58amiconnBagder: The bom makes *some* sense even for utf-8 - as an indicator that this file is utf-8 encoded
10:34:07markunit's just a marker, like "fLaC" at the start of a flac file
10:34:09DerPapstwho uses notepad.exe anyways?
10:34:13Bagderyeah I guess
10:35:44amiconnLinusN: My suspicion is bad (read: wrong) ata timing, plus our power consumption issue on PP502x
10:35:49GodEateramiconn: it may make sense in that way - but apparently it breaks gcc
10:36:33LinusNamiconn: then we agree
10:36:53amiconnWe need to RE how to set ata timing on PP. Then we can fix the ata sleep on Nano. My suspicion is that the nano flash (and CF in mini, etc) come up in PIO2 from sleep - and we're trying to talk in PIO4 to it...
10:37:17 Part ddalton
10:37:25aliaskOh, what was the outcome for turning off the HDD on PP502x targets? Did it affect the battery life much?
10:37:27amiconnThe Toshiba hdds (and luckily also the microdrives in small H10 and mini) just don't care
10:37:52LinusNaliask: roughly 10% iirc
10:37:55 Quit donutman25 (Read error: 104 (Connection reset by peer))
10:38:16aliaskExcellent! Not quite OF levels yet though?
10:38:18 Join PaulJam [0] (
10:38:26amiconnRoughly 10% on PP5020 with 1.8" hdd, and roughly 20% on PP5022 with 1.8" hdd
10:38:37amiconnNothing on microdrive PP targets
10:41:39amiconnJdGordon: RED
10:41:39 Join spiorf [0] (
10:41:52PaulJamhi, is it possible to make genlang highlight changes that were only done to the voice string too (like in r15507)?
10:42:19JdGordonah bugger
10:42:29JdGordoncourse its not the sim i test!
10:43:22 Join metzelmetzger [0] (
10:49:03JdGordonthat error doesnt even make sensee
10:51:24JdGordonumm... did i break swcodec sims with the last commit?
10:51:55*JdGordon glares at beer bottle.... your making me look bad...!!
10:57:41amiconnIt does make sense
10:58:13amiconnObviously, talk_date() and talk_time() are still emtpy macros on hwcodec sims
10:58:27amiconnHence, you broke the hwcodec sims for targets with rtc
10:58:38JdGordonnothing is obvious after beer....
10:59:10peturbeer should only be allowed for people who can handle it :P
11:01:51JdGordonpixelma: woops.. i owe you an apology.. i didnt realise you were talking about my previous commit...
11:03:26***Saving seen data "./dancer.seen"
11:05:53 Quit metzelmetzger ()
11:07:20 Join metzelmetzger [0] (
11:08:14 Quit XavierGr (Nick collision from services.)
11:08:15 Join XavierGr_ [0] (n=xavier@
11:08:51GodEatertabs in our sourcefiles are a no-no yes ? But what about the Makefiles ?
11:08:56 Join barrywardell [0] (
11:09:06Bagdermakefiles require tabs
11:09:26 Quit DerPapst ("university (away for ~3h30min)")
11:09:28peturheh hey do indeed
11:09:29Bagderalthough gnu make isn't as strict as other makes
11:09:51*petur once had them replaced with spaces.....
11:10:19 Quit crashd (Read error: 101 (Network is unreachable))
11:10:35markunpetur: me too
11:10:36JdGordonlol @ the changelog entry
11:11:01JdGordonyou just cant win
11:11:33GodEaterare there any source files where we're deliberately keeping tabs instead of spaces ? Such as codecs we've grabbed from other projects ?
11:11:37peturaren't we against commits that change more than a thing at the time?
11:12:00BagderI am at least ;-)
11:12:00JdGordonyeah, but im lazy and the 3 were in 3 files
11:16:44 Quit JdGordon ("Konversation terminated!")
11:16:54n1saliask: any progress on the s port?
11:17:06 Join Thundercloud [0] (
11:18:07aliaskn1s: Nope, finished my exams today and I plan on working on it from now
11:18:13 Join JdGordon [0] (
11:18:45n1saliask: great, where are we at? does interrupts work and can we load a firmware file from disk?
11:19:11markunJdGordon: if there are in 3 files it's even easier. I would also prefer 1 change per commit.
11:19:40*JdGordon goned
11:21:25aliaskn1s: Interrupts aren't working, but I think that's because we're still only loading the bootloader and not the main binary
11:22:01 Join crashd [0] (
11:22:19 Quit z35 (Read error: 110 (Connection timed out))
11:22:21aliaskAnd no, the firmware file doesn't work when loaded.
11:22:53 Join z35 [0] (
11:24:38n1saliask: I have been studying data sheets and if the spi interface works might have coded a simple driver for the rtc in the mc13783 :-) haven't tested yet though, not too keen on having to boot windows every time and not yet clear on how to modify and upload new firmware safely
11:27:25aliaskn1s: There's an RTC on the i.mx31, which is better (I can't remember exactly why though), but it seems to be only read only for some reason
11:28:33aliaskThe mc13783 one works well though I think (never tried writing to it, but reading worked fine)
11:28:37n1saliask: I know it's there but not why it would be better, if we use the mc13783 rtc we should be able to wake on alarm even if we power down the cpu
11:29:11aliaskI think it was more accurate, or it was able to store the year or something
11:29:24aliaskThough the alarm function probably outweighs any benefit
11:30:14aliaskShould we move this to #gigabeat?
11:32:35LinusNaliask: sounds like a #rockbox discussion to me
11:35:24 Quit metzelmetzger ()
11:35:38 Quit RaRe` (Read error: 113 (No route to host))
11:36:23 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
11:36:35*amiconn still wonders why there is a separate #gigabeat
11:37:01 Nick fxb__ is now known as fxb (
11:38:35Bagderyeah, keeping a specific port's discussions elsewhere isn't really a very good idea, imo
11:38:45 Quit JdGordon (Read error: 104 (Connection reset by peer))
11:40:25Nico_Pthe logging is nice
11:40:54 Join JdGordon [0] (
11:41:44Bagderthe #gigabeat one?
11:42:35 Quit OlivierBorowski (Remote closed the connection)
11:42:57aliaskI like it because I can just read the logs and see anything gigabeat specific
11:43:07aliaskRather than wade through #rockbox logs
11:43:22BagderI'm interested in all targets, and I'd rather not join 26 IRC channels
11:43:24LinusNi dislike it for the same reason
11:43:40LinusNBagder: amen
11:45:06Bagderand I believe I could add some support and input to people with gigabeats
11:45:15Bagderbut I won't be bothered to join a specific-port channel
11:45:54Bagderand yes I am grumpy ;-)
11:48:12markunBagder: #gigabeat is not only about rockbox but also gigabeat dock mods for example
11:48:32Bagderwell, for non-rockbox stuff I fully get it
11:48:40 Quit Lambuntu (Remote closed the connection)
11:48:45Bagderand I wouldn't want to read about that... :-)
11:49:00markunalthough I don't think it's ever about the OF :)
11:50:15n1saliask: here's the patch
11:51:01n1smost of it just moves code from the sansa rtc driver which is need by this one too (calculating a date from a number of seconds)
11:51:34 Quit Nico_P (Remote closed the connection)
11:51:41n1sI would be interested if any sansa users could try it out and tell me if the rtc still works too
11:52:11n1sand if anyone has a better place for the helper functions to live than in firmware/common/timefuncs.c
11:52:29JdGordonn1s: ill check on my sansa
11:53:29 Join vnl [0] (n=nitin@
11:53:35n1sJdGordon: thanks
11:54:01 Part vnl
11:54:18 Nick fxb is now known as fxb__ (
11:56:48GodEaterI've just done a complete tab cleanup of all *.c|*.h files in the rockbox code tree with this :
11:56:50JdGordonn1s: looks fine
11:57:01n1sJdGordon: setting date too?
11:57:03GodEateris it worth submitting the result as a patch ? Or would one of you chaps with commit access care to do similar ?
11:57:07JdGordonI had to reset the date though.. but i dont know if it was set correctly before hand
11:59:02n1sJdGordon: thanks, it's at least not horribly broken then :-)
12:02:56barrywardellGodEater: I'd say we probably want to keep the tabs in a lot of them to allow easier syncing of code that's copied from elsewhere (eg. the codecs)
12:03:10 Join DontFret [0] (i=423cb164@gateway/web/cgi-irc/
12:03:31GodEaterbarrywardell: I knew you'd say that :(. Why didn't anyone say that when I asked the question earlier ?
12:05:01DontFretrockbox is pretty amazing, just installed it today, im a little late
12:05:17barrywardellGodEater: there are still quite a few files which have no reason to have tabs in them, but I wouldn't be 100% sure which ones they are
12:05:19 Quit jhMikeS (Nick collision from services.)
12:05:25 Join jhMikeS [0] (n=jethead7@rockbox/developer/jhMikeS)
12:06:01 Quit DontFret (Client Quit)
12:08:23GodEaterthere's quite a lot of files outside the codecs tree (which I could leave alone easily enough) - but do we copy much code from other places to locations outside our codec tree ?
12:08:51n1sok, got the imx31 datasheet to open and it seems the built in rtc fires an interrupt when an alarm occurs so I assume the chip must be in some state of powered on
12:09:52n1swhile the mc13783 datasheet says this "A Time Of Day (TOD) alarm function can be used to turn on the phone and alert the processor."
12:09:55LinusNGodEater: i think all files not in codecs/ and plugins/ are safe to fix
12:10:13 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
12:11:36GodEatercould we fix plugins which are rockbox specific (i.e. not doom, zxbox or rockboy) ?
12:11:47aliaskn1s: Well that raises a good question, when the thing is off, how off is it? I think that the battery is tied to an interrupt to power the unit on, so as soon as you connect the battery (with the switch), it turns the unit on.
12:11:58LinusNGodEater: of course
12:12:07GodEateror would you rather leave everything in plugins alone completely ?
12:12:12LinusNGodEater: show me the list of files
12:12:38aliaskIf that's the case, then the best we can do in terms of turning the unit off completely would be to put it to sleep, as turning it off would just turn it back on
12:12:56GodEaterLinusN: (I've removed codecs already)
12:15:22markunaliask: we also haven't found a way to turn off the S3C2440 (Gigabeat F/X)
12:16:41aliaskmarkun: It also is just "asleep"?
12:17:59LinusNGodEater: revised list:
12:18:14GodEaterLinusN: thanks
12:18:47JdGordonGodEater: can you get your script to fix the keywords on all the files also?
12:20:41GodEaterwhat keywords would those be ?
12:20:52GodEaterLinusN: I have a patch now - flyspray ?
12:20:58markunaliask: yes, consumes about 0.78mA while sleeping
12:21:02LinusNGodEater: sure
12:21:20linuxstbGodEater: Regarding tabs, ipod_fw.c is just a copy of the IPL make_fw.c, so maybe tabs should stay there.
12:21:37LinusNGodEater: we will be eternally hated by the keepers of other patches :-)
12:21:50amiconnLinusN: sh2d is imported, isn't it?
12:22:03LinusNamiconn: i didn't know
12:22:04markunaliask: and if we turn off RAM we could probably lower that quite a bit more (right, amiconn?)
12:22:18Soappondlife, That is a good point (adding a sticky about the Nano thread since locking) I didn't think about. I was running off the assumption that now all the relevant information is in the first and last pages (the only two pages people are appearing to read), but I didn't consider the fact now that the thread is locked it will slip further and further down the page until it isn't seen by a casual viewer (the ones I was targeting) at all.
12:22:40LinusNamiconn: we have imported plenty of files in firmware/include too, but i think they are fairly stable and easy to sync manually
12:22:54GodEaterlinuxstb: oops - too late =/
12:23:11amiconnHmm. sh2d also saw several fixes by us...
12:23:38amiconnmarkun: If it's ~1mA or less, the RAM is already sleeping
12:23:50Soappixelma (for the logs) - I didn't mention/quote the discussion on Rockbox's power consumption on PP "supporting" the problem because I felt it was opening as many questions as it answered. Perhaps I should add that as well?
12:24:25 Join webguest78 [0] (i=522e52e0@gateway/web/cgi-irc/
12:25:49SoapThough, pondlife / pixelma, the front page is arguably a better place for a "notice" on the issues some Nanos are experiencing than the forums perhaps?
12:26:03*linuxstb isn't sure he likes these cosmetic patches - it breaks "svn annotate"
12:26:36GodEaterlinuxstb: surely though it only breaks it once
12:26:41linuxstbThat's enough...
12:27:30GodEateryou seem to do enough tab policing yourself however...
12:27:56linuxstbTrue, which is why I'm starting to doubt it's usefulness.
12:28:44GodEaterwell someone seems to have done it on some of the files I'm trying to keep a patch sync'd against
12:28:52GodEaterso I just thought I'd try to get the whole tree in one hit
12:30:01aliaskn1s: I haven't tested your patch yet, but it seems that the spi code is unreliable. I'll have to look into it tomorrow.
12:30:37GodEateranyway - unless someone chooses to commit that patch, then you won't have to worry will you ?
12:30:54linuxstbGodEater: I know it's been done a lot in the past, I'm just asking everyone else if they still consider it beneficial.
12:31:11GodEaterand the sound of the silent response is deafening ;)
12:31:32 Quit aliask ("sleep")
12:31:59amiconnImho it is.
12:33:48GodEaterit seemed at least as if some of the tabs in the files weren't even leading ones - they appeared in some very odd places
12:34:03linuxstbI know ffmpeg have hooked into their SVN a check for tabs and trailing whitespace, and reject commits containing them. It could be helpful if we do something similar, to prevent the need for post-commit fixes.
12:34:16GodEaterI was thinking similarly
12:34:25GodEateror even get it to silently change them
12:34:37linuxstbPresumably we could make file-by-file exceptions.
12:34:40amiconnHmm, but we can't do that in general
12:35:49 Join MajorC [0] (
12:36:21GodEaternot unless someone's volunteering to maintain the list of exceptions
12:36:29GodEater<−− isn't volunteering
12:37:49amiconnWell, everything in subdirs under apps/codecs/ needs to be excluded, and all files named 'Makefile'
12:37:55*pondlife thinks manual code policing is fine
12:39:51linuxstbGodEater: That's easy - if you attempt to commit a file with tabs, and it's rejected, then that's the time to add the file to the list of exceptions. Hopefully the exceptions file would be in SVN. It's a very small percentage of Rockbox source files.
12:40:49linuxstbAnd I disagree about apps/codecs - a lot of files are Rockbox specific, others come from ffmpeg (same no-tabs rule as us).
12:42:24GodEater <−− codecs with tabs in
12:43:27GodEaterfeel free to edit and I'll submit another diff with ones you think shouldbe tidied too
12:43:30linuxstbOK, the ffmpeg-derived files shouldn't - i.e. libffmpegFLAC and libwma, and I don't think adx.c or sid.c should either (IIUC they're Rockbox-specific codecs).
12:45:00*GodEater has no idea what adx or sid do
12:46:40GodEateramiconn: I only did the policing for files named *.c or *.h - so that skipped all Makefiles
12:47:11linuxstbsid.c is the SID codec. Looking at it, the TABs look accidental, so I think they can go. adx.c is the ADX codec written for Rockbox by hcs, and tabs again look accidental.
12:47:17GodEaterI'm not sure of the rules on .S files either - so I left them alone
12:48:45linuxstbGodEater: BTW, how did you remove the tabs in your patch? Looking at adx.c as an example, some will need to be replaced by 8 spaces, some by 4...
12:49:01GodEaterI used tab2space
12:50:20GodEaterI made the assumption that people who were accidentally using spaces would at least have their tabstops set to 4 chars.
12:50:26GodEaterperhaps this is in error however
12:50:55linuxstbYes, I think it will be optimistic to think it can be done automatically.
12:51:30linuxstbYou could test with adx.c and see how it looks...
12:51:35GodEaterI already did
12:51:40GodEatermostly it looks fine
12:52:21GodEateractually - I lie - I've not run it on adx yet
12:52:25GodEaterit looks dreadful already
12:52:31amiconnGodEater: .S files should be fixed too
12:53:09amiconnlinuxstb: The files directly in apps/codecs/ should be fixed too imho, but not files in subdirs of that
12:53:26*pondlife gets the daily mention of #7738 in...
12:53:45GodEaterrunning tab2space on it actually imroves it
12:54:09pondlifeGodEater: What's holding #7738 up? Removal of the setting to a fixed value?
12:54:47linuxstbamiconn: Not even the ffmpeg-derived ones?
12:55:37GodEaterpondlife: no idea
12:55:54GodEaterpondlife: that's why I just sync'd it again
12:56:10pondlifeAh, ok.. I thought you (and jhMikeS) had some reason..
12:56:34GodEaterpondlife: I'm in no position to say
12:56:49pondlifeYou at least have an iPod...
12:56:50linuxstbpondlife: Do you think there needs to be a setting?
12:56:58GodEater <−− list of files including .S
12:57:10pondlifelinuxstb: No big opinion, but I'd hope not
12:57:17pondlifeAssuming the OF doesn't need one
12:57:50pondlife I don't even have a scrollwheel... but that patch seems to have been hanging around too long.
12:58:16linuxstbI think that's the general view. So the setting needs to be removed, and we need to ensure a sensible value for the configurable parts of the patch for each ipod.
12:59:01GodEaterand currently we've only had feedback (that I know of) from a video owner (me), and a nano owner (Llorean) - both of whom agree that the current defaults appear "sensible"
12:59:29linuxstbI guess I should test then ;) And also a mini2g owner (amiconn?)
13:02:05 Quit CaptainSquid (Remote closed the connection)
13:03:28***Saving seen data "./dancer.seen"
13:07:48preglowi think it's fine for nano too
13:07:53preglowand yes, it really should be commited now
13:08:02preglowbut the settings need to go and fine-tuning commence
13:10:33 Join mrkiko [0] (
13:11:46 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
13:12:48amiconnpondlife, GodEater: Does it work for mini g1, and 1st..3rd gen now?
13:12:53 Quit spiorf (Remote closed the connection)
13:13:25pondlifeSorry, I don't know.
13:13:37pondlifeMaybe you could test?
13:14:06preglowlinuxstb: also, where are we with removing the hw eq?
13:15:31amiconnsafetydan needs to fix his patch
13:20:51preglowwouldn't the bug be easier to fix by someone with an actual g5?
13:22:10LinusNwhat bug?
13:22:11amiconnIt's not a functional bug
13:22:29preglowthere is a functional bug, afaik
13:22:48amiconnThe patch contains conflict markers in the english.lang changes, and one change doesn't apply anymore
13:23:52preglowman, flyspray's search sucks
13:23:53GodEaterthere's nothing in the patch that relates to the mini1g or 1st -> 3rd gen as far as I can see
13:24:28 Join jpt9 [0] (
13:24:36 Quit Toxicity999 (Remote closed the connection)
13:24:36 Quit jpt9 (Client Quit)
13:25:10preglowamiconn: i can't see any conflict markers
13:25:26GodEaternor me - the one I sync'd applies cleanly here
13:25:38preglowthe previous patch had them, but that was updated quite some days agio
13:25:42amiconnGodEater: Those targets have a different button driver, and don't use button-clickwheel.c
13:26:35preglowa week, actually
13:27:44GodEaterpreglow: oops - have I sync'd an old version then ?
13:28:30preglowGodEater: i don't think we're talking about the same thing
13:28:38preglowGodEater: i'm talking about the remove-hw-equalizer patch
13:29:37GodEaterah ok
13:36:23GodEateramiconn: are you saying you're opposed to 7738 being commited until it supports all ipods ?
13:41:46preglowit would be a nice bonus, but unless it's going to happen soon, i say we commit it anyway
13:44:16 Quit Seed (Nick collision from services.)
13:44:19 Join Seedy [0] (
13:45:40linuxstbpreglow: Yes, fixing that bug is on my to-do list. Although I think what's needed is a rewrite of the driver so it's using shadow registers, which would also fix help the broken-playback-after-recording issue (at least, if I understood jhMikeS correctly).
13:46:18 Join CaptainSquid [0] (
13:47:05GodEaterI'm not sure we can realistically expect Bushel to get 7738 working on a bunch of devices he doesn't own =/
13:47:13preglowhell no
13:47:17GodEaterso I don't think that sort of acceleration is likely anytime soon
13:49:19preglowis there any reason 1-3g can't use the same code with a few mods?
13:49:30preglowdo all 1-3g models use a physical kind of wheel?
13:51:30 Quit J3TC- (Read error: 113 (No route to host))
13:51:54GodEaterpreglow: the 3g is a touch wheel
13:52:03GodEaterI think the 1st and 2nd gens use a physical one
13:52:07GodEaternot sure on the mini1g
13:52:59 Join RaRe [0] (
13:54:41preglowthis is weird, the crossfeed settings lists are reversed on nano
13:55:04 Join J3TC- [0] (
13:55:22preglowa lot of them are
13:55:24preglowvolume, balance
13:55:34 Join kugel [0] (i=kugel@unaffiliated/kugel)
13:55:41preglowstereo width..
13:56:02LinusNpreglow: have you seen #8106?
13:56:19preglowLinusN: yes, this is what made me notice at all
13:56:33JdGordonpreglow: clockwise scroll shoud increase
13:57:22preglowi don't really think of it in terms of "increase" or "decrease", i think of it in terms of "up" and "down"
13:57:26JdGordonwould anyone object to a patch which made pressing a button while a splash is displayed which is blocking force it to close early?
13:57:31preglowand i doubt other people do any differently
13:57:43preglowif we stop using lists, it'll be different
13:58:06n1sJdGordon: where would it be useful?
13:58:45JdGordonit drives me nuts when a splash comes up like the "nothing to resume" and just sits there...
13:59:16preglowthen people might miss those splashes if they push buttons too quickly
13:59:18preglowi say no
14:00:04linuxstbJdGordon: What if you press the button just after the splash disappears by itself?
14:00:17n1sJdGordon: maybe shorten the delay of the splash?
14:01:12 Quit mrkiko ("leaving")
14:05:45JdGordonsomething like
14:06:02JdGordonit wont exit unless the buttong going in was released...
14:06:10JdGordonand pressing late shouldnt be a problem
14:07:33JdGordonexcept it should be _OK instead of _CANCEL
14:08:01JdGordonpressing too early seems to be no problem either
14:08:08JdGordoncommon... pretty please??? :p
14:09:33JdGordonexcept it doesnt yield now though which might be a problem
14:09:35n1sJdGordon: I think the behaviour is kind of strange with splashes disappearing without user action anyway, how 'bout making splashes stay until a key is pressed?
14:09:54JdGordonno, thats not very nice
14:10:16JdGordoni dont tihnk thats nice anway ;p
14:10:47n1sI would think that behaviour is expected and it would allow enough time for reading the splash and no waiting time...
14:11:56n1sof course it wouldn't work for all splashes
14:12:01JdGordonalthough... I guess it does actually make sense to wait...
14:12:15JdGordonthe splashes which dont block would stay non blocking
14:12:29JdGordonanyone else want to comment?
14:12:41*JdGordon doesnt think waiting indefinatly would be good for voiceing
14:13:21preglowi'm just kind of opposed to splashes being easily aborted by accident
14:13:26preglowthe whole point is making the user notice them
14:13:34JdGordonits not as easy as it sounds
14:14:11JdGordonactually... it should be _CANCEL ... makes it even harder to press accidently
14:14:14preglowstuff like "nothing to resume" is quite ok to abort, but there are more important splashes
14:14:33preglowJdGordon: i guess that makes sense
14:15:04linuxstbJdGordon: What do you mean by "and pressing late shouldnt be a problem" ? If the user presses it late, Rockbox will do something else with the keypress...
14:15:22n1sright voice wouldn't like indefinate waiting
14:16:05JdGordonlinuxstb: because the splashes are up for a few seconds.. you have a very small window between it exiing and the other scrww redrwing when you could accidently press a button
14:16:35JdGordonits not like you could see it come up and then try to pres but its gone already
14:16:48JdGordon... except on the splashes which dont block
14:16:55amiconnSplashes without timeout should stay non-blocking. Everything else would be nasty...
14:17:52*JdGordon doesnt see how adding this makes them more nasty
14:18:21amiconnGodEater, preglow: 4th gen and higher have the wheel hooked up to the opto controller. 1st..3rd gen and mini 1st gen have it on gpio ports
14:19:02preglowamiconn: perhaps 1-2g would be better served using the sansa wheel code directly? both are physical rotary wheels
14:19:03JdGordonlinuxstb: its impossible to press too late... unless you start pressing when your already past get_action() in the other screen which means ts redrawn already... the context change means it will ignore imput until a release
14:19:17amiconnOnly the 1st gen wheel is a mechanical wheel
14:19:34 Join Daniel_S [0] (i=57b0dd24@gateway/web/cgi-irc/
14:19:39amiconn2nd gen and up are touch wheels, but the hookup is the same (except the wheel power switching)
14:20:10linuxstbJdGordon: I'm talking about 1) Brain tells finger to press button; 2) Rockbox removes the splash; 3) finger presses button. I agree it's a small period of time though.
14:20:36amiconnI'd hate blocking splashes because I love to be able to press buttons in advance
14:20:39JdGordonI highly doubt anyone could press it in that time period...
14:21:01 Join webguest27 [0] (i=c05b9323@gateway/web/cgi-irc/
14:21:01JdGordonamiconn: im not suggesting adding blocking to any currently non blocking splashes
14:21:04 Quit webguest27 (Client Quit)
14:21:13linuxstbWhat about leaving the button on the queue, but making it remove the splash?
14:21:45GodEateramiconn: not sure why you're telling me this though...
14:22:05*linuxstb doesn't see blocking splashes as a problem anyway, he's never in that much of a hurry...
14:22:09JdGordonlinuxstb: then that would trigeger the action in the previous screen?
14:22:37linuxstbWouldn't it trigger the action in the next screen?
14:22:53JdGordonthe one that called the splash..
14:22:57linuxstbOr rather, whatever screen next checks for actions after the splash is removed...
14:23:31JdGordonwithout giving apps/ access tothe button queue thats not possible
14:23:57preglowand let's keep this simple...
14:24:19preglowi really don't see this as much of a problem, i'm fine with the way it is
14:24:22preglowi almost never see splashes
14:24:44linuxstbDoesn't apps/ already have access to the button queue?
14:24:53JdGordonnot dirctly
14:25:14linuxstbbutton_queue_count() ?
14:25:46linuxstbpreglow: I agree, it doesn't annoy me at all.
14:26:00JdGordoncould use that...but that makes accidently missing the splash waay to easy
14:26:28JdGordonthe _REL event will be in the queue for most tagrtes whent he splash is shown
14:26:33 Join ilgufo [0] (
14:27:48 Nick fxb__ is now known as fxb (
14:28:17 Join PrintHorizon [0] (
14:28:57n1sinteresting, coldfire v2 pipeline description was committed to gcc mainline (4.3) should help avoiding stalls :-)
14:29:23 Join webguest06 [0] (i=42d27746@gateway/web/cgi-irc/
14:29:50 Join daurnimator [0] (i=daurn@unaffiliated/daurnimator)
14:29:52PrintHorizonHey, I'm trying to get the "Album Artist" tag in my MP3s to work for the database feature of Rockbox. foobar2000 uses this tag by default, but Rockbox doesn't recognize it. Anyone know what exactly the tag should be called for Rockbox to recognize it?
14:29:59preglowtoo bad coldfirev2 doesn't have much of a pipeline
14:30:53PrintHorizonI'm thinking fb2k simply formats the title a certain way and Rockbox wants the title a different way. I'm just not sure how to see. :)
14:31:03n1spreglow: I am fairly certain it didn't know about move.l stalls at all before
14:31:06preglown1s: where do you find this out, btw?
14:32:14webguest06For those contributors present - and those other places - THANKS!
14:32:32 Nick webguest06 is now known as sawyer_HP140 (i=42d27746@gateway/web/cgi-irc/
14:33:38preglown1s: i guess we should try to go 4.3, then
14:34:23preglowwould rock if we had someone involved with gcc onboard, i'm pretty sure we're one of the larger bases of code running on coldfire publically available
14:34:41n1spreglow: I intend to test with the next weekly snapshot (released on friday nighs) to see if it got nicer
14:35:30n1spreglow: you mean someone that would code up some emac support for gcc :-P
14:36:05PrintHorizonAh, from the Wiki, it appears that the tag Rockbox wants from MP3s for "Album Artist" is "BAND"
14:36:37preglown1s: nah, i've given up on that being possible to do in a nice way, but there are plenty of other things that could be nicer
14:37:26 Join roolku [0] (
14:39:23roolkuPrintHorizon: rockbox uses the id3v2 tags TPE2 and TP2 for album artist
14:40:17PrintHorizonThe "Album Artist" section of the database seems to read from the "ARTIST" tag by default
14:40:22 Quit sawyer_HP140 ("CGI:IRC")
14:40:22roolkuPrintHorizon: I don't know what foobar translates that into :(
14:40:32PrintHorizonWhich is fine because all I need to apply an album artist to various compilations
14:40:39PrintHorizonAnd one semi-split EP
14:41:11roolkuPrintHorizon: if it doesn't find TPE2/TP2, it will use TPE1/TP1 (artist)
14:42:15roolku(but only for the database - the WPS will be empty for album artist)
14:42:29 Quit CaptainSquid ("Miranda IM!")
14:42:35PrintHorizonThanks. Well, I saw in the wiki at that the albumartist tag is read from "BAND" for MP3s. I applied that to a set of files and it seems to work great.
14:42:58PrintHorizonSo I guess TPE2/TP2 = BAND
14:43:17roolkuTPE2/TP2 might well be called BAND
14:43:40PrintHorizonOkay, this is cool. :) Thanks.
14:44:59PrintHorizonI like when I have a few releases by a band, and their compilation contributions are grouped in with the rest of their releases... but I dislike having a bunch of artists with one or two tracks populating the database. So browsing by Artist and Album Artist is like having the best of both worlds...
14:45:38 Quit PrintHorizon ()
14:46:59 Quit daurnimator ("Cyas later...")
14:48:02 Nick fxb is now known as fxb__ (
14:59:43 Join XavierGr [0] (
15:02:59kkurbjunbarrywardell: yes, I have a JTAG connection on one of my gigabeat's
15:03:12kkurbjunI used a Xilinx Parallel III jtag cable
15:03:16 Join Siku [0] (
15:03:32***Saving seen data "./dancer.seen"
15:03:56kkurbjunyou should also be able to use a JTAG3 from digilent - I havn't tried it, but I hear they are compatible
15:04:25kkurbjunJdGordon: the commit to i_system.c was intended
15:05:00kkurbjunthe gigabeat has a user timer now so I was removing an unnecessary condition
15:06:14kkurbjunJdGordon: have you tried a battery runtime test now with the backlight work that eigma did?
15:06:58JdGordono point really...
15:07:16kkurbjunI would think that and keeping the core from spinning when it needs to sleep would have the largest effect
15:07:23kkurbjunyeah, i suppose
15:07:26JdGordoni did the firs one so we could see the battery curve
15:07:38 Join sawyer_iHP_140 [0] (i=42d27746@gateway/web/cgi-irc/
15:07:50JdGordonexept I dont think ive done anything with the resut yet
15:08:05kkurbjunok, no worries
15:09:11 Quit Thundercloud (Remote closed the connection)
15:09:15kkurbjunbarrywardell: openocd is great for arm debugging too by the way - I've never tried it on a portal player chipset so I don't know what the chain looks like
15:11:12*preglow starts chanting "sansa dev board"
15:12:08 Quit XavierGr_ (Read error: 110 (Connection timed out))
15:12:28sawyer_iHP_140back again
15:12:31sawyer_iHP_140I have and iHP-140, running rockbox r15492-071106, and its locking up on various files that play fine with the 1.66 ver of the iriver ware. It is happening on various types - wav, mp3 and also doesn't matter if its plugged in or not.
15:13:16preglowhow locking up?
15:13:48sawyer_iHP_140sorry to ask such a question but after looking through the docs I'm not finding a direction.
15:14:06sawyer_iHP_140It shows the song title, and the song length
15:14:24sawyer_iHP_140but no sound, and it shows sometimes the play or pause icon on the bottom
15:14:45GodEaterMoB issues ?
15:14:56sawyer_iHP_140only way to get out is to press reset
15:15:09sawyer_iHP_140MoB = motherboard>
15:15:17preglowno, metadata on buffer
15:15:19preglowa recent addition of ours
15:15:23preglowpertaining to playback
15:15:37preglowsawyer_iHP_140: could you supply one of these files?
15:15:47sawyer_iHP_140sure, how
15:15:57preglowany way convenient for you
15:17:20sawyer_iHP_140hehe, I did it again. It seems to happen with alot of the files I have.
15:17:36LinusNsawyer_iHP_140: is it consistent?
15:17:38sawyer_iHP_140I think I could email one of the files.
15:18:10LinusNsawyer_iHP_140: does it happen if you put one of the problem files in a separate directory and play it there?
15:18:12sawyer_iHP_140yes - it is consistent for me. Do you think since it is so many of the files I have its an issue with my player and not a rockbox issue>
15:18:34sawyer_iHP_140let me tru to put it in a different dir.
15:18:50LinusNthatks, that helps us isolate the problem
15:20:07preglowif it always locks up in the same place for a given file, then having that file would be interesting
15:21:04 Quit MajorC (Read error: 104 (Connection reset by peer))
15:21:12sawyer_iHP_140the thing is - it locks up before I get any sound
15:21:45preglowdoesn't matter, as long as it does so consistently
15:22:22sawyer_iHP_140hmm, you guys are smart! The same file that locked up every time now plays after copying to a new directory. What gives? =]
15:22:41preglowindeed, what gives...
15:22:47preglowsounds like mob bugs, then
15:23:11pondlifeSounds like the playlist is picking up a later file in the same folder and MoB is choking on that file...
15:23:52sawyer_iHP_140this is the first time for me using rockbox, although I did read the faqs and manual
15:23:53pondlifesawyer_iHP_140: Could you try setting up a new folder with ALL of the same files in and see it that gives a problem?
15:24:03pondlifei.e a folder copy
15:24:06sawyer_iHP_140all the music was on prior to rockbox install
15:24:12 Quit crashd (Remote closed the connection)
15:24:16 Join crashd [0] (
15:24:17 Join daurnimator [0] (n=fake@unaffiliated/daurnimator)
15:24:42pondlifeThen, try deleting files from the end of that folder one at a time until the problem goes away...
15:24:42sawyer_iHP_140although I'm jammin to live custard pie right now!
15:24:52pondlifeMmm.. jam...
15:25:20sawyer_iHP_140BTW - I'm not using a playlist just going into a dir and selecting files to play
15:25:28sawyer_iHP_140K hang on a sec
15:27:01sawyer_iHP_140err a couple minutes
15:27:08sawyer_iHP_140it is a big directory
15:27:37sawyer_iHP_140so, the solution is to re-copy all the files? No big deal to me if thats the fix.
15:28:20sawyer_iHP_140yeah, jam good
15:29:08sawyer_iHP_140noodles bad !! =]
15:30:46 Join mf0102 [0] (n=michi@
15:31:12 Quit ilgufo ("So Long, and Thanks For All the Fish -")
15:31:16pondlifesawyer_iHP_140: No, the theory is that one of those files is causing the problem.
15:31:35pondlifeBy making a copy and then deleting it file-by-file, we should be able to work out which one.
15:32:32pondlifeAlternatively, make sure repeat is disabled, then start playback on the last file and see if that works. If so, stop and restart playback on the previous file. Repeat until crash.
15:35:30sawyer_iHP_140OK, but what has been happening is it locks on the first file I play; is that affected by other files in that dir? Also to add trouble, I'm having some CRC errors during copy so it seems I need to attend to the HD status before I suspect anything else...
15:35:38sawyer_iHP_140sorry for any trouble
15:35:49pondlifeAh, that's probably the issue then
15:35:56pondlifeCHKDSK /F /R
15:36:02pondlifeOr fsck
15:36:21sawyer_iHP_140or toss it in the bin and replace with an 80GB =]
15:36:36pondlifeIf you can find one.
15:36:52 Quit sawyer_iHP_140 ("CGI:IRC (EOF)")
15:37:04 Join TMM [0] (n=hp@
15:37:09pondlifeI'd love an 80GB upgrade, but I can only find them second-hand for a silly price.
15:37:10preglowwell, if of was able to play them...
15:37:16pondlifeHmm, good point
15:37:21 Join sawyer_iHP140 [0] (i=42d27746@gateway/web/cgi-irc/
15:37:33sawyer_iHP140ack closed the wrong window
15:37:55preglowsawyer_iHP140: and you're sure the original firmware can play those files fine?
15:37:57pondlifesawyer_iHP140: Do you know that all files in that folder play ok in the original firmware?
15:38:41pondlifeBear in mind that we buffer the following files after the one that's playing... so a read error in one of the later ones might kill it.
15:38:56pondlifeAlthough a nice error, rather than a lockup would be good
15:40:46sawyer_iHP140yes, they play OK in the original firmware, although once in a rare while (with orig firmware) my player will either play a snippet of a different song or skip to the next song in the middle of a file.
15:41:02sawyer_iHP140hmm seems more and more I have hardware issue
15:41:59sawyer_iHP140damn chkdsk is going to take awile
15:43:01 Quit JdGordon (Remote closed the connection)
15:43:50XavierGrsawyer_iHP140: be sure to plug it on the AC
15:43:59sawyer_iHP140you guys are great to point me in the right direction - I think I will call it a night and let chkdsk run (I'm in China so its 10PM!) and hit you all up another day. Great project!
15:44:03XavierGrchksdsk can take ages in those disks
15:44:05sawyer_iHP140Yep, plugged in
15:44:20sawyer_iHP140through the 240 to 110 converter =]
15:45:26sawyer_iHP140electricity is scary over here
15:49:53 Join japc [0] (n=japc@
15:54:30sawyer_iHP140yep - looking at the chkdsk output the hd is definitely in question.
15:58:06barrywardellkkurbjun: sorry. I was away
15:58:16barrywardellI already have openocd working with my H10
15:58:35barrywardellbut I was wondering where you got the patch for accessing the flash?
15:58:43kkurbjun:), nice, the patch?
15:58:48kkurbjunoh, on openocd?
15:58:53kkurbjunI made that myself
15:59:20barrywardellah, ok. what would be involved in making one for the H10? I can read the flash at 0x0 already...
15:59:26 Quit sawyer_iHP140 ("CGI:IRC")
15:59:39kkurbjundo you know who makes the flash?
15:59:48kkurbjunis it CFI compatible?
16:00:25barrywardelli have a datasheet:
16:00:51 Quit Zagor ("Client exiting")
16:02:12kkurbjunit looks like it's CFI
16:02:12barrywardellkkurbjun: also, do you have trst hooked up? Is it necessary? I'm using a homemade Xilinx Parallel III which doesn't have it afaik
16:03:19kkurbjunI don't have it connected, I've never tried using gdb with my cable, I think openocd can deal without it, but you can issue a full reset and hold from the begining of the program as far as I know
16:03:40kkurbjunyou should be able to use the builtin CFI driver in openocd
16:04:07 Part LinusN
16:04:10kkurbjunsee if that works, if it doesn't you might need to add another device to allow it to support your flash
16:04:34barrywardellwhat does the patch in the wiki do? Is it necessary?
16:04:59kkurbjunthe flash writer code I have should also be able to help you - the flash on the gigabeat isn't fully cfi compatible, but it does have most of the full spansion cfi commands
16:05:17 Join scorche|w [0] (n=42c007b2@rockbox/administrator/scorche)
16:05:20kkurbjunso theres some extra info you have to give to openocd so that it knows how to deal with the flahs
16:05:30kkurbjunI have to go to work though
16:05:33kkurbjunI can talk later
16:05:59barrywardellok, thanks for the info
16:06:12kkurbjundon't use the gigabeat flash code till you have the flash writing through jtag working though
16:06:35kkurbjunyou can always write a script to recover it without support of openocd though too
16:07:46barrywardellDon't worry, I don't plan on doing any flash writing just yet
16:11:31 Quit Echelon (Remote closed the connection)
16:11:34 Quit hcs ("Leaving.")
16:16:40 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
16:21:49 Part Daniel_S
16:25:16 Nick fxb__ is now known as fxb (
16:39:01 Join Crash91 [0] (n=evil91@
16:45:49 Join Bubelgum [0] (
16:49:52 Nick parafin|away is now known as parafin (
16:50:11 Join vnl [0] (n=nitin@
16:50:35 Part vnl
16:50:42 Join jgarvey [0] (
16:57:00preglowmy lord, how much can i stand
16:57:15preglowneither audacity, sweep or sox can pan a stereo file for me without crashing or malfunctioning badly
17:01:29 Join ApotheoZ [0] (
17:02:38 Quit ApotheoZ (Client Quit)
17:03:35***Saving seen data "./dancer.seen"
17:03:56 Join sonofsam [0] (i=4516ca84@gateway/web/cgi-irc/
17:04:55sonofsamguys i have scandisk e200 if i download rockbox will it let me veiw russian text
17:05:34sonofsamHELP: i have scandisk e200 if i download rockbox will it let me veiw russian tex
17:08:22sonofsamHELP: i have scandisk e200 if i download. rockbox will it let me veiw russian text?
17:08:50Crash91sorry, rockbox doesnt support players from "scandisk"
17:08:58sonofsamok thank
17:09:02Crash91but yes, it can
17:09:08Crash91and its sandisk :P
17:09:30sonofsamwait wut?
17:09:59sonofsamdoes support scan disk or not?
17:10:03scorche|wsonofsam: rockbox supports unicode, so it should if you use a unicode font
17:10:19Crash91rockbox, doesnt support players from "scandisk" but it supports players from "sandisk". It can also let you view Russian text, yes
17:10:42Bubelgumthe sandisk e200 is from sandisk
17:11:00sonofsamcrash91 : ow wow i'm an idiot i ment sandisk
17:11:26Bubelgumquite obvious :p
17:11:54Crash91just messing around
17:13:02Crash91i dont quite understand what you meant by "ment"
17:13:05Crash91:P :P
17:14:56 Quit sonofsam ("CGI:IRC (EOF)")
17:15:27 Join sonofsam [0] (i=4516ca84@gateway/web/cgi-irc/
17:16:17sonofsamhey crash91: i'm a noob at this sooo after i download and unzip rockbox for my 60 gb ipod
17:16:32sonofsamhow do i fromat my ipod software
17:17:09Crash91i redirect you to the holy grail of these things
17:17:14kclafI recommand you either read documentation or stick with ipod software
17:17:56Crash91and only do this if you know what you are doing, understand the risks of bricking and if you are really going to use the extra RB features
17:18:34sonofsamyea i like your themes that you have
17:18:55sonofsamand i need it to recoginze russian text
17:18:56scorche|wCrash91: what risks of bricking are those?
17:19:20GodEaterThere is no risk of bricking an ipod. It's not possible by installing Rockbox.
17:19:25Crash91scorche|w: how the hell am i supposed to know :D
17:19:39scorche|wthis is a sansa, but the issue some people have still isnt bricking ;)
17:19:40Crash91ah well, i thought that
17:20:11Crash91wait...he asked if he could use it on his sansa, Then asked howto install on his video
17:20:38*Crash91 wonders why someone would need both
17:20:44scorche|weither way, it isnt bricking ;)
17:21:48sonofsamwell my brother has sansa and i have ipod video
17:22:01GodEaterand some people own more than one target...
17:22:19Crash91GodEater:thats devs
17:22:43GodEaterand me
17:22:45Crash91GodEater: and rich people :P (or non-teenagers, unlike me)
17:22:49GodEaterI own three targets
17:22:54Crash91GodEater: arent you a dev
17:23:01*scorche|w waits for his sansa to come
17:23:19Crash91or more appropriately: _
17:23:21*GodEater didn't know sansa's could get turned on like that
17:23:24scorche|wCrash91: you can identify people by their rockbox cloak
17:23:48*scorche|w thinks that GodEater is missing out on quite a bit
17:24:00GodEaterI ought to buy one obviously
17:24:05*Crash91 wonders whether it is the sex
17:24:13scorche|wwell, you missed the cheap price
17:24:17GodEatercan you upload a video of it to beautifulagony ?
17:24:27GodEaterscorche|w: but woot don't ship here :)
17:24:29*scorche|w moves over to -community
17:25:36 Quit sonofsam ("CGI:IRC (EOF)")
17:26:27 Join OlivierBorowski [0] (
17:28:36 Quit webguest78 ("CGI:IRC")
17:30:28 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
17:30:49 Join miepchen^schlaf [0] (
17:33:08 Quit Xerion (Read error: 110 (Connection timed out))
17:39:19 Quit TMM ("Ex-Chat")
17:42:22 Quit Calcipher ("I-n-v-i-s-i-o-n 2.0 Build 3515 with A Pack Fix By")
17:44:37 Join Frazz [0] (
17:50:24 Join feisar_ [0] (
17:50:24 Quit feisar (Read error: 104 (Connection reset by peer))
17:57:49preglowwe already have the code to do pwm dimming, did anyone check to see if it can be reused for the wheel light at all?
17:58:48GodEaterpreglow: FS8110 ?
17:59:02preglowyes, the one you just closed an reopened...
17:59:12GodEaterlinuxstb shouted at me
17:59:25preglowwell, i agree, it's an honest request
17:59:35GodEaterso did I after I thought about it
17:59:41GodEaterhence I re-opened it
18:00:05linuxstbGodEater: I didn't shout, I just questioned why it was different from regular backlight dimming...
18:01:12GodEaterI imagined it as a shout
18:01:17preglowon that note, we need another fade in/out setting, 250ms
18:01:21preglowi think the current ones are too slow
18:01:21*GodEater has an active imagination
18:01:24Dark_ApostropheMissing start packet code prefix F4EB at 4000, followed by Missing packet start code prefix: 2355 at 0
18:01:34Dark_ApostropheI'm getting that with many videos
18:01:55Dark_ApostropheThe first one appears when I select the videos, then the second when I select "play from beginning"
18:01:56linuxstbGodEater: Not at all - sorry if it came across that way.
18:02:00Dark_ApostropheIs this a known bug or something?
18:02:07preglowalso, ten seconds fade out? come on...
18:02:42Nico_Pheh, slow fade like the sun
18:03:03pondlifeIndeed :)
18:03:11pondlifeYou read the forum too....
18:03:34Nico_Ppondlife: no, that was FS #8110 :) I'd like to see the forum thread
18:03:52pondlifeHmm, just imagine a backlight that takes 24 hours to brighten and dim.
18:04:20Dark_Apostrophepondlife: Imagine the battery lifeime of said MP3 player ;)
18:04:43pondlifeFusion powered, I guess.
18:06:06preglowhrm, seems like anything lower than 500ms requires further changes...
18:06:32Dark_ApostropheCan anyone please help me out?
18:07:35n1sNico_P: I saw in the #gigabeat logs that you downloaded the linux bsp for the imx31, anything interesting in there? I am mostly thinking about the mc13783 as the "datasheet" available is quite lacking...
18:08:34Nico_Pn1s: I'm not really competent enough to understand the stuff in there but if you tell me what to look for I can tell you
18:09:37markunDark_Apostrophe: I have no idea what it is
18:10:08Dark_Apostrophemarkun: I get that with certain videos
18:10:36Dark_Apostrophemarkun: I can send you one of the videos so that you can see if you get the error, as well
18:11:25n1sNico_P: I'm not sure what I'm looking for really, just stuff missing from the datasheet like controlling the voltage to the cpu and brightness/fading of the backlight are mentioned in there but never described...
18:11:58n1sif the mc13783 driver is self-contained maybe you can pastebin it or something?
18:12:09Nico_PI'll search for it
18:16:39linuxstbDark_Apostrophe: How did you encode your videos?
18:17:48linuxstbCan you put an example video online somewhere?
18:18:00Nico_Pn1s: what does the chip do?
18:18:20n1spower management mostly
18:18:47n1sI saw that it is used on the dev board the linux bsp is supposed to be run at
18:19:21Dark_Apostrophelinuxstb: Someone might get pissed, as it's a Simpsons episode
18:19:24n1sbut it also has audio ADC and DAC touchscreen interface and some other stuff
18:19:38Dark_Apostrophelinuxstb: I can DCC you one if you want
18:19:45 Quit barrywardell (Remote closed the connection)
18:20:05n1sNico_P: in the gigabeat is basically just power management/backlight/rtc
18:23:25 Join lee-qid [0] (
18:24:59Nico_Pn1s: would you know what file it is if I gave you the list?
18:25:39 Part pondlife ("Gone")
18:25:55linuxstbn1s: You're talking about the Gigabeat S I assume? Have you started running code on it?
18:26:05n1sNico_P: no, I'm in no rush really and will try to download the image once freescale's site lets me, thanks for looking tho :-)
18:26:09 Join joh1 [0] (
18:26:45n1slinuxstb: not yet, I have been reading datasheets though and the mc13783 datasheet is lacking a lot of details
18:27:00 Join DerPapst [0] (
18:27:11DerPapst'lo all
18:27:16n1slike it mentions it can set brightness and fade backlight leds but not how to make it do that...
18:27:21linuxstbWe _really_ need to find an easier way to run code on the S...
18:27:44Nico_Pn1s: oh I think I found the file
18:27:55n1slinuxstb: agreed, aliask said earlier today he now has more free time and will start looking at the port again
18:29:13*n1s <3 the imx31 reference manual, 2000+ pages of gory details :-)
18:30:50Nico_Pn1s: does mc13783.c sound interesting to you?
18:31:03n1sNico_P: yes, yes indeed :-)
18:33:04Nico_PI took it from one of the patches
18:34:16 Quit Bubelgum (Remote closed the connection)
18:35:17n1sNico_P: thanks
18:35:47Nico_Pn1s: there seems to be a lot more mc13783-related things in that patch... I can DCC it to you if you want
18:36:37n1sNico_P: I'm not sure dcc works for me only tried it once though, but sure go ahead
18:36:49 Join bertrik [0] (
18:37:55 Join pepie34 [0] (
18:38:00 Quit kugel (Read error: 104 (Connection reset by peer))
18:42:48 Quit B|utEngel (Read error: 104 (Connection reset by peer))
18:43:08 Quit petur ("work->home")
18:45:31 Join andrewg877 [0] (
18:49:15n1sNico_P: unrelated: can the buffering code now handle large data structs on the buffer associated with a track (for midi)?
18:50:33 Quit idnar (Nick collision from services.)
18:50:35 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
18:50:46 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
18:51:16lostlogicn1s: define large and probably, would just need to have a type defining the proper behavior for it
18:52:35lostlogicn1s: the problem would be knowing where to put them so that they aren't in the way.
18:53:00Nico_Plostlogic: we should start experimenting with plugins maybe
18:53:24lostlogicNico_P: for handling what to do when we finish buffering a file?
18:53:26n1slostlogic: a midi file would need a set of patches to be available for as long as it's playing, the largest patch I have is 2.2 MB
18:53:29lostlogicbecause that's what came to my mind
18:54:18 Quit pepie34 ("Ex-Chat")
18:54:20lostlogicn1s: what's involved in generating those? To do it with simple modifications only they would need to be available to the buffering system at roughly the same time that the midi is being buffered
18:54:28Nico_Plostlogic: never though of that... I was thinking of giving plugins access to the buffering API
18:54:32lostlogicn1s: ie would get_metadata be able to tell the system what to stick on?
18:54:40Nico_Pand seeing how existing ones could benefit
18:55:14n1slostlogic: they are just read from disk and possibly byteswapped and yes get_metadata should be able to do it
18:55:15Nico_Plostlogic: for example, the midi player is a plugin AFAIK
18:55:26lostlogicn1s: that would work beautifully then.
18:55:44lostlogicI gotta run to work, but this has definite promise
18:56:41n1sNico_P: the plan is to turn the midi player into a codec
18:57:05n1sbut I got to go too now so ttyl
18:57:10linuxstbDo we have a freely distributable set of patches for midi?
18:57:11 Quit n1s ()
18:57:47Nico_Plinuxstb: there is one on the wiki
18:57:59linuxstbI know about that one, but I thought the license was unknown?
18:58:11Nico_Pah then I know nothing more than you do
18:59:24linuxstbAh, a readme for that patchset has been added since I last looked...
18:59:28 Join spiorf [0] (
19:00:03 Join Arathis [0] (
19:00:27linuxstbHmm, but the readme doesn't really help...
19:02:27 Join ilgufo [0] (
19:03:39***Saving seen data "./dancer.seen"
19:08:53 Join Xerion [0] (
19:08:53 Quit andrewg867 (Read error: 110 (Connection timed out))
19:12:40 Join Thundercloud [0] (
19:12:43*preglow has speex at 41kb for coldfire and expects at least 10kb more can go, easily
19:14:46linuxstbSmaller than the Rockbox logo on a lot of targets...
19:15:29preglowi spent quite a lot of time hunting for some rogue floating point code that turned out to be in codecs/speex.c ...
19:15:43preglowa lovely divide and multiply, all in one glorious line
19:15:55 Join jumijoze_ [0] (
19:16:29 Quit jumijoze (Nick collision from services.)
19:16:39 Nick jumijoze_ is now known as jumijoze (
19:18:56lostlogichmm, the only problem I can see wiht the idea of midi as a codec with its samples on the buffer would be if the midi + samples are larger than the buffer. I think since midi files themselves are small, that's an acceptable situ
19:19:32linuxstbThe patchset is over 20MB - so that rules out a few swcodec targets....
19:19:36preglowlostlogic: the midi itself is tiiiiiiny in comparison
19:19:48preglowi don't see that as a problem at all
19:20:35roolkuvery few midi files will use all of the patches (instruments)
19:21:02roolkuso a lot less than 20Mb is required
19:22:49 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
19:23:10 Join miepchen^schlaf [0] (
19:23:45 Join BigBambi [0] (n=alex@rockbox/staff/BigBambi)
19:24:57 Join Llorean [0] (n=llorean@
19:25:06 Quit linuxstb ("Client Exiting")
19:25:16 Quit miepchen^schlaf (Read error: 104 (Connection reset by peer))
19:25:21 Join Exiro [0] (
19:25:30Exiroello there
19:25:37 Join miepchen^schlaf [0] (
19:27:33 Quit kkurbjun (Read error: 113 (No route to host))
19:27:55 Quit spiorf (Remote closed the connection)
19:28:26lostlogicpreglow: I figured... it should work great then... just hafta work out who's in charge of putting the samples into the buffer after get_metadata knows which ones the song needs.
19:28:29ExiroHi, I'm having a problem on the current build for my iPod Nano. I'm hearing some weird Techno-like Noise in the background.
19:29:19 Quit Crash91 ("Bye Bye!")
19:29:23ExiroI can't really explain what kind of sound, but it's definitely not part of the music I listen
19:36:04 Quit jgarvey (Read error: 110 (Connection timed out))
19:36:54 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
19:37:40hcsNico_P: I'm considering reverting to the old ADX codec until the backwards seeking issue can be resolved, any objections?
19:38:34Nico_Phcs: none. I haven't thought about how to do it better yet :/
19:40:13hcsNico_P: I consider it a two-stage process, to fix.
19:40:16hcs1) fix the seeking issue (rebuffer the file in the space that is already allocated for it? throw out what's in the space where the requested stuff would be?
19:40:43hcs2) implement some way to not dump the buffer for files where we know we'll need it
19:41:05hcsThe first to make it work, the second to make it efficient.
19:41:11lostlogichcs: what happens when an adx is > bufsize?
19:42:42hcsthen we've no choice but to dump it
19:42:45hcspriority: current buffer=2, reserved buffer=1, other tracks=0
19:43:12hcsnot that I have any experience with the buffering code to figure out how to implement it
19:43:16*lostlogic hates special handling for one type of audio.
19:43:41lostlogicADX is like the bastard child of the small atomic codecs and a packet codec
19:44:29hcslostlogic: the second stage is optional, I'd just be happy with it working even if it has the reload the file
19:44:58lostlogichcs: yeah, realistically once 8092 is fixed, it should work again, especially on flash players w/o a spinup
19:45:32lostlogicbut then making it work well remains difficult −− needs some kind of flag that says "don't shrink me unless you strictly must"
19:45:39lostlogicwhich actually wouldn't be unreasonable.
19:45:50lostlogicshould that be a different type or a flag?
19:45:54hcslostlogic: any plans in place to fix 8092 or should I start looking into it
19:46:12lostlogichcs: I _think_ Nico's working on it. I'm bitter/jaded/refusing.
19:46:23lostlogic(and yes, I know how childish it is)_
19:46:52hcsI'm just going to go ahead and revert ADX, then, so at least we have something usable for now
19:47:24lostlogichow well does it work? don't you still run into trouble around the buffer wrap?
19:47:38hcsIt works flawlessly, as far as I can tell
19:47:46lostlogichcs: oh, cool −− revert away then :)
19:48:23hcslostlogic: observe the section commented "dance with the devil in the pale moonlight", it's supposed to handle that, but I can't remember exactly why it works
19:48:50lostlogichahaahha, I'll take a look.
19:50:45 Quit Exiro (Read error: 110 (Connection timed out))
19:50:57hcsdrat, I'm on a public wireless network, is there secure access for svn?
19:51:15lostlogicshouldn't it default to ssl?
19:52:04hcsI thought I recalled someone asking about this a while ago and it wasn't possible.
19:52:20 Join Bagder_ [0] (
19:53:01lostlogicask Bagder_
19:54:35 Join Domonoky [0] (
19:55:34hcsBagder_: can we do svn over a secure channel?
19:55:44JRoT|Stagedoes anyone knows if the old keymap for sansa ever comes back?
19:56:11*bertrik gazes crystal ball
19:56:50JRoT|Stagei don't really like the new one, it made everything harder, why did you guys chose for the new one?
19:57:11DomonokyJRoT|Stage: not likely.. for reasons read in the forum.. :-)
19:57:22JRoT|Stagek thnx
19:57:25lostlogicJRoT|Stage: we're actually in the business of making software as hard to use as possible. It's in our charter.
19:57:29lostlogic::shiftyeyes:: <−− liar.
19:57:57JRoT|Stagelostlogic i didn't want to flame
19:58:00JRoT|Stagesorry for that
19:58:11lostlogicJRoT|Stage: no no, no worries, I was just being an ass ;)
19:58:15 Nick JRoT|Stage is now known as JR0T (
19:59:29 Quit Thundercloud (Remote closed the connection)
20:01:36 Join r121 [0] (
20:03:44hcsha, tsocks + ssh -D ftw
20:06:40LloreanJR0T: It's not harder... And it certainly doesn't make EVERYTHING harder since it only affected a few buttons in a few specific screens
20:06:53LloreanPlease, try using it for more than a couple days before judging it. Right now it's new and you haven't gotten used to it.
20:07:32lostlogicJR0T: and if you continue to hate it... you can always build the devkit and compile a patched build with the old keymap pretty easily :)
20:07:50LloreanNot to mention the patch has been in the tracker for months, and I gave a weeks warning both in the forums and the mailing list, so this change shouldn't have taken people interested in "what's going on" (as people updating regularly should be) by surprise.
20:08:21 Quit Bagder (Read error: 110 (Connection timed out))
20:09:39 Join mrkiko [0] (
20:10:01 Quit japc (Read error: 110 (Connection timed out))
20:11:25 Join karashata [0] (
20:12:08mrkikoHi all! When I'm in the track-list with my H340, I can make rockbox go mad just pressing the "next" or "prev" key quickly! Does someone experience this with this or another player?
20:12:41lostlogic"track list"?
20:12:55hcs"go mad"?
20:13:06*lostlogic high fives hcs
20:13:31karashatajust wanna mention to whoever's been playing around with playback stuff that the recent SVN commits broke FLAC playback on my H10
20:13:37 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
20:13:52lostlogickarashata: please be specific
20:14:04 Quit ilgufo ("So Long, and Thanks For All the Fish -")
20:14:07lostlogickarashata: which cmomit broke it and for which flacs (I play flacs pretty frequently without trouble)
20:15:00mrkikohcs: "go mad" = the player will not detect the next key release;
20:15:16karashataI'm not sure which one, the most recent one (r15517) and the one I tried before it that I noticed didn't work (r15513) are the only two builds I've used with FLAC recently
20:15:20mrkikolostlogic: "track list": file browser, the window I get in when selecting "files"
20:15:53lostlogicmrkiko: why would you press next repeatedly in that window (not that I'm justifying it breaking us, but ???)
20:16:26lostlogickarashata: so 15513 worked and 15517 broke, right?
20:16:41karashatano, 15513 didn't work either
20:16:51lostlogickarashata: oh, what build worked?
20:17:05mrkikoThere are 3 tracks, A, B and C. I would press next repeatedly and quickly if I need to skip from A directly to C.
20:17:12karashataI don't remember, let me check the other builds I have backed up and see if I can't find it
20:17:18 Join ilgufo [0] (
20:17:32lostlogicmrkiko: wouldn't... wouldn't you do that from the WPS not the browser??
20:17:34mrkikolostlogic: note that I'm blind, so I can't use the file browser agibly.
20:17:56Domonokymrkiko: so you are using voicefiles ?
20:18:08mrkikoDomonoky: yes
20:18:22*mrkiko needs to say that his braille display is slow in reacting...
20:18:50lostlogicmrkiko: considering that I had no idea you were blind until you said so, I'm rather impressed by your reading and typing speed ATM.
20:19:32preglowi've had this bug myself
20:19:33preglowseveral times
20:19:36preglowrockbox would just hang
20:19:44mrkikolostlogic: thank you. I would do better with a functional braille display. This one is having some problems right now! :)
20:20:13lostlogicpreglow: how do you trigger it exactly?
20:20:23karashataokay, apparently it broke for me before 15493 even...
20:20:32preglowlostlogic: by switching tracks very fast
20:20:34lostlogickarashata: are the flacs > 30m in size?
20:20:40preglowlostlogic: i can't make it happen consistently :/
20:20:45lostlogicpreglow: does voice need to be enabled?
20:20:45karashatano, none of them are
20:20:48preglowlostlogic: nope
20:21:25lostlogickarashata: can you further describe "breaks"?
20:21:27 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
20:21:42karashatathe player completely freezes when trying to play a FLAC file back, no response to anything
20:21:48karashataconstant disc activity
20:21:52mrkikolostlogic: in my case at least it's different from the fs#8003.
20:21:53karashataer, disk
20:21:55Lloreanpreglow, lostlogic, mrkiko: I think you may be talking about two different things?
20:22:07LloreanIt sounds like "mrkiki" is scrolling up/down in a list rapidly
20:22:10 Quit ilgufo (Client Quit)
20:22:11Lloreaner mrkiko
20:22:13 Join ilgufo [0] (
20:22:22lostlogickarashata: at the beginning before even starting to play or smoetmie randomly in the middle?
20:22:24LloreanSince "next" and "previous" in the filetree wouldn't skip tracks, but enter and leave folders.
20:22:32karashataright at the beginning
20:22:35lostlogicLlorean: that's what I was confused about.
20:22:37karashatait doesn't play the file at all
20:22:58karashataloads the metadata fine though for it and the next track on 15493
20:22:59mrkikofs#8003 (still verifying) will crash my player; this problem instead causes the player to act like the last key is being pressed again.
20:23:00lostlogickarashata: but other codecs work? What kind of tags are on your FLACs?
20:23:30karashataI'm not sure, whatever tags Winamp uses to tag FLAC files
20:23:39lostlogicmrkiko: so the problem you are reporting today acts as though you've held the next button down and keeps repeating that action?
20:24:01lostlogickarashata: lemme find a build for you to try.
20:25:57lostlogickarashata: can you try 15464 for me? It's possible that stripping leading tag data from the buffer is causing your trouble.
20:25:59mrkikolostlogic: yes
20:27:00lostlogicmrkiko: on the H340, which direction button is it that we are speaking about? Left or Down?
20:27:04karashatalostlogic: if I can find it, I don't have SVN to grab builds with
20:27:25 Join Lear [0] (i=chatzill@rockbox/developer/lear)
20:27:56mrkikolostlogic: /me is trying to read as much text as possible - please wait
20:28:27karashatalostlogic: the closes daily build to 15464 is 15466
20:28:51lostlogickarashata: crap.
20:28:59mrkikolostlogic: ... left button
20:29:15mrkikolostlogic: the left part of navi
20:29:40mrkikoAniway: probably I can trigger it with "right"...
20:30:01karashataif someone could build me a build to check I'd be glad to test, but I don't have SVN and the time it would take to get a build environment set up would be a while...
20:30:02 Join bertrik_ [0] (
20:30:03mrkikolostlogic: it triggers frequently especially when I generate a key sequence needed to skip from a directory to the next...
20:30:30 Join merbanan [0] (n=banan@
20:30:34lostlogicmrkiko: and once it triggers, what do you hear that tells you that it is skipping repeatedly?
20:31:12lostlogickarashata: I hear ya, I'm pondering... I wish flacs weren't so large, do you have a small-ish one that causes the problem that you could send me?
20:31:48karashataI have a few that might qualify as smallish, where do you want me to dump one for youÉ
20:32:17karashataack, apparently Pidgin changed the keyboard layout again on me, be back in a sec...
20:32:19mrkikolostlogic: if "beep" is enabled, I can hear many many beeps. If it's not I can detect it from the fact that my IDE disk will simply not spin down. When I arrive to school I should reset the player aniway when this happens. In some cases the player will shut down ...
20:32:23 Quit karashata ("Leaving.")
20:32:37lostlogickarashata: my email server will accept attachments up to 20m if yours will send them
20:32:48 Join karashata [0] (
20:32:52lostlogickarashata: my email server will accept attachments up to 20m if yours will send them
20:33:14 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
20:33:22karashatahmm, I can try, I don't know if mine sends up to that
20:33:23lostlogicmrkiko: I think I understand the bug pretty well. I assume that you have tried on the latest build already?
20:34:05mrkikoYes - I got up this mornign early to test it.
20:34:08 Join akuma [0] (n=chatzill@
20:35:06mrkikolostlogic: I use daily builds
20:35:19mrkikoAiway I have here a ready build environment...
20:35:50*akuma i just installed rockbox on my ipod nano adn installed some 3 or 4 themes ...the problem is that whenever i play any song i keep hearing squeeking noises in the background
20:36:22*scorche|w wonders if akuma added any oil to his nano recently
20:36:24akumawhat do i do?
20:37:26dionoeaNico_P: cool, you fixed it :) Thanks.
20:37:36mrkikolostlogic: aniway I'm not a native english speaker, so tell me if I'm clear...
20:37:41 Quit ilgufo (Remote closed the connection)
20:38:35akumathe noise is like someone rubbing two balloons in the background
20:38:49 Quit r121 ("Leaving")
20:39:27 Join ilgufo [0] (
20:39:35Nico_Pdionoea: thanks for the report ;)
20:39:54dionoeaWell I only read the svn log... haven't had time to test yet :)
20:41:51*akuma waits paiently for someone to help him...
20:42:50 Join jgarvey [0] (
20:43:22mrkikoakuma: stupid but may be helpful question - can you try booting your iPod with the original firmware and test the problem disappears?
20:44:06akumayes ...
20:44:24 Join GodEater_ [0] (n=bryan@rockbox/staff/GodEater)
20:44:27 Quit bertrik (Read error: 110 (Connection timed out))
20:44:44 Nick bertrik_ is now known as bertrik (
20:44:45akumawhen i use the original firmware i dont hear the squeeking sounds....
20:45:04 Quit GodEater (Read error: 113 (No route to host))
20:45:13 Nick GodEater_ is now known as GodEater (n=bryan@rockbox/staff/GodEater)
20:45:34mrkikoAkuma: are you trying with mp3 files?
20:46:25mrkikoakuma: mhm... I don't have any idea... right now
20:46:33akumashall i delete the .rockbox folder and reextract
20:46:54mrkikoakuma: mm... it won't help
20:47:02LloreanIf it's a Nano, it could simply be the Nano bug
20:47:03mrkikoakuma: probably...
20:47:18lostlogicmrkiko: hmm, would it be possible to use current builds
20:47:25lostlogicmrkiko: you're being very clear.
20:47:49akumai am using thecurrent build
20:48:25mrkikolostlogic: ... ok: now I checkout and compile, ok?
20:49:14BigBambimrkiko: Or just download it
20:49:22BigBambiUnless you are adding patches
20:49:25 Join Echelon [0] (
20:49:43mrkikoBigBambi: ... downloading it now
20:50:13mrkikoplease be patient - I can hardly read ...
20:50:53BigBambimrkiko: That was a smiley, not a hurry up!
20:51:16 Join ruinedtwice [0] (
20:51:50mrkikoBigBambi: ... I know, I warned only ...
20:52:55 Quit jhulst ("Konversation terminated!")
20:53:24 Quit BigBambi (Remote closed the connection)
20:54:09akumawell overwriting tockbox folder worked...
20:55:10pixelmaaccording to firefox JdGordon's fix of Zagor's is still not right. Looks correctly if the encoding is set to utf-8 now which makes me wonder what encoding it was before...
20:55:16 Join atsea- [0] (i=atsea-@gateway/tor/x-48fc633452b3f594)
20:55:30akumaare tehre any tried and tested themes for rockbox that i can use?
20:56:48*mrkiko is writing the firmware to the player
20:56:51 Join Toxicity999 [0] (n=bryan@unaffiliated/Toxicity999)
20:57:22*pixelma adds a "name" to her previous sentence
20:57:29 Quit Toxicity999 (Read error: 104 (Connection reset by peer))
20:57:34 Join Toxicity999 [0] (n=bryan@unaffiliated/Toxicity999)
20:58:18mrkikook - booting rockbox
20:58:52akumamrkiko: when i am installing the themes do i even copy hte .DS_Store file?
20:59:16 Join BigBambi [0] (n=alex@rockbox/staff/BigBambi)
20:59:46 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
21:00:17akumawhen i am installing the themes do i even copy the .DS_Store file?
21:00:31pixelmano, you don't need to
21:00:32mrkikoOk - It didn't crash while in the track list
21:00:42mrkikobut It crashed hard on the main menu
21:00:57mrkikothe one saying "files" - "Database" and so on
21:01:08akumaok thanks.
21:01:10pixelmaakuma: looks like something that the operaring system of the theme author added
21:01:11mrkikobut this is fs#8003: the problem I mentioned initially has been solved
21:01:27lostlogicmrkiko: I'm glad that at least one problem is solved :)
21:01:57mrkiko:) I'm happy also
21:01:58 Quit hcs (Read error: 113 (No route to host))
21:02:33mrkikodoes someone know a way to get the H340 out of a crash without the need to press the reset button' It's so difficult! :)
21:02:48mrkikoand plugging the reader to the usb doesn't help, seems
21:02:58akumapixelma: do you use a ipod?
21:03:02lostlogicmrkiko: and so you know, you are much better spoken, in English, and more able to respond and manage your computer/player than many people who are both sighted and native English speakers. I feel kinda humbled.
21:03:16pixelmaakuma: no
21:03:41***Saving seen data "./dancer.seen"
21:04:02mrkikolostlogic: thank you so much!
21:04:15akumapixelma: ah you wouldnt be able to suggest me some theme...?
21:04:28pixelmanot at all
21:04:34Lloreanmrkiko: Unfortunately, the whole point of the "Reset" button is for when there is no other way to reset it.
21:04:51 Quit Nimdae_ ("reboot")
21:05:24mrkikoLlorean: no problems, I will reset it later when I find something appropriate :)
21:05:40akumapixelma: so what theme do i use?
21:05:41 Quit mf0102 ("Verlassend")
21:05:42pixelmaakuma: also, themes are made for a specific screen (size, colour depth) so there is no them for Ipod
21:05:58 Join bernd_b [0] (
21:06:02Lloreanmrkiko: The reset button really basically disconnects and reconnects power, so it's not something handled by programming but a much more "physical" solution.
21:06:25pixelmaakuma: in general, I mean. I use my own theme...
21:07:37mrkikoLlorean: .. ehm... the problem basically (at least for H340) is that to reset you need something very very very fine
21:08:13Domonokyakuma: you could try holding the "power on" button (i think its play on h3x0) for about 15sec, for many crashes this works to shutdown.. :-)
21:09:46akumaah...fixed it by re installing rockbox....
21:09:54Lloreanmrkiko: I know. It's unfortunate. But it's just not a "button" like the other ones, so you can't even reprogram another easier to use button to do the same thing or anything. It's one of those things we just can't really fix.
21:09:57 Join linuxstb [0] (n=chatzill@rockbox/developer/linuxstb)
21:10:03akumathink i messed with some setting...
21:10:18 Quit ilgufo ("So Long, and Thanks For All the Fish -")
21:10:41 Join ilgufo [0] (
21:10:50*akuma decides its time to sleep....rockbox rules....
21:11:03akumathanks all for helping....
21:11:15mrkikoDomonoky: unfortunately with this crash the thing doesn't work.
21:11:15 Part akuma
21:12:00Domonokyit only works if rockbox isnt completly crashed..
21:12:37 Quit jhulst ("Konversation terminated!")
21:13:37 Join iamben_ [0] (
21:14:40 Join spiorf [0] (
21:16:05bertriki get a bit of a cracking sound on some mp3s, on the bass/snare, but it happens at low volume too
21:16:21bertrikdoes this sound familiar to anyone?
21:17:08 Quit Siku ()
21:17:12Domonokybertrik: you use any eq or bass enhancment in rockbox ?
21:17:25bertrikno, disabled eq and crossfeed
21:18:38bertrikthe same mp3 sounds clean on my pc
21:18:52zewith same headphones?
21:18:59bertrikze: yes
21:19:16Lloreanbertrik: You have all settings cleared in Rockbox?
21:19:40bertrikLlorean: no, I'll try that, but I think I already tried that
21:22:21bertrikIt's still there with cleared settings, I think I'll try to going back some versions and pinpoint it
21:23:17 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
21:24:10hcslostlogic: well, predictably, I was standing in line and suddenly I was blasted with loud static, looks like the old ADX isn't as perfect as I'd thought
21:24:22lostlogichcs: wheee
21:24:57hcsI'm really getting tempted to just bypass the buffer
21:25:17 Quit iamben (Read error: 110 (Connection timed out))
21:25:26ruinedtwicethe wiki cites the frontend "winff" and it's underlying executable "ffmpeg" as good candidates for video encoding for rockbox. My searches through the winff / ffmpeg docs, the wiki, and the forums have not produced for me a useful answer to the following question...
21:25:36 Join akuma [0] (n=chatzill@
21:25:46akumathe squeeking is back
21:25:59lostlogichcs: how?
21:26:20hcslostlogic: by just opening the file
21:26:46mrkikoCan someone try to trigger fs#8003 on another player different from h320 and h340?
21:27:16lostlogichcs: the sound you hear is hdd based players screaming in pain
21:27:48akumai have nano
21:28:12akumaits a flash drive player right?
21:28:39 Quit amiconn (Nick collision from services.)
21:28:44Lloreanakuma: There are multiple conversations going on. That's why people precede their comments with who they're directing them at in many cases. For example the one about hdds was to HCS, not you
21:28:47 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
21:28:48ruinedtwicenamely, I would like winff to drive ffmpeg to produce well behaved split files. Please, what additional command line parameter should one enter in winff in order to compell ffmpeg to split the output file into, say, 50mb parts.
21:29:00Lloreanakuma: As I said quite some time earlier, it's very likely you're simply experiencing the current problem on Nanos, which there is no yet fix for.
21:30:16Lloreanruinedtwice: Why do you want to split files, there's not a very practical reason to do this in Rockbox?
21:30:23akumaLlorean: ah when will the next build going come out?
21:30:57akumasoon ?
21:31:25 Join Angryman [0] (i=50e0a118@gateway/web/cgi-irc/
21:31:27LloreanIt's not fixed in the next build...
21:31:41lostlogicakuma: new builds come out almost constantly, but it's not likely that any particular one will fix your specific problem.
21:32:02akumasoi got to report a bug?
21:32:09 Quit Angryman (Client Quit)
21:32:15Lloreanakuma: As I said, it's a known issue...
21:32:23 Join Angryman [0] (i=50e0a118@gateway/web/cgi-irc/
21:32:27 Join TMM [0] (
21:32:29ruinedtwiceLlorean: I had rockbox choke on a file that happened to be "large" −− esprit d'experimentation
21:33:17 Quit krazykit ("Connection reset by beer")
21:33:33Lloreanruinedtwice: Did you know that Rockbox is supposed to work on any length video?
21:33:53 Quit ompaul (Client Quit)
21:34:31LloreanIf a file isn't working as it's supposed to, you should post a proper bug report...
21:34:37 Join handmadematters [0] (
21:34:40ruinedtwiceLlorean: yes, and I hope and intend to get my installs of it (h340 and sansa 280) to behave better. At the moment either they are misbehaving or else (more likely), I am an incompetent guardian.
21:34:44 Join tokam_one [0] (
21:34:57tokam_oneis itpossible to get a java plugin for rockbox?
21:35:07tokam_onei want to be able to execute j2me code on it
21:35:13 Join krazykit [0] (
21:35:20Lloreantokam_one: If you write one.
21:35:29hcstokam_one: that'd be a neat trick, but I don't know of any such project in development
21:35:30tokam_oneLlorean: :D
21:35:59tokam_onei would have fun to start one
21:36:07tokam_onebut i do not hav time for more new projects
21:36:28tokam_onei have a typo3 project running to januar
21:36:53handmademattersMaybe another stupid question: it seems that i cant get the adc values read out of the plugin section? Any advice? And yes i know how to register it (plugin.c plugin.h) :)'s useful to be reminded that rockbox is meant to support "any size" of video file. That encouraging. I'd still like to know the syntax to make winff produce files split to order. I guess it's my narcissism showing.
21:37:41lostlogichandmadematters: that is a bad idea (giving plugins direct access to an adc)
21:38:05lostlogicruinedtwice: ask in a winff channel/forum?
21:38:08handmademattersBut i need to test some smoothing code...
21:38:19lostlogichandmadematters: ah, ok, good though :)
21:38:20handmademattersand i thought this would be the best place
21:38:21bertriksmoothing code?
21:38:40Lloreanruinedtwice: Well, considering the fact that nobody here wrote WinFF, perhaps you should seek questions that aren't actually related specifically to Rockbox to the creators of that software instead...
21:38:40handmademattersH10 got jumpy values
21:38:42lostlogichandmadematters: you would need to add a pointer in plugin.h and an assignment to a function in plugin.c
21:38:46tokam_oneI guess if a j2me plugin for rockbox will be released, more developers will work on rockbox
21:38:52handmademattersdid that
21:39:05handmademattersreads the very first value then it stops
21:39:07tokam_onethe java guys (like me) will start to develope programms and games for rockbox
21:39:14Lloreantokam_one: Considering the ones we want are ones who will work on the core, I'm not sure that's a huge incentive.
21:39:18lostlogichandmadematters: stops or keepds getting same values?
21:39:21tokam_oneand than we can might use other programms written for j2me
21:39:21lostlogic(ie does it freeze?
21:39:29Lloreantokam_one: Rockbox is a music playing firmware, we really care very, very little about plugins/games, overall.
21:39:33handmademattersgets the same values
21:39:36handmademattersno stops
21:39:40LloreanThey're more "fluff and added bonus" than a focus in any way.
21:39:57lostlogichandmadematters: the adc is read over an i2c bus (IIRC) you might need to wait some ticks between polls?
21:40:03handmademattersThe plugin works further only the adc doesnt respons any further
21:40:04tokam_onewhat does the word incentive mean?
21:40:36krazykittokam_one, a reason to do something
21:40:36Lloreantokam_one: "Encouragement for doing something"
21:40:39linuxstbruinedtwice: Tried the "-fs" option to ffmpeg?
21:40:55handmademattersAlso thought about that... But the strange thing is that half a year ago exactky the same procedure worked...
21:40:58tokam_oneLlorean: Apple ipod is also a music player, rockbox can only be better if it is more than a music playing software
21:41:04handmademattersBut i will give this a try
21:41:16Lloreantokam_one: Or simply a better music player.
21:41:32lostlogichandmadematters: there have been significant timing changes to the adc drivers on many targets in that time...
21:41:32tokam_oneLlorean: why it is better, if it just can play music and video like others too?
21:41:39Lloreantokam_one: You could take every single plugin out of Rockbox and 99% of current users would probably still use it.
21:41:53handmademattersYep thats the reason why i start my work on that again...
21:41:54 Quit akuma ("ChatZilla [Firefox]")
21:41:54Lloreantokam_one: Seriously, do you not know what features Rockbox has?
21:42:00tokam_oneLlorean: you mean to implement new drivers for other hardware with the "work on the core" ?
21:42:03*bertrik agrees with Llorean
21:42:03 Quit Frazz (Read error: 104 (Connection reset by peer))
21:42:06lostlogicjava on rockbox is a nonstarter, IMO −− the JVM requires a lot of threads just to run itself and we don't want to have that many threads living
21:42:10ijuz_Llorean: not the bubbles! ;)
21:42:15lostlogicbut then that's just my opinion and I'm also against making rockbox preemptive
21:42:28tokam_oneijuz_: lol :D
21:42:37Lloreantokam_one: A wide, wide range of supported formats, parametric equalizer, gapless playback, dithering, replaygain, crossfeed, crossfade, advanced on-player playlisting and database management, themeability, just to name a few features.
21:42:50tokam_oneLlorean: i think that rockbox is better, but is only because it can do more
21:43:01tokam_oneif it could just do the same like others it would not be
21:43:22Lloreantokam_one: Yes, but you can do "more" while still focusing on music and media related features.
21:43:50tokam_oneLlorean: you can do more with a lot of things....
21:43:51Lloreanlostlogic: My understanding is that the Phone Edition VM would be within the limits of our players.
21:43:52 Join Buschel [0] (
21:44:01Lloreantokam_one: Yes, but as I said, it's not a focus of the project.
21:44:04tokam_oneLlorean: many users know how to convert formats ..
21:44:12LloreanConverting formats degrades quality.
21:44:17LloreanThis is obvious...
21:44:27tokam_oneLlorean: do you think, that rockbox will oneday be able to play wma without coprocessor?
21:44:34LloreanIt already can
21:44:37BigBambitokam_one: It does
21:44:39tokam_oneiT CAN?
21:44:46tokam_onei will do an update now :D
21:44:47*Domonoky thinks rockbox would happily accept a java vm if someone wrote one for rockbox... but someone has to volunteer... :-)
21:44:48LloreanSeriously, learn about the program before commenting on it, please.
21:44:52tokam_oneeven on ipod 5?
21:45:01tokam_onesince when?
21:45:03linuxstbIt always has done - what made you think it didn't?
21:45:04BigBambitokam_one: Apart from the portalplayers, what about all the single core daps we have?
21:45:15LloreanDomonoky: I'm quite sure it would. All I ever said was that "attracting plugin developers is not an incentive for Rockbox core devs to spend resources on it"
21:45:22ruinedtwicelinuxstb: yes −− and I suspect that it "should" do the trick. I'm embarrassed to confess that I haven't yet guessed the required syntax for that option: "-fs 50", or "-fs 50m", or "-fs 50mb", or "-fs 50 mb", and so on. None of my attempts so far have produced anything other than abort upon execution. So sad. I need to get this little tweak working to make the sky stop looking like a Munch etching.
21:45:32Lloreantokam_one: Since inclusion of WMA support, it's always run on one core.
21:45:48tokam_oneok, i will look after it now
21:46:00DomonokyLlorean: true, but we dont have any ressources to distripute.. its all volunteer.. :-)
21:46:01Lloreanruinedtwice: It's in bytes, so -fs 50000 would be 50000 bytes.
21:46:29tokam_oneflv format would be nice
21:46:32tokam_oneflash video
21:46:39BigBambitokam_one: Then get coding....
21:46:40Lloreanruinedtwice: You can add units though. But it doesn't split for you, so you'd need to know how long the file is in time, and convert a second time, with the time offset starting at that time to make a second 50mb file, and it might be imperfect for overlap
21:47:07Lloreantokam_one: Considering the necessity to resize for the iPods screen anyway, while converting audio is a no-do, converting video is essentially a necessity anyway...
21:47:17bertrikmmm, I even get crackle from SVN 15000... :(
21:47:26tokam_one^^ Llorean, ok
21:47:34tokam_oneLlorean: it suxx :) hehe :D
21:47:59tokam_oneLlorean: an encoding software inside rockbox would be too slow would it?
21:48:09LloreanThese players are very, very slow.
21:48:10LloreanSo yes.
21:48:59tokam_oneLlorean: have you developed for the rockbox project?
21:49:13tokam_onehave you developt might the textviewer?
21:49:30LloreanMostly I've just fixed a few bugs.
21:50:02 Quit Angryman ("CGI:IRC (EOF)")
21:50:10 Quit spiorf (Connection timed out)
21:50:15tokam_onei have a little improvement idea, but i heared that it was allready planed or developed, but it did not work on my ipod 5, but before telling more i will install now an update :D
21:50:24 Quit Rick ("updating mirc")
21:50:28 Join spiorf [0] (
21:50:51 Join Frazz [0] (
21:51:13tokam_onei am talking about a funktion with reminds textposition not only for 1 file, also for more 5-10 or so
21:51:23tokam_one-k +c
21:51:46 Join Rick [0] (
21:51:48BigBambiI thought it already did remember the last few files position
21:52:53tokam_onei will download the update now
21:52:59tokam_onemine is 2 -3 month old
21:53:02BigBambiIt has done for ages
21:53:12tokam_oneit remembers only the position for one page on my ipod 5
21:53:17*mrkiko resetted the player
21:53:21BigBambitokam_one: Please don't comment or report problems with anything less than the current build
21:53:45BigBambitokam_one: It is incredibly annoying trying to help with a 'bug' that has been fixed ages ago
21:53:52tokam_onebut in a forum i was told, that it should allready remember it for some more files
21:54:01BigBambiAnd so it does
21:54:01tokam_oneBigBambi: you are right
21:54:18tokam_onewhen i was told i downloaded the latest update :)
21:54:27tokam_onebut i will do now again, and test it ;) :D
21:54:36tokam_onerockbox is nice
21:54:40BigBambitokam_one: You should always have the latest before questioning anything
21:54:54tokam_oneBigBambi: you are right
21:55:21tokam_onequestioning yes, but no bug reporting with an old version
21:55:35BigBambiQuestioning no
21:55:47BigBambiIf you had updated you would know it worked
21:55:57BigBambiThen we wouldn't have ad to have this conversation
21:56:03tokam_oneBigBambi: the question for j2me was ok i guess
21:56:24BigBambitokam_one: I'm just talking about current rockbox functionality
21:56:38tokam_oneyes, there you are right ;)
21:57:15 Join Llorea1 [0] (n=llorean@
21:57:30 Quit Llorean (Nick collision from services.)
21:57:45preglowamiconn: had a chat with some people in #gcc, and they said it'd be perfectly ok to ping the binutils mailing list with the bugzilla entry if nothing happens
21:57:45 Nick Llorea1 is now known as Llorean (n=llorean@
21:57:53preglowamiconn: how's brightness stuff coming along, bnntw?
21:57:55 Quit XavierGr (Nick collision from services.)
21:57:56 Join XavierGr_ [0] (
21:59:16tokam_oneis it the right way for an update to replace the .rockbox folder?
21:59:23BigBambiyep, just overwrite
21:59:49Buscheldid anyone ever try to set hi 5G ipod to 100MHz?
21:59:52tokam_onethan i gone reboot now
22:01:00preglowBuschel: why would we want to? :>
22:01:12tokam_onedamn nice
22:01:22tokam_onetextviewer works how i want it :)
22:01:25Buschele.g. to have 16:9 video at 24fps?
22:01:28tokam_oneyou do nice work
22:01:56 Join JRoT|Stage [0] (
22:02:30tokam_oneif there would be an j2me plugin for rockbox even a noob like me will be able to write java programms for his ipod or so
22:02:51tokam_onei think it is a nice idea, but with c i am not good enaugh to write an implementation
22:03:46scorche|wthen sit back and wait for someone else to if it happens at all
22:03:56BigBambitokam_one: I think it unlikely, but should someone write one I'm sure it shall be considered
22:04:12BigBambiHowever, I do not think any of the core devs will be jumping on this one
22:05:22LloreanI think there was someone trying to drum up interest in a JAVA forum, but again, it was full of people saying "Someone ought to make this" rather than people saying "I'm willing to make this" or trying to get work done.
22:06:06tokam_oneand nobody was able to manage it?
22:06:21BigBambitokam_one: More like nobody bothered
22:06:39tokam_oneBigBambi: do you mean with your second post, that nobody whant to switch his project and change work form rockbox core development to j2me development
22:06:43hcsI'm in the planning stages of implementing a different virtual machine, but that's another issue entirely...
22:06:45pixelmaBuschel: check last weeks log for w1lliam (or similar)
22:07:01BigBambiIt is all very well saying, someone should do this, wouldn't it be great etc., but until someone who wants it writes it it won't happen
22:07:08tokam_oneBigBambi: nobody bothered without java?
22:07:24BigBambitokam_one: That none of the core devs could give two hoots about java on rockbox
22:07:30Domonokytokam_one: people develop only what they want to develop.. so if noone has interesst..
22:07:45BigBambitokam_one: No, no one bothered to write a java interpreter for rockbox
22:07:47Lloreantokam_one: Someone who wants JAVA should work on it, but there are lots of bugs and other things that most of us consider far more important.
22:08:13pixelmaBuschel: starting here
22:09:06Buschelpixelma: thanks, will take a look
22:09:06 Quit Domonoky (Read error: 104 (Connection reset by peer))
22:09:10 Join Thundercloud [0] (
22:09:33 Join Domonoky [0] (
22:09:52 Quit hcs ("Leaving.")
22:09:53 Quit karashata ("Leaving.")
22:10:37 Join karashata [0] (
22:10:38tokam_oneBigBambi: and what about it now?
22:10:44tokam_oneBigBambi: it did not work?
22:10:51BigBambitokam_one: Did what not work?
22:11:09tokam_oneBigBambi: the java interpreter for rockbox
22:11:24scorche|wtokam_one there is what people are telling you
22:11:25BigBambitokam_one: There isn't one
22:11:39BigBambitokam_one: No-one who wants one has bothered to write one
22:12:13 Nick Bagder_ is now known as Bagder (
22:12:18tokam_onescorche i am not as good in english as BigBambi is i guess
22:12:19 Join keanu [0] (n=keanu@unaffiliated/keanu)
22:12:42tokam_onei look up many words of him :D do you come from the usa BigBambi ß
22:12:50BigBambitokam_one: I'm British
22:12:51 Quit ilgufo ("So Long, and Thanks For All the Fish -")
22:13:07 Quit nicktastic ("Leaving")
22:13:12tokam_oneBigBambi: ^^ i see, you talk like a native speaker
22:13:22BigBambiThat's because I am
22:13:39 Join XavierGr [0] (
22:13:42scorche|wenglish didnt originate int he us...
22:13:54BigBambiI'm the original native speaker!
22:14:09tokam_onehcs: on what kind of vm do you work?
22:14:27tokam_one(22:06:35) hcs: has he left?
22:15:06scorche|w*** hcs [n=agashlin@rockbox/contributor/hcs] has quit ["Leaving."]
22:19:07 Join midgey [0] (
22:19:49 Quit JR0T (Read error: 110 (Connection timed out))
22:23:42 Quit ruinedtwice ()
22:26:12 Quit XavierGr_ (Read error: 110 (Connection timed out))
22:28:20Buschelok, got my 5.5G 30GB pod running at 100MHz :o) works fine for the last minutes and tests
22:30:18 Quit merbanan ("Leaving")
22:33:47 Quit Buschel ()
22:36:01tokam_onehehe he quit
22:36:12tokam_onetomorrow i have to ask bushel and hcs
22:36:23Nico_Ptokam_one: about?
22:36:24tokam_onebushel for 100mhz ipod and hcs for java on rockbox
22:37:06scorche|wwhy do you want to run the ipod at 100MHz?
22:37:08Nico_Pjava on rockbox sounds like pretty unlikeliy
22:37:21Nico_Pand a bit useless too
22:37:35scorche|wand hcs said nothing about java
22:37:46tokam_onehe did
22:37:54tokam_onescorche isn't it faster
22:37:57Bagderthat java crap pops up every once in a while
22:38:00Bagderit'll die again
22:38:11tokam_onescorche but the batteriy will lowered down faster will it?
22:38:23scorche|wtokam_one yes...why would you want to run it faster?
22:38:43tokam_onescorche fast is nice, but batterie is more important i guess
22:38:50scorche|wand no he didnt...he was talking about something different than java
22:38:53tokam_oneBagder: I hope not so
22:38:59scorche|wbut why would being fast be nice?
22:39:06Lloreanscorche|w, tokam_one: There's no real harm in changing the "boosted" speed as long as the unboosted speed stays the same, as long as boosted is within spec, I think
22:39:06scorche|wwhy would you need it faster for?
22:39:22Lloreani mean, you draw more power, but at less time, and iiuc the ratios work out so that it should be the same net power draw, roughly
22:39:28scorche|wLlorean: i am trying to get him to tell me his reasoning of why he wants it faster
22:39:35scorche|wyou dont need to reference me ;)
22:39:43LloreanIt _could_ help with APE.
22:39:56BigBambiand java? :)
22:40:20tokam_oneBigBambi: you want to know why i need java?
22:40:24scorche|wLlorean: oh yes...there are plenty of reasons why one would want it faster, but i was curious if he would say one of them, or wanted it for no reason
22:40:28 Join Zagor [0] (
22:40:44Lloreanscorche|w: True, plenty of people just think faster is better.
22:40:49BigBambitokam_one: I was just making an unfunny joke
22:41:05BigBambiAbout java needing speed
22:41:14LloreanBut really, we don't need to go faster unless we're fixing a problem with it at this point, since we have to optimize everything for the 80mhz iPods either way.
22:41:14tokam_onescorche|w: it is what Llorean mentioned
22:41:31tokam_onebut the fast lowered power is the bad aspect of it
22:41:46*Llorean just said it shouldn't increase power consumption
22:42:11 Quit Dark_Apostrophe (Read error: 104 (Connection reset by peer))
22:42:12LloreanNow the crazy thing some people have done where they increased their unboosted speed to 60mhz (I will never understand that, it really baffles me) will destroy what minimal battery life we already have on the iPods.
22:42:31tokam_oneLlorean: why not, if my notebook runns on 2x2ghz it is warm fast, on 2x800mhz not
22:42:47Lloreantokam_one: Rockbox runs by boosting when it needs to
22:43:07LloreanIt would be like your laptop running at 800mhz most of the time, and changing to 2000 to catch up if it's not far enough ahead any more
22:43:22LloreanIf it went to 1500 to catch up instead of 2000, it'd use less power when at 1500 but have to be there longer so it doesn't really gain anything.
22:44:26tokam_onebut to give rockbox the possibility to go up to 100 i have to enable it?
22:44:47LloreanThere's not a good reason to.
22:45:04LloreanThere's much more suitable lossless compression for audio.
22:45:20tokam_oneLlorean: ok
22:45:28 Join bV [0] (i=4128a885@gateway/web/cgi-irc/
22:45:47tokam_onei have to go now
22:46:36tokam_oneis there a special c compiler for rockbox systems?
22:47:15Lloreantokam_one: There are plenty of developer documents in the wiki.
22:47:15linuxstb"special" is an apt word for it...
22:47:21tokam_onei never wrote programms in c, but have i to work vry closed to the hardware, or is there an interface given by ther rockbox core?
22:47:45tokam_onei use usually java :)
22:48:05tokam_onethere isn't an api like in java i guess :)
22:48:24BigBambiTHERE IS NO JAVA!
22:48:28bVI've got a quick question. Has anyone experienced the inability to get into the manufacturer mode after "unbricking" a Sansa e200r series player?
22:48:34Lloreantokam_one: Spend some time with the documentation for a while, please.
22:48:53tokam_oneLlorean: ok
22:48:58bVI can't attempt a re-installation of Rockbox because I can't run the bootloader modifier again
22:48:58tokam_onegood night to all...
22:49:04 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:49:07tokam_onethanks for your work, rockbox is nice!
22:51:12LloreanbV: What happens instead?
22:52:58 Quit Domonoky ("Trillian (")
22:52:59bVBlack screen. The scroll wheel lights up when I hold it down as it should, and instead of the text that says "Welcome to manufacturer mode" or whatever, I just get a black screen. I end up needing to kill it by holding down power.
22:54:08 Quit keanu (Read error: 110 (Connection timed out))
22:54:20LloreanManufacturer mode never had text...
22:54:25LloreanRecovery mode had text
22:54:40LloreanManufacturer mode is a blank screen and a lit up wheel, exactly as you're describing.
22:57:03bVOkay, I may be remembering where the text came from wrong - but I know I received some kind of notification that I had entered manufacturer mode (maybe a popup on the computer)?
22:57:27LloreanYour computer may have "recognized" the device for the first time
22:57:36LloreanBut there is normally no special notification of any sort.
22:59:47bVOkay then! I guess I must have been smoking something. I just tried the patcher again and it recognized it, and patched it. Amazing... thanks. Going to try the installation again
23:02:37 Quit bV ("CGI:IRC (EOF)")
23:03:44***Saving seen data "./dancer.seen"
23:04:22 Quit Lear ("ChatZilla [Firefox]")
23:05:53 Quit bernd_b ("Verlassend")
23:05:54*ender` yawns
23:06:35mrkiko... Now I may go...
23:07:03mrkikoHas someone said something to me? I Haven't detected this yet...
23:08:19preglowthen probably not
23:08:31 Quit Arathis ("Bye, bye")
23:08:58mrkikoI'll look at the web log tomorrow with a good braille display
23:12:21mrkikoBye! Good chatting!
23:12:25 Part mrkiko ("rockbox")
23:15:01 Quit Frazz ("Leaving")
23:17:23 Join safetydan [0] (n=safetyda@rockbox/developer/safetydan)
23:19:13 Nick parafin is now known as parafin|away (
23:19:25 Join petur [0] (n=petur@rockbox/developer/petur)
23:21:39 Quit midgey ()
23:21:59bertrikhmmm, can't reproduce the mp3 cracks on the sim
23:23:17 Join darkless [0] (
23:23:46Lloreanbertrik: Have you tried the original firmware then?
23:24:00bertrikeh, no
23:24:39bertrikgood idea to try while waiting for compile
23:27:01 Quit tokam_one (Remote closed the connection)
23:28:10 Quit jhulst (Connection timed out)
23:33:52bertrikno cracks in OF
23:36:44LloreanWhat hardware is this on?
23:38:47 Quit z35 (Read error: 110 (Connection timed out))
23:41:26bertriksansa e200
23:42:17LloreanStrange, there used to be such crackling back with older builds
23:42:42LloreanBut assuming you're using a current build and not raising the volume above 0, and none of your DSP settings are enabled, it shouldn't happen.
23:43:00bertrikit only happens on certain mp3s, but those mp3s sound fine in the sim, and it also happens at low volume
23:44:11LloreanI'm assuming you're using an official SVN revision and bootloader?
23:44:53bertrikyes, the firmware is SVN but the bootloader is an older version
23:45:10LloreanHow old?
23:45:22bertrik3 weeks or so
23:45:36LloreanThat shouldn't be a problem.
23:46:00bertrikmy build was done on cygwin though, not with the officially approved compiler, so I'll try that next
23:46:22LloreanWhy did you compile your own build if it's unpatched anyway?
23:47:22 Quit jgarvey ("Leaving")
23:47:25bertrikI was compiling my own because I was working on some patches
23:47:58LloreanWell wait, so are there source code changes or no?
23:48:36bertrikno, I reverted those to be sure
23:48:54LloreanOne thing to be test might be to simply decode the MP3 to WAV on your computer and then playback the WAV file.
23:49:54bertrikI'll try that, tomorrow. The official build also crackles by the way
23:50:14 Join Seed [0] (
23:50:27 Join hachi [0] (
23:50:52 Quit bertrik ("bye")
23:51:02hachiI've got a 5.5 80GB ipod here, the interface on rockbox occasionally hangs while it still plays music
23:51:12hachinormally I just do a hard reset of the ipod and go on with my day
23:51:22hachibut the reset isn't working, do I have any other options?
23:52:01 Part handmadematters
23:53:33 Quit davina ("xchat on Ubuntu 7.04")

Previous day | Next day