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

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

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

#rockbox log for 2007-12-05

00:00:18BagderKlevi: well, more less all companies do it to some extent
00:00:46Zagorwe actually might want to check that always. word 82, bit 3.
00:01:32LloreanZagor: I was about to say, those seem like the sort of thing we should always check. :)
00:02:05 Quit MethoS- ("Konversation terminated!")
00:04:16amiconnThe sleep issues are most probably not an issue with the sleep command, but with the pio mode when waking up again
00:04:44Zagorif we want to detect flash specifically, there is a "CFA" bit for that
00:04:58DerPapstamiconn: http://papsti.dyndns.org/upload/fat32_dump_5.5G.bin.zip
00:05:04amiconnWe're communicating at pio4 speed, but a disk that's sent to sleep might wake up in a lower pio mode
00:05:14 Quit kubiix (Read error: 110 (Connection timed out))
00:05:36amiconnDerPapst: Yeah, got that already, as well as a dump from a working rockboxed G5.5
00:06:00Zagoramiconn: well that is only a problem if we go to sleep :-)
00:06:01amiconnSo far we're lucky that most hdds didn't care about such things
00:06:35amiconnWith proper ata transfer speed handling, sending a flash disk to sleep and waking it up again should not cause problems
00:07:36 Join linuxstb__ [0] (n=chatzill@i-83-67-212-170.freedom2surf.net)
00:08:48Zagorthe question remains: does sleep have any purpose on a CF card? if not, why bother with it?
00:10:02amiconnFor those I know, it does not
00:10:11amiconnThey usually auto-sleep when not in use
00:11:59amiconnEven the microdrive used in the ipod mini seems to autosleep. At least disk poweroff has no measurable effect on mini
00:14:07 Quit Rondom ("Ex-Chat")
00:15:32 Quit ompaul (Client Quit)
00:15:45 Quit OlivierBorowski ("Konversation terminated!")
00:16:01lostlogicbagh, sansa just froze.
00:16:20 Quit Klevi ("Im forcing myself to sleep for trimester exams this week.... Be well everyone =)")
00:16:31 Join Thundercloud [0] (n=thunderc@resnet06.nat.lancs.ac.uk)
00:17:23 Quit Febs ()
00:17:49 Join Febs [0] (n=chatzill@38.98.196.75)
00:17:59 Quit Febs (Remote closed the connection)
00:19:30 Join EspeonEfi [0] (i=espeonee@STRATTON-THREE-EIGHTEEN.MIT.EDU)
00:20:02 Quit EspeonEefi (Read error: 104 (Connection reset by peer))
00:20:42 Quit Delvien (Read error: 104 (Connection reset by peer))
00:20:59 Quit linuxstb (Read error: 110 (Connection timed out))
00:21:57 Quit EspeonEfi (Client Quit)
00:22:15 Join Delvien [0] (n=dm@cpe-65-24-167-246.columbus.res.rr.com)
00:22:53 Join radinp [0] (n=pradin@vpnwl-229-104.net.rpi.edu)
00:22:55CtcpIgnored 1 channel CTCP requests in 0 seconds at the last flood
00:22:55*preglow needs a better fixed point log :/
00:24:28 Part Daolan
00:24:46preglowamiconn: you got a 8.24 flog() lying around? :)
00:25:08 Nick Delvien is now known as DM| (n=dm@cpe-65-24-167-246.columbus.res.rr.com)
00:25:32 Quit Langly- (Read error: 113 (No route to host))
00:25:33amiconnNo, still not
00:25:50preglowseems my problem with eq plotter is that i need to shift away too many frac bits
00:28:26preglowargghhh, i miss floating point :/
00:28:55pregloweven with a perfect 24 bit log, the results aren't good enough
00:30:16preglowthat hardwired approach is starting to sound a lot more temptiung
00:31:45 Join J3TC- [0] (n=jetc123@pool-72-76-187-15.nwrknj.east.verizon.net)
00:33:02 Join l7 [0] (n=l7@evil-wire.org)
00:33:41 Quit petur ("Zzzzz")
00:35:13safetydanhttp://astroinfo.svn.sourceforge.net/viewvc/astroinfo/trunk/astroinfo/source/FixedPoint.c?revision=624&view=markup might has some inspiration for a fln() function at least
00:37:22preglowat least that one looks rather accurate
00:37:32preglowbut not too fast
00:37:39 Join EspeonEefi [0] (i=espeonee@STRATTON-THREE-EIGHTEEN.MIT.EDU)
00:39:01 Quit spiorf (Remote closed the connection)
00:39:23amiconnThe principle used for exp() and log() in gray_core.c is easily adjustable for other fixed point formats
00:39:55amiconnIt is based on http://www.quinapalus.com/efunc.html
00:41:20preglowamiconn: do you know how accurate they are?
00:42:20***Saving seen data "./dancer.seen"
00:43:12 Nick linuxstb__ is now known as linuxstb (n=chatzill@i-83-67-212-170.freedom2surf.net)
00:43:15 Join FOAD_ [0] (n=dok@dinah.blub.net)
00:49:09amiconnThe article has some remarks regarding accuracy. I only know that they are accurate enough for gamma calculation
00:52:10 Join SirFunk [0] (n=Sir@206-159-155-246.netsync.net)
00:52:25 Quit Zagor ("Client exiting")
00:52:41 Quit scorche|w ("CGI:IRC (Ping timeout)")
00:52:51 Part dan_a
00:53:59 Join Calcipher [0] (n=Calciphe@ool-18bab657.dyn.optonline.net)
00:55:03Calcipherbeen like 2 weeks or more maybe, I wonder what great advances have been implemented
00:55:55safetydanCalcipher: http://www.rockbox.org/twiki/bin/view/Main/MajorChanges
00:56:40markunCalcipher: and for a very detailed report: http://www.rockbox.org/since-4weeks.html
00:56:41*DerPapst thinks that the MajorChanges link should be on the front page
00:57:04preglowit still isn't?
00:57:12DerPapsti can't find it
00:57:27*DerPapst tries harder
00:57:27Calcipherthanks all of you, I'll take a look
00:57:38safetydannope, not on the front page
00:58:18markunamiconn: that algorithm for calculating log's is pretty cool
00:58:19LloreanIf it were, the fact that someone had to look harder would've meant it's not prominent enough anyway, I think. It seems like it, and "Rockbox Features" should be two of the most obvious links.
00:59:41 Quit FOAD (Read error: 110 (Connection timed out))
00:59:42 Nick FOAD_ is now known as FOAD (n=dok@dinah.blub.net)
00:59:42 Nick parafin is now known as parafin|away (i=parafin@paraf.in)
00:59:49 Join Lebowski_ [0] (n=jc@pool-71-106-183-237.lsanca.dsl-w.verizon.net)
01:00
01:00:28Lebowski_is it impossible to get e200R to get recognized by a Mac?
01:01:08Calcipherno, I see it on my mac, but mine is rockboxed
01:01:08*DerPapst doesn't find "Rockbox Features" on the frontpage either :S
01:01:25Calcipherrunning 10.4.9
01:01:36*DerPapst blames his glases
01:01:38 Quit ender` (" Variables won't. Constants aren't.")
01:01:52Lebowski_It never mounts for me
01:02:28Calcipherwierd, I even accessed it at work on one of theyre Imacs
01:02:38Lebowski_you have the R model?
01:02:49Calcipherwhat mode do you use it in?
01:02:52Calcipheryeah I do
01:02:54Lebowski_MSC
01:03:08Calcipherme too
01:03:17 Quit l7 ("Lost terminal")
01:03:38markunLebowski_: this might help: http://www.rockbox.org/twiki/bin/view/Main/SansaE200RInstallation
01:04:07 Quit Jon-Kha (Remote closed the connection)
01:04:09markunhm, maybe not
01:04:36 Join BMill [0] (n=chatzill@pool-71-191-145-253.washdc.fios.verizon.net)
01:04:37Lebowski_if doesn't show up with the orignal firwmare Rockbox sure wont help
01:04:39Calcipherso the speex changes were completed?
01:04:54LloreanLebowski_: R series don't have a choice named "MSC". What's the actual name of the choice you've picked?
01:05:10Lebowski_Rapsody
01:05:20Lebowski_mode
01:05:33LloreanHave you tried mounting it manually?
01:05:35Calcipherso I see it was noted that users should update theyre voice files, is there a detailed description of how to do this for the new speex set up?
01:05:48LloreanCalcipher: The new instructions are up exactly where the old ones were.
01:06:08Calciphercool thanks Llorean
01:06:19 Quit BMill (Client Quit)
01:06:23 Join BHSPitMonkey [0] (n=stephen@unaffiliated/bhspitmonkey)
01:06:57Lebowski_does anyone know this error code when I use etool "Control message (-71, Protocol error)"
01:07:47 Join nicktastic [0] (n=nick@unaffiliated/nicktastic)
01:10:37 Quit lee-qid (Read error: 110 (Connection timed out))
01:11:18advcomp2019Lebowski_, are you in manufacturing mode
01:13:05Lebowski_i believe so... Blue Ring is light up (nothing on screen)
01:13:45 Join donutman25 [0] (n=chatzill@65.75.87.48)
01:14:04 Quit donutman25 (Client Quit)
01:15:04 Quit DM| ("*bashes head against keyboard*")
01:17:46 Quit J3TC- (Read error: 113 (No route to host))
01:21:34 Quit lastebil (Remote closed the connection)
01:21:37 Join lastebil [0] (n=truck@cube.lomal.la)
01:23:40 Join zicho [0] (n=martin@c-5f9fe355.68-7-64736c14.cust.bredbandsbolaget.se)
01:24:15advcomp2019Lebowski_, did you hold the REC button while running the e200tool if that is what you are running
01:26:05 Part linuxstb
01:26:45Lebowski_yes sir
01:29:30 Quit Nico_P (Remote closed the connection)
01:31:54 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
01:32:12 Join Febs [0] (n=chatzill@207-172-204-33.c3-0.rdl-ubr4.trpr-rdl.pa.cable.rcn.com)
01:33:25Nico_PI don't mean to start a big VCS debate again, but for amiconn and linuxstb, here is an interesting read: http://keithp.com/blog/Repository_Formats_Matter/
01:33:37Calcipherhey, heres a quick question, if I was going to generate a new voice file, and I just updated from svn after 3 weeks
01:33:59Calcipherdo I need to make regular before I make voice?
01:34:05Nico_Pamiconn: some of the points in there are those I was trying to make when we last talked about the subject
01:34:22 Nick _pill is now known as pill (i=pill@sloth.shellfx.net)
01:34:42Calcipherso I would have to run configure and run it on normal, then configure-advanced-voice
01:34:46Calcipherthen make voice
01:36:21 Quit kugel (Read error: 104 (Connection reset by peer))
01:36:23advcomp2019Lebowski_, what linux distro are you using
01:37:04Lebowski_ubuntu
01:37:08CalcipherI'm using cygwin since I had trouble accessing my debian VMware share
01:37:25Calcipheroops, thought you were asking me
01:38:35safetydanCalcipher: I'd say so. You'll need to build the new voice file encoding tool at least (rbspeexenc)
01:39:24Calcipheroh? is that a separate process from the normal compiling?
01:39:55safetydanCalcipher: no it should be built for you if you follow the normal voice building process
01:40:34Calcipheroh ok, hey how do I remove a directory again? I thought it was the rm command
01:40:49CalcipherI wanted to clean out my build dir
01:42:12 Quit PaulJam (".")
01:42:24safetydanit is the rm command
01:42:28Calcipher-R?
01:42:33safetydanlowercase -r
01:42:45safetydanNico_P: that is an interesting article.
01:42:56Calcipheroh, so what is upper case r in this case?
01:43:17Nico_Psafetydan: yeah, I think a lot of it relates to rockbox
01:43:27safetydanCalcipher: turns out it's an alias for -r anyway so it wouldn't matter
01:43:40Calcipherhaha, ok thanks
01:45:03 Join Mouser_X [0] (n=mouser_x@207.155.176.3)
01:45:19safetydanNico_P: well it pretty much sold me on git :)
01:45:26Nico_Phaha
01:45:37Nico_Pweren't you already, though?
01:46:04safetydanpretty much, but that article added some extra push
01:46:13advcomp2019Lebowski_, so you followed the steps that i provided on ABi as a link and you got that message
01:46:38safetydanthe distributed backup and offline operation thing feels like something we need given the unreliability of access to rockbox.org
01:47:35LloreanIs there a way to push to the central repository WITHOUT having four million revisions added by someone who compulsively commits locally with each added function?
01:47:36Lebowski_advcomp, yes
01:47:44*Llorean knows very little about git
01:47:58Nico_PLlorean: yes, but that requires the person to rebase
01:48:26Nico_PLlorean: it's more a question of good practice than VCS
01:48:38amiconnThe distributed backup argument is about the only one that makes some sense to me
01:48:49LloreanI know there's value in having small individual commits to highlight changes as they describe, but there's also value in having the commit that adds the feature add the whole feature, instead of the whole development process for the feature.
01:49:06Llorean#2 is, I think, completely pointless for Rockbox
01:49:08amiconnBut git being fast on linux doesn't make it magically faster on cygwin or native windows, or easier to learn
01:49:41DerPapstiirc the mozilla project doesn't use git because it's a pain on windows
01:49:45Llorean#3 is nice, and #1 seems an issue that could be debated since it's really not dependent on what your central repository is using.
01:49:54Nico_PLlorean: MoB was a (semi-)private branch for some time
01:49:56 Join countrymonkey [0] (i=4b05639a@gateway/web/cgi-irc/labb.contactor.se/x-b99951301adb8ac5)
01:50:02LloreanNico_P: Yes, and yet we're using SVN
01:50:06 Join J3TC- [0] (n=jetc123@pool-72-76-187-15.nwrknj.east.verizon.net)
01:50:24Nico_PI used git to develop MoB ;)
01:50:25Lebowski_good grief even the new Sansa View is having Mac USB problems
01:50:49LloreanNico_P: I know. My point is more or less "We don't have to move the central repository to git for people to get the advantages of it.
01:50:49amiconnsafetydan: I really don't get the point in the 'offline operation' argument
01:50:59amiconncvs could not do offline diffs, svn can.
01:51:04Nico_PLlorean: I'm not saying we have to
01:51:15Lloreanamiconn: I think the point is "local revision history and commits"
01:51:27safetydanamiconn: but it can't do offline commits. Which is nice if you're a compulsive saver/committer of changes
01:51:48LloreanI think there's a *very* strong argument for keeping the central repository SVN, since a wide variety of our scripts work off that, and as well it maintains better windows compatibility still.
01:51:52 Join Langly [0] (i=Langly@c-24-21-39-47.hsd1.mn.comcast.net)
01:51:52Nico_Pamiconn: it's a matter of what is cached locally... another read: http://blogs.gnome.org/newren/2007/11/24/local-caching-a-major-distinguishing-difference-between-vcses/
01:52:05amiconnWhile I do understand the need of an occasional offline diff, I cannot understand what offline commits would be good for
01:52:18amiconnI have absolutely no use for that
01:52:48 Quit Lebowski_ ()
01:53:08Nico_PLlorean: the post-commit hook could be changed to a post-psuh hook
01:53:35advcomp2019ok he/she left
01:53:47LloreanNico_P: That doesn't solve windows compatibility issues, though. What's the argument *for* making the central repository use git?
01:53:47Nico_Pbut the windows compatibility argument is compelling, though subject to change
01:54:06LloreanWhat *can't* you do right now using svn-git or whatever it's called, that you could do if the central repository were changed?
01:54:30safetydanamiconn: while working on a feature it sometimes nice to rollback to a previous implementation state without compulsively creating diffs of each change
01:54:41Nico_PLlorean: git-svn adds some limitations, like the need to constantly rebase
01:56:14amiconnThere's one detail that I am mssing in svn compared to cvs, but other than that, I don't need more
01:56:57Nico_PI do want to stress that I'm not trying to push a change of central repo VCS... I was merely following up on a discussion we had earlier. I agree that git has little to offer as a central repo, but it does have very nice features for individual devs
01:57:01amiconnAnd the "better merging capabilites" of git aren't explained? Better in what way?
01:57:06 Quit mud-rb (Read error: 104 (Connection reset by peer))
01:57:14Nico_Pamiconn: no need to track the merges manually
01:57:19amiconn?
01:57:38amiconnI just 'svn up' every now and then
01:57:45Nico_Pin svn you need to know what you've merged and you haven't.
01:57:58Nico_Pmerging is related to branches
01:58:12amiconnEither there is a conflict, or not. If there is one, I can't see how any other scm could magically avoid that
01:58:29amiconnAh, that. Well, we don't use branches
01:58:51safetydanNico_P: the followup post to the on you originally linked gives an idea of how you might change the SCM :) http://keithp.com/blogs/Tyrannical_SCM_selection/
01:59:05amiconnIn case it matters - the cvs feature I am missing is the 'attic' directory
01:59:07Nico_Pamiconn: we don't release either... that doesn't mean we shouldn't ;)
01:59:56 Quit countrymonkey ("CGI:IRC (Ping timeout)")
02:00
02:00:37Nico_Psafetydan: nice, thanks :)
02:01:21LloreanIt seems to me the best plan is just to put up a wiki page and list the Pros and Cons of each VCS as they come up
02:01:23safetydanLlorean: Java has the potential to be faster than C/C++ with a JIT and runtime profiling
02:02:03Nico_PLlorean: we have one about git ;)
02:02:49LloreanNico_P: I meant more of a side by side of SVN and Git. A list of Pros ("things it has that the other doesn't") and Cons ("weaknesses it has that the other doesn't, but don't qualify as real features that the other has")
02:03:52Lloreansafetydan: You're saying that a program can be written in JAVA in such a way that it's impossible for a C program to do the same thing, only faster?
02:03:55Nico_Pyeah, quite a good idea
02:05:18Lloreansafetydan: I don't disagree that JAVA could be faster than C under certain circumstances, but mean to imply that in terms of absolute speed, the C version of a program should always have the potential to be made in such a way as to be faster than the JAVA version on the same hardware (assuming it's not that theoretical native JAVA cpu or other such chicanery)
02:06:18safetydanLlorean: it's a subtle point. Basically C code can only be optimised for one case (say -mcore2-duo). Since Java has an intermediate representation it can be optimised for whatever target your code happens to be running on.
02:06:32LloreanYes. I don't disagree with that.
02:07:47 Join mokkurkalve [0] (n=eivind@084202137013.customer.alfanett.no)
02:07:47safetydanLlorean: for example http://shootout.alioth.debian.org/gp4/benchmark.php?test=all&lang=gcc&lang2=java
02:08:00safetydanbut the embedded space is a different kettle of fish and Java isn't particularly suited for that
02:08:14*safetydan waits for the "offtopic" call :)
02:08:19Lloreansafetydan: Does the edit to my post suit you then? Or should I elucidate a bit more for their sake?
02:10:21safetydanLlorean: nah, that's plenty. The only other point is that Java is not an acronym
02:11:32LloreanAh, yes
02:11:36safetydanon a slightly more on-topic note, the iPod's are all clocked at 80 MHz at the moment aren't they?
02:11:44LloreanYes
02:11:58LloreanI always want to type it JAVA because I lived one island over from Java, and it differentiates them in my mind.
02:12:47safetydanFair point.
02:13:50LloreanBut java also works, and looks less acronymy
02:17:58 Quit Soap (Read error: 110 (Connection timed out))
02:18:14amiconnsafetydan: All PP targets use 30MHz and 80MHz for cpu scaling.
02:18:14 Quit Nico_P (Read error: 104 (Connection reset by peer))
02:20:28safetydanamiconn: the cpu supports up to 100 MHz doesn't it? Any reason we don't allow boosting that high?
02:20:33safetydanapart from heat
02:20:44amiconnNo it does not, or rather, we don't know
02:20:47Calcipherdamnit, I forgot the syntax for providing tts options, to specify a voice to use
02:21:08Calcipherwhen using configure
02:21:18amiconnPP5020 (ipod G4, color, mini G1, iriver H10) supports up to 80MHz according to the product brief
02:21:37Calcipherwas it "voice:"VW Kate"
02:21:40 Join Soap [0] (n=Soap@rockbox/staff/soap)
02:21:45amiconnPP5022 (ipod mini G2 (only!), Sansa c200) supports up to 100MHz
02:22:04amiconnipod Nano and Video use PP5021 for which there is no product brief
02:22:05safetydanah, so it's only one cpu that might support up to 100 MHz then
02:22:06DerPapstnight all
02:22:15 Quit DerPapst ("So Long And Thanks For All The Fish!")
02:22:54*karashata randomly ran across this: http://web.tagus.ist.utl.pt/~rui.neves/se2005/basedoIpod_5002_brief_0108_Public.pdf
02:23:03amiconnThe PP5021 reports itself as a PP5022C in software, so I think it's a PP5022 that doesn't reach the full PP5022 specs
02:23:18amiconnFor safety, we assume it allows up to 80MHz
02:24:15amiconnAlso, the PP5022 might cause heat problems in the tiny Mini G2 and especially Sansa c200 cases when running at 100MHz
02:24:39safetydanmakes sense. I'm assuming the original firmware never clocked them that high?
02:25:51amiconnThe PP5002, however, is specced 90MHz, and (1) our PP5002 targets don't have tiny cases, plus (2) we don't have power consumption problems anymore on PP5002, unlike on PP502x
02:26:13amiconnSo to compensate for the broken cache, we could go 90MHz on PP5002
02:27:05 Quit radinp (Read error: 110 (Connection timed out))
02:28:15amiconnAll PP targets of which I have analysed the PLL setup after bootup without rockbox intervention boot either at 24MHz (the PP default clock - ipods), or 80MHz (iriver H10, Sansas)
02:28:39CalcipherI'm recieving the following error when trying to "make voice" using a current build updated from svn, /bin/sh/ -c: line 1: Syntax error near unexpected token '<<<'
02:28:44amiconnAnd the sansas have PP5022 (c200) resp. PP5024 (e200)
02:30:05amiconnAll that said, I once tried 96MHz on my mini G2. It was perfectly stable (or rather, no more unstable than usually) even with a lot of the stuff enabled which I don't normally use (EQ etc)
02:30:32Calcipherfollowed by /bin/sh/ -c: line 1: '<<<<<.mine' then make: *** [voice] error 2
02:31:10amiconnCalcipher: You have conflict markers in your working copy
02:31:17 Join psycho_maniac [0] (i=psycho_m@ppp529.hk.centurytel.net)
02:32:04CalcipherI have no idea what that means, or how to resolve that issue
02:32:38amiconnI think its better to keep the general 80MHz limit for PP502x. Going higher would make the power consumption problems even worse, while it would not solve performance problems in general, since not all PP502x targets could use a higher clock
02:34:21amiconnWe're already better than ipl, which runs PP502x at 75MHz, and PP5002 at only 66MHz, even if 75MHz are intended (sic!)
02:35:39LloreanI think at this point we really should be able to overcome almost all our performance problems without clock increases anyway, right?
02:35:48LloreanWell, with the exception of perhaps monkey's audio, which doesn't count
02:36:33Calcipheranyone have an idea of how I can resolve my compiling issue?
02:36:59CalcipherI don't know what went wrong
02:39:55amiconnkarashata: That's the PP5002 product brief which is (or was) publicly available from portalplayer, as well as the PP5020 and PP5022 product briefs
02:40:19amiconnUnfortunately they don't tell much apart from very general stuff
02:40:53*karashata nods slightly
02:40:58karashataI figured that much
02:41:30Calcipherneed to get rid of that espeak voice before i go to bed
02:41:41 Join raulh [0] (n=raul@71-37-217-131.phnx.qwest.net)
02:41:54raulhhello everybobody i was wondering..
02:42:04raulhcan you turn ipod into linux using linux??
02:42:08Calciphersomeone please help!, should I dump the entire source dir? and get the source again
02:42:18Lloreanraulh: This is #Rockbox. Perhaps you meant to join #ipodlinux?
02:42:24***Saving seen data "./dancer.seen"
02:42:26raulhnooo
02:42:52raulhall i mean is rockbox linux transferable on ubuntu
02:43:00scorcherockbox isnt linux
02:43:34raulhwhat is it -_-
02:43:35scorchedid you mean to ask if you can use your ipod under linux with rockbox?
02:43:41raulhyes
02:44:01scorchethen you can
02:44:04Lloreanraulh: Rockbox is from scratch.
02:44:26Calcipherconflicting makers in working copy, so what should I do?
02:44:43raulhoh.. well ipodlinuyx.org there server dosn';t hve ther downloads anymore..
02:45:02raulhis this rockbox just as good?
02:45:11raulhbecuase i haaaatteee apple.. and windows..
02:45:33Lloreanraulh: I would suggest the manual.
02:45:43*Calcipher hears the crickets in the distance
02:45:46safetydanCalcipher: try "svn revert -R ." in your source directoyr
02:45:58safetydanyou'll lose any changes you've made to the source though
02:47:44Calcipheroh, so maybe I screwed up the source sometime long before and had never noticed, oh no, there is an edit I made to the configure file, to let it allow me to input tts voice options with spaces, for specifying voices like "VW Kate", because otherwise it would fail
02:47:50 Quit DrCrow ("Leaving")
02:48:00Calcipherthats the only edit to the source that I care about
02:48:28safetydanCalcipher: it's not likely something you've done, just svn merge issues on update. The revert command will force SVN to restore everything to a pristine state
02:49:20raulhok well llorean can i install roxbox with linux os? as in (i'm using ubuntu)
02:49:29Calcipheroh ok, so I will save a back up of my edited configure file, so I can compare with the new one, I think I remember what lines it was
02:49:44amiconnCalcipher: YOu don't need to change 'configure' in order to use sapi voices with spaces in the name. I do this all the time
02:49:53scorcheraulh: there are linux instructions in the manual..
02:49:54Lloreanraulh: Please see the manual. There wouldn't be Linux installation instructions in it if you couldn't...
02:50:08amiconnJust put the whole argument into single (!) quotes, i..
02:50:29amiconni.e. in your case: '/voice:VW Kate'
02:51:32Calcipherok, I will try the svn revert, then the single quotes technique, thanks amiconn and safetydan
02:52:33psycho_maniacjust to make sure Calcipher its "svn revert -R ." everything in that quotes. i know some people have messed that up. including me
02:55:14Calcipheryou werent kidding psycho_maniac, I was trying -r and was getting a syntax error
02:56:21raulhgaaaaah
02:56:40raulhanyone know how to find a mounting point with linux?>ubuntu
02:56:41safetydanCalcipher: yeah that's one case where -r/-R are different :)
02:56:55safetydanraulh: type mount and it will list all mounted devices
02:57:01raulhty
02:57:57Calcipherwhere am I going wrong
02:58:35psycho_maniacwhat do you mean?
02:58:38Calcipheroh, I see, I have to run "make" then "make zip"
02:59:25CalcipherI was getting some error about apps dir not existing and other things, because after setting configure for normal build I tried make zip
02:59:43Calcipherwithout first doing make, which I have running now
03:00
03:00:49Calcipherlooks like things should go smoothly from here, I remember the process more now, I had only really done it once a few weeks ago
03:01:36CalcipherI'm going to retry these steps in vmware debian, since I got rid of zonealarm
03:04:36raulhLlorean, thanks for your help, and scorche thank you too!
03:07:19raulhuumm.. one more last question guys...
03:07:33raulhi just manually installed it and am i done? if not what else i need to do
03:09:18scorcheraulh: the manual should tell you everything you need to know :)
03:10:13raulhok nvm i did thanks for all ur help bye
03:10:25raulhneed get shuteye for school...
03:11:19raulhomg its so nice! bye!!
03:11:22 Part raulh
03:16:47 Quit Billenium ("Ex-Chat")
03:18:09 Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net)
03:20:33Calciphergreat, its working perfectly in cygwin, thanks safetydan and amiconn
03:23:07 Quit pill ("changing servers")
03:23:28 Part pixelma
03:23:30 Quit marcosource (Remote closed the connection)
03:23:34 Join marcosource [0] (i=marco@cakebox.net)
03:24:44 Join raulh [0] (n=raul@71-37-217-131.phnx.qwest.net)
03:25:07raulhnow i feel dumb.. how do i add music files to my ipod rockbox?
03:25:21raulhi cannot find any info on it on the manual...pdf
03:25:54 Quit XavierGr (Read error: 110 (Connection timed out))
03:26:57safetydanraulh: make a folder on your iPod called "music" and copy your songs there
03:27:07safetydanit's really up to you how you do it, but that's the general idea
03:27:20 Join radinp [0] (n=pradin@psyduck-08.dynamic.rpi.edu)
03:27:28raulhthnx wouldn't that make x2 as much files???
03:27:58safetydanwhy?
03:28:30raulhbecuz i got a 2gb lol..
03:28:31safetydanunless you're copying them twice I can't see why there would be twice as many files
03:29:20raulheh, ok thanks
03:39:23 Join alienbiker99 [0] (n=alienbik@ool-44c126d4.dyn.optonline.net)
03:46:51raulhlol sooo many games..
03:47:16 Join bonbonthejon [0] (n=jon@69.61.203.3)
03:47:22 Nick raulh is now known as rockboxpwns (n=raul@71-37-217-131.phnx.qwest.net)
03:51:31 Quit bonbonthejon ("Leaving.")
03:53:49 Join bonbonthejon [0] (n=jon@69.61.203.3)
03:56:34 Quit EspeonEefi ("さよなら")
03:56:40 Nick rockboxpwns is now known as linuxftw (n=raul@71-37-217-131.phnx.qwest.net)
03:56:46 Part linuxftw
03:58:26 Join parafin|sleep [0] (i=parafin@parafin.dialup.corbina.ru)
04:00
04:10:05 Quit parafin|away (Read error: 113 (No route to host))
04:10:39 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
04:10:47 Part radinp
04:16:27 Quit Thundercloud (Remote closed the connection)
04:26:19 Join billenium [0] (n=billeniu@c-69-249-243-110.hsd1.pa.comcast.net)
04:31:47 Quit JdGordon (Remote closed the connection)
04:39:57 Quit psycho_maniac (" bye")
04:40:15 Part hcs
04:41:26 Part bonbonthejon
04:42:28***Saving seen data "./dancer.seen"
04:42:56 Join iwantanimac [0] (n=iwantani@124-169-19-238.dyn.iinet.net.au)
04:49:06 Join animeloe [0] (n=animeloe@unaffiliated/animeloe)
04:51:46 Quit iwantanimac ("iMac, out.")
04:58:30 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
04:58:57 Join miepchen^schlaf [0] (n=hihi@p54BF5406.dip.t-dialin.net)
05:00
05:08:06 Quit animeloe ("Leaving")
05:10:10 Join animeloe [0] (n=animeloe@ool-44c2be34.dyn.optonline.net)
05:10:12 Join Veinor [0] (n=veinor@wikipedia/Veinor)
05:10:28 Part Veinor
05:11:24Lloreansafetydan: I really don't understand why the frequently suggested solution to "it runs to slow" is "speed up the processor." I remember on guy came around and started running his iPod video at 90mhz. Of course, being a bit nutty, he raised the unboosted speed to 60 for absolutely no known reason.
05:14:23safetydanLlorean: tweaking a couple of values to up the clock rate is much easier than having to think about how to make something genuinely faster
05:14:50safetydanwhat irks me is that there's good code in iboy, but none of the rockboy fans seem willing to work to merge that code in
05:16:04LloreanMaybe they don't know?
05:16:31LloreanI heard iBoy was faster, but that seemed vague, and I also heard it doesn't have sound, which suggested to me at least that maybe it was faster for that very reason
05:17:25safetydanthe wiki page for iboy says it supports sound
05:17:42 Quit animeloe ("Leaving")
05:17:46LloreanWell, there's hearsay for yah
05:19:01LloreanThere was a sound sync patch in the tracker that actually made our Rockboy's sound quite good, but somehow what got accepted didn't work out the same
05:20:09safetydanhrm, the ipodlinux forums seem to indicate they're about the same speed
05:20:46 Join animeloe [0] (n=animeloe@unaffiliated/animeloe)
05:20:50LloreanI imagine they're optimized in different areas though, so some cross pollination couldn't hurt?
05:21:06safetydanand then I see posts saying iboy is much faster
05:21:07 Quit homielowe (Read error: 110 (Connection timed out))
05:22:55safetydan" i dont even use rockbox(i dont know why i even put it on my ipod, cuz all it has is doom and iboy and music playback "
05:23:16LloreanMaybe we need a slogan.
05:23:21safetydanI like how music playback came last and with such a disdainful tone.
05:23:52LloreanSomething like "Rockbox. It Rocks the Box. With Music. Because that's what it's for. Music. If you don't want good Music features, don't install it. For real."
05:24:15ShaidWouldn't 'Rockbox. It's for music' just do the job?
05:24:35scorchesurely not
05:24:41Shaidthen whenever anyone asks about other features you just say 'it's for music'
05:26:44safetydanman, reading the ipl forums makes me appreciate the heavy moderation on ours all the more
05:29:38LloreanI kinda wish we didn't need to moderate like we do. It's wearing on me, and tends to result in me crossing the line from time to time, but it kinda seems like you have to choose between "too much" and "too little" sometimes
05:29:56 Quit Langly ("I dont have anything to say, thats why I'm quitting in the first place.")
05:30:40safetydanI imagine it's hard being the bad guy all the time.
05:31:52LloreanNot my favorite duty ever, but it kinda feels like if we I ever stop I'm either thrusting it on someone else, or I'm going to have a mess to clean up if I ever start again
05:41:55 Join animeloe_ [0] (n=animeloe@unaffiliated/animeloe)
05:45:24 Join Langly [0] (i=Langly@c-24-21-39-47.hsd1.mn.comcast.net)
05:50:43 Quit karashata ("I'm a fluffy dragon, anyone who gets in my way will be snuggled to death!")
05:52:05 Join webguest02 [0] (i=cbdc7c7e@gateway/web/cgi-irc/labb.contactor.se/x-9d9091806a3c42ab)
05:53:19 Quit webguest02 (Client Quit)
05:58:14 Quit animeloe (Read error: 110 (Connection timed out))
06:00
06:06:08 Quit Calcipher (Read error: 104 (Connection reset by peer))
06:12:01 Join pradin [0] (n=pradin@psyduck-08.dynamic.rpi.edu)
06:14:43 Join b_eb [0] (i=44cc4bdc@gateway/web/cgi-irc/labb.contactor.se/x-133b6c44a3d37e60)
06:15:27 Quit b_eb (Client Quit)
06:15:30 Join b_ebo [0] (i=44cc4bdc@gateway/web/cgi-irc/labb.contactor.se/x-8e4ff1893a0850fa)
06:16:13 Quit axionix (Read error: 104 (Connection reset by peer))
06:16:24b_eboI'm having a problem after installing rockbox. windows comes up with a new hardware found wizard asking me to install drivers. if I boot the ipod in ipod mode it doesnt complain
06:16:44b_ebothe device it finds is "rockbox media player"
06:18:30b_ebocant seem to find any information in the forums
06:19:02Mouser_Xb_ebo: Did you read the manual?
06:19:19Mouser_XThat's Rockbox's USB mode, which currently is only good for charging.
06:19:21b_eboyes.. the ipod works fine
06:19:28b_eboah
06:19:49b_ebook, so its fine then. everything is ok, I just have to boot into disc mode then?
06:19:55Mouser_XIn other words, when you want to charge within Rockbox, that's what will come up. If you want to actually transfer files, you'll need to use the OF.
06:20:07b_ebogotcha
06:20:41b_ebothanks, that answers my questions. I read where it said it was not functional, but it didnt mention anything about hardware manager
06:20:59Mouser_XOf course not. You're supposed to ignore that.
06:22:19Mouser_XI know this is mentioned, but it might only be mentioned in the MajorChanges wiki page, or the commit itself. I'm not entirely sure.
06:22:20 Quit b_ebo ("CGI:IRC (EOF)")
06:33:21 Join james01 [0] (n=james@EV-ESR-74-215-26-69.fuse.net)
06:38:55 Quit pradin (Connection timed out)
06:42:29***Saving seen data "./dancer.seen"
06:42:43 Join axionix [0] (n=axion@cpe-74-70-239-192.nycap.res.rr.com)
06:56:16 Join mud-rb [0] (n=mud@dialup-4.156.9.217.Dial1.Boston1.Level3.net)
07:00
07:08:09 Join karashata [0] (n=karashat@207.61.208.119)
07:14:21 Quit animeloe_ ("This computer has gone to sleep")
07:27:53 Join woodensoul [0] (n=woodenso@72.86.92.80)
07:28:45 Part safetydan
07:28:55 Join Jon-Kha [0] (i=jon-kha@80-248-247-190.cust.suomicom.fi)
07:29:50woodensoulDon't laugh, but I'm making a WPS for the iRiver H120/H140 with album art. Should I go ahead and have irfanview create grayscale .bmps or just let rockbox convert them?
07:32:04 Join pradin [0] (n=pradin@psyduck-08.dynamic.rpi.edu)
07:37:08 Join mud-rb_ [0] (n=mud@dialup-4.156.222.22.Dial1.Boston1.Level3.net)
07:46:36 Join mud-rb__ [0] (n=mud@dialup-4.156.222.24.Dial1.Boston1.Level3.net)
07:46:51 Quit mud-rb (Nick collision from services.)
07:46:59 Nick mud-rb__ is now known as mud-rb (n=mud@dialup-4.156.222.24.Dial1.Boston1.Level3.net)
07:47:36DarkStylistisnt xm alot like the mod format?
07:47:51DarkStylistor s3m?
07:54:55 Quit TMM ("Ex-Chat")
08:00
08:06:56 Quit mud-rb_ (Read error: 110 (Connection timed out))
08:07:16 Quit mud-rb (Remote closed the connection)
08:07:59 Join GodEater_ [0] (n=bryan@rockbox/staff/GodEater)
08:09:37 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
08:14:18 Join animeloe [0] (n=animeloe@unaffiliated/animeloe)
08:15:06 Join j0tt [0] (n=j@unaffiliated/jott)
08:24:00 Join animeloe_ [0] (n=animeloe@unaffiliated/animeloe)
08:27:33 Quit jott (Read error: 110 (Connection timed out))
08:31:40 Quit animeloe (Read error: 113 (No route to host))
08:33:55 Join gromit` [0] (i=gromit@ras75-5-82-234-244-69.fbx.proxad.net)
08:33:57 Quit Llorean (Read error: 104 (Connection reset by peer))
08:37:35 Join Llorean [0] (n=llorean@cpe-70-113-103-34.austin.res.rr.com)
08:38:18 Quit woodensoul ()
08:40:17 Quit gromit`` (Read error: 110 (Connection timed out))
08:41:19 Join Rob2222 [0] (n=Miranda@p54B15F49.dip.t-dialin.net)
08:41:53 Part pradin
08:42:32***Saving seen data "./dancer.seen"
08:44:35 Quit BigBambi (Read error: 110 (Connection timed out))
08:50:04 Join LinusN [0] (i=linus@rockbox/developer/LinusN)
08:52:22 Quit jhulst ("Konversation terminated!")
08:55:32 Quit idnar (Nick collision from services.)
08:55:34 Join idnar_ [0] (i=mithrand@unaffiliated/idnar)
08:57:15 Join safetydan [0] (n=safetyda@rockbox/developer/safetydan)
08:59:33 Join advcomp2019_ [0] (n=advcomp2@unaffiliated/advcomp2019)
08:59:43 Quit Rob222241 (Read error: 110 (Connection timed out))
08:59:47 Quit advcomp2019 (Nick collision from services.)
08:59:50 Join JdGordon [0] (n=jonno@rockbox/developer/JdGordon)
08:59:53 Nick advcomp2019_ is now known as advcomp2019 (n=advcomp2@unaffiliated/advcomp2019)
09:00
09:03:56 Join petur [0] (n=petur@rockbox/developer/petur)
09:07:04 Join Darksaboteur [0] (n=chatzill@75.73.87.203.static.nsw.chariot.net.au)
09:07:10 Quit Darksaboteur (Client Quit)
09:08:09 Quit miepchen^schlaf (Read error: 110 (Connection timed out))
09:11:43 Join ender` [0] (i=krneki@84-255-206-8.static.t-2.net)
09:14:39 Quit gromit` ("Coyote finally caught me")
09:17:46 Quit Mouser_X (Read error: 104 (Connection reset by peer))
09:18:31 Join Zagor [0] (n=bjst@46.35.227.87.static.tab.siw.siwnet.net)
09:19:32*Zagor is getting seriously annoyed by all the spurious PM requests in the tracker
09:19:51LloreanPM requests?
09:20:13Zagorproject manager request
09:20:27Zagor"please reopen this bug, I want more features" etc.
09:21:04ZagorI think I'll limit it so only developers can submit such requests
09:21:14 Join homielowe [0] (n=chatzill@d207-81-67-190.bchsia.telus.net)
09:21:45LloreanOnce or twice a reopen request has been useful because the closer misread the task, but yeah. People can just open a new task for those, perhaps
09:22:20LloreanCan you get rid of that silly Vote thing?
09:22:22rasherZagor: can't devs reopen tasks anyway?
09:22:30markunthey can always ask in here for us to reopen it
09:22:47Zagorrasher: yeah I think you're right
09:32:37 Join mrkiko [0] (n=mrkiko@host169-98-static.34-88-b.business.telecomitalia.it)
09:32:42mrkikoHi all! I'm reading GPL V3.
09:32:53GodEater_ooh - light bedtime reading if ever there was some
09:33:04mrkiko:)
09:33:45mrkikowhat is the problem with GPLV.3 and rockbox? I think I understood it, but here stands people which probably understood it better...
09:34:10Bagdermrkiko: v3 is incompatible with v2
09:35:12mrkikobut probably you would change the rockbox' license if you don't have good motivations in not doing so...
09:35:18mrkikoright?
09:35:35GodEater_that makes no sense
09:35:35mrkikoI'm trying to catch those motivations...
09:35:54rashermrkiko: getting the permission of every copyright holder isn't easy.
09:36:18GodEater_I also seem to recall we're split roughtly 50/50 on those who like the v2 license, and those who like the v3 license
09:36:22mrkikoeheh... right :
09:36:34markunalthough we agreed that people who change just a few lines are not the copyright holders of those files
09:37:01markunGodEater_: and then there is the group who doesn't really care about either
09:37:07GodEater_that too ;)
09:37:10JdGordondidnt we decide rockbox hadnt definatly said v2 only so we could go v2 or later/v3 without a problem?
09:37:51rasherJdGordon: I really disagree with that. The headers point to a specific file, containing a specific version. I don't see how that's not definite.
09:38:04 Quit karashata ("I'm a fluffy dragon, anyone who gets in my way will be snuggled to death!")
09:38:14mrkiko... any way guys: I would like to respect your decisions. Would it be intereginst in your opinion to ask, with the rockbox forum may be, if one exist, what people holding those copyrights thinks?
09:38:20GodEater_so no, we never decided ;)
09:38:28markundo we use any code from other projects (codecs, plugins) which is v2 only?
09:38:36Lloreanrasher: The file they point to says "v2 or later" though, I thought
09:38:37GodEater_mrkiko: most people who hold the copyrights don't read the forums
09:39:19mrkikohmh... I expected this.
09:39:45mrkiko... aniway, if we want, probably we can get theyr opinion without being "pedantic"...
09:39:46Bagdermarkun: we have linux code
09:39:49mrkiko... ??
09:40:09rasherLlorean: No. It contains the text "If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation."
09:40:33rasherBut I maintain that pointing to docs/COPYING which is v2, does count as specifying a version
09:40:54 Quit JdGordon (Remote closed the connection)
09:41:03Zagorand the question is what the definition of "The Program" is. if it's the whole archive, it definitely says v2. but each specific source file doesn't say that.
09:41:08GodEater_mrkiko: why do you care which version Rockbox uses ?
09:41:26DarkStylistisnt rockbox based on linux?
09:41:31rasherZagor: it says "the file COPYING" though, which is v2...
09:41:42rasherDarkStylist: No.
09:41:49mrkikomhm...
09:41:56 Join linuxstb [0] (n=linuxstb@rockbox/developer/linuxstb)
09:41:59mrkikofor mainly two reasons.
09:42:03ZagorI asked FSF this during their big irc q&a session, but they didn't want to answer in public... :-)
09:42:06DarkStylisthas it something in common with another operating system?
09:42:10LloreanZagor: Also, if the file that proclaims "v2" is the "v2" license that also says "or any later version", I don't see how it can work otherwise, unless you have the GPLv2 license included (as the mandate requires you do) while somehow stripping it of all mention of it being version 2?
09:42:14GodEater_DarkStylist: no
09:42:19DarkStylistoh ok
09:42:21mrkiko1 - I red the GPL V 3 and It doesn't seems to me so different from V2 in many important aspects, but this is an opinion.
09:42:27mrkikoanother motivation is the rockbox-espeak problems...
09:42:29rasherLlorean: the v2 doesn't say "or any later version"
09:42:48Lloreanrasher: But the COPYING file included with our source does, I thought.
09:42:57rasherLlorean: Where?
09:42:58linuxstbrasher: How do you interpret the sentence in section 9 of COPYING then? "If the Program does not specify a version number of this License, you may choose any version ever published by the Free Software Foundation."
09:43:21Zagorlinuxstb: but the program specifies v2, by specifically including the v2 license
09:43:30rasherExactly.
09:43:34LloreanZagor: But how can you do v2 or later then, ever?
09:43:38linuxstbZagor: Then in what case would section 9 apply?
09:43:46rasherLlorean: by specifying "v2 or later" in the header files?
09:44:04rasherHeader files.
09:44:07ZagorLlorean: this is not a clear-cut situation. valid arguments exist for both viewpoints.
09:44:21rasherOr "v2 as included in COPYING or any later version"
09:44:22linuxstbZagor: And it doesn't specify "v2" at all - it specifies that the license is in "COPYING", and that license says that any version of the GPL can also be used.
09:44:26GodEater_I can't believe we're having this discussion again =/
09:44:27 Join pixelma [0] (i=pixelma@rockbox/staff/pixelma)
09:44:49Zagorlinuxstb: what is "it"? "the program"? if so, first you must define "the program"...
09:44:51mrkikoGodEater_: just my fault! :)
09:44:52rasherlinuxstb: But only if no version has been specified - and it has, since it specifies "the license in COPYING", which is v2
09:45:15linuxstbZagor: Yes, "it" is all source files with the standard Rockbox header.
09:45:46Zagorlinuxstb: why only the source files? why not the entire source archive, which includes the v2 license? you see how this is not an obvious situation...
09:46:18rasherI really don't understand how you can argue that no version has been specified. The header clearly states that the file is under the license in COPYING, which is *specifically* version 2
09:46:30linuxstbrasher: Yes, and "v2" of the GPL has a clause that if the program doesn't specify a license, any version can be used. The License is definitely not part of the program IMO.
09:46:34mrkikoOk: I'll follow the lesson now...
09:46:52rasherlinuxstb: But the program *does* specify a version.
09:46:57mrkikoSorry to all those annoied people which doesn't want talk about licenses
09:47:03rasherlinuxstb: That is, the version that is in COPYING, being v2
09:47:06Zagormrkiko: :-)
09:47:17krazykitmrkiko, it is an important topic, though.
09:47:25Bagdermrkiko: we're used to this subject, we have it repeatedly ;-)
09:47:37GodEater_aaaaaahhh! Recursive argument. stkov!
09:48:39Bagderhttp://daniel.haxx.se/blog/2007/12/05/i-solved-the-ams-sansas-firmware-checksum-puzzle/
09:48:57 Join dryahetzeph [0] (n=HydraPhe@209-62-215-8.mobiletel.com)
09:48:59 Nick dryahetzeph is now known as hydrapheetz (n=HydraPhe@209-62-215-8.mobiletel.com)
09:49:42GodEater_good effort at changing the subject Bagder :)
09:49:49Bagderhaha
09:49:52linuxstbrasher: The license is not part of the program though.
09:50:02rasherBagder: hurray!
09:51:26rasherlinuxstb: the files do not exist in a vacuum. What if the header had simply said "see COPYING for copyright details.". Would it then be impossible to use the files, since they don't explicitly specify a license individually?
09:52:48rasherI think using this loophole is a very dodgy at best, and an underhanded way to change licensing.
09:53:10linuxstbI don't think it's underhand, it's pretty clear in the license text.
09:53:28rasherEven if it may just be technically correct, I think it'd be morally wrong to use it.
09:54:51safetydanIsn't one of the objections to GPL V3 the fact that it places additional restrictions on potential users of the source? THe anti-tivoisation clause thingy.
09:55:20rasherlinuxstb: it is not in any way clear.
09:55:43linuxstbrasher: In what way?
09:55:49GodEater_safetydan: that's my objection to it certainly - but it's why linuxstb likes it :)
09:56:00rasherlinuxstb: about 42
09:56:05linuxstbsafetydan: I believe so - but then it comes down to a matter of philosophy, so there's no point arguing... ;)
09:57:13safetydanBSD and be done with it I say! :)
09:57:30rasherBagder: so, could you run code on the thing? (if you owned one)
09:58:07mrkikoaniway...
10:00
10:00:36mrkikoThe problem was relative to adopting espeak on rockbox. Without real-time speech on rockbox, no real accessibility will be possible. The work to do then, if you solve the problem, would be aniway VERY MUCH... but it may be aniway interesting...
10:01:46krazykitmrkiko, espeak built from when it was GPLv2 could be used, though.
10:01:46markunit would be a very nice feature, but I don't agree that rockbox isn't really accessible without it
10:02:36mrkikomarkun: but in how many cases you may be able to select a directory ?
10:03:20mrkikomarkun: in how many cases you will need to search simply a file?
10:03:29 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
10:03:37rashermrkiko: have you used Rockbox with voice and .talk clips?
10:14:15pondlifeYes, I'd think it's pretty much accessible without espeak...
10:14:31markunbut would take a lot less work with it
10:14:57pondlifeI think I'd prefer to use pregenerated voices anyway; espeak sounds worse IMHO.
10:16:12GodEater_surely we'd leave that as an option anyway ?
10:16:23GodEater_I know you windows boys like your SAPI...
10:16:55pixelmaI use espeak as a SAPI voice =)
10:17:32pondlifeWe would, of course. But I'd still like to know what real-time would bring that would be so useful. Reading out a book in text format is one I can come up with.
10:17:34pixelma(just for the only known (to me) free possibility to have a swedish voice)
10:18:22linuxstbpondlife: I would think it would be more convenient (for filenames and, if someone does it, tags) than having to pre-generate clips.
10:18:42 Join sanlist [0] (i=c27f0811@gateway/web/cgi-irc/labb.contactor.se/x-2e47169fc4ad0cf5)
10:19:12pondlifeHmm, I suppose so. It's not exactly hard to run Voicebox though..
10:19:31sanlistSHould the latest commit by Mark Arigo improve runtime on sansas?
10:19:40pondlifeWould be neat for tags.
10:20:51rasherpondlife: clips for tags can be generated from the database files though. See FS #7984
10:22:22 Join lee-qid [0] (n=liqid@p549674EE.dip.t-dialin.net)
10:22:41linuxstbCan rbutil create .talk clips?
10:22:42rasherIt might not apply anymore due to recent changes in voice.pl, but should be trivial
10:22:59rasherlinuxstb: I believe it can, but it requires a bit of work on Windows
10:23:40rasherIdeally it'd interface with sapi directly, but I think it currently requires make_voice.vbs
10:24:36pixelmalinuxstb: I think it could but wasn't adapted for speex yet (IIRC)
10:24:48rasherAh.
10:25:07*linuxstb reads the rbutil changelog and sees the SVN version can now use rbspeexenc (but not the binary release)
10:25:12pixelmaand I never tried...
10:26:54pixelmalinuxstb: thanks for pointing that out :)
10:27:08linuxstbWould anyone disagree with renaming the current RockboxUtility wiki page to something like RockboxUtilityOld and renaming RockboxUtilityQt to RockboxUtility?
10:27:28rasherSeems reasonable.
10:27:30BagderI'm in favour of that
10:27:37*linuxstb politely requests that Bagder makes it so...
10:27:42Bagderthe 'qt' suffix is mighty weird
10:27:52linuxstbI'll update the links in the manual.
10:29:02Bagderdone!
10:29:49linuxstbmanual updated too.
10:30:37markunsanlist: it sounds like it
10:30:53rashermight make sense to trigger a manual rebuild
10:31:33linuxstbYes, the link on the RockboxUtilityQt error page to Rockboxutility isn't very obvious...
10:31:53Bagderhehe
10:32:17linuxstbThose twiki developers didn't pay much thought to the UI... Or is it just our skin?
10:32:30 Quit mrkiko (Remote closed the connection)
10:33:05sanlistmarkun: was it a response to a known bug? Or just a result of code inspection?
10:33:09BagderI created a new *qt page now with nothing but a pointer
10:34:05sanlistmarkun: there's not FS mentioned in the commit message, that's why I'm asking. But I also understand that you can't know for sure.
10:34:47pixelmamarkun, sanlist: I also hope it makes the radio support more stable - sometimes on my c200 the frequencies were all off (a reboot often helped). Will try in a bit
10:36:30Zagorpixelma: you are lucky. my c200 radio is a lot more troublesome. the 1sec-delay-per-0.05Mhz is just a start...
10:37:02LloreanMy c200 radio seems to work fine except that my headphone jack isn't firmly connected and requires a little wiggling
10:37:07markunsanlist: I don't even know who Mark Arigo is. My guess was only based on the commit.
10:37:12*Llorean got lucky in the c200 lottery, apparently
10:37:24sanlistmarkun: so was mine :-)
10:37:30pixelmawell sometimes I need to use the radio in the original firmware shortly to fix that
10:37:39pixelmamarkun: low_light
10:38:19 Join bughunter2 [0] (n=Administ@ip565fbeaa.direct-adsl.nl)
10:38:50markunpixelma: ah, then I know him :)
10:39:05pixelmaLlorean: have you ever tried using the radio after listening to music (files) before? That's when the radio gets "confused" most often on mine, when starting the radio right after boot there are no problems
10:40:20Lloreanpixelma: I think only once
10:42:23bughunter2hey guys, yesterday i've been discussing if the SanDisk E280 would be any good. it works with rockbox too. But after reading a review and a bunch of comments, i've found out they say the player freezes all the time. I think this does not apply when rockbox is installed?
10:42:35***Saving seen data "./dancer.seen"
10:44:02Bagderbughunter2: it doesn't apply to me at all, OF nor rockbox
10:44:14bughunter2hmm, you have the E280 exact?
10:44:28Bagderno, but the e200 series are all the same except the nand flash size
10:44:41bughunter2some people say it freezes after 4 days or sometimes 2 weeks. also, SanDisk doesn't create replacement USB cables? is it some special USB cable?
10:45:10Bagderit's a custom one, yes
10:45:16Bagderand a custom plug in the player
10:45:24bughunter2too bad there is no replacement?
10:45:53bughunter2any who, E280 seems a good choice still? rockbox works with the mechanical scroll wheel too?
10:46:31GodEater_wouldn't be much use if it didn't....
10:46:51bughunter2well yes
10:47:23bughunter2Bagder: can you recommend the SanDisk player?
10:47:51Bagderyes, at least the v1 so far
10:47:59linuxstbpreglow: What made you think that Lear wote the ac3 metadata parser? My memory is telling me I wrote it...
10:48:06bughunter2okay thanks
10:55:01pixelmahmm... should a warning about new/other generations of players (like Nano G2/G3, the Sansa v2 players, maybe Ipod Classic) be also in the manual (part of the installation instruction along with the other notes)?
10:55:39bughunter2doesn't rockbox work on Sansa v2? any chance that i'll get a v2 when buying sansa e280 now?
10:55:51linuxstbpixelma: I think so, yes. Preferably with a way to distinguish them.
10:56:06Bagderbughunter2: if you buy a new, you will most likely get a v2
10:56:12bughunter2and rockbox won't work?
10:56:20Bagderhttp://daniel.haxx.se/sansa/v2.html
10:56:46bughunter2ouch
10:56:53pixelmalinuxstb: yes, I'll try to explain this too
10:57:08bughunter2why do manufacturers keep doing that
10:57:35linuxstbTo upgrade to newer, faster, cheaper, more efficient processors...
10:57:42rasherbughunter2: Well it's a new device, essentially
10:57:48bughunter2hmm :)
10:58:14bughunter2i could help making rockbox work on it
10:58:16pixelmalinuxstb: I even think someone (bluebrother?) started a wiki page on how to distinguish players and found a helpful link for the Ipods...
10:58:21bughunter2although i'm not going to tear the player apart when i have it
10:59:12linuxstbpixelma: I've also started to add support to ipodpatcher to recognise the newer ipods. I think it's working, but haven't got round to committing it yet. But after I do that, rbutil should be able to give a message to users if it finds an unsupported ipod.
10:59:32linuxstbI've no idea if we can do the same with the v2 Sansas though...
10:59:48pixelmaah, IdentifyPlayers it is
10:59:58pixelma-s
11:00
11:01:01linuxstbpixelma: Maybe the manual could say something like "if your ipod/sansa was purchased after [date], then it may be a "classic/v2/..." and Rockbox will not work on it. See IdentifyingPlayers for instructions on identifying your exact model."
11:01:53bughunter2oh heck, the original firmware is fine too i think. at least no iTunes-clone and you can just drag n drop files onto it
11:02:14bughunter2i'll consider buying one, and maybe help getting rockbox work onto it
11:02:16bughunter2:)
11:04:07pixelmalinuxstb: nice (about ipodpatcher). The note should still be before the automated installation part, I would think. And I also think that the note shouldn't be too long, so the link to the wiki (as done in other parts) would be better than writing long essays
11:08:24 Join gromit` [0] (i=gromit@ras75-5-82-234-244-69.fbx.proxad.net)
11:08:40bughunter2gotta go to school :)
11:08:44bughunter2have fun
11:09:33 Part bughunter2
11:13:59 Part sanlist
11:14:34preglowlinuxstb: svn blame
11:14:40preglowlinuxstb: feel free to change
11:18:30 Join Bagder_ [0] (n=daniel@1-1-5-26a.hud.sth.bostream.se)
11:22:44 Quit Bagder (Nick collision from services.)
11:22:49 Nick Bagder_ is now known as Bagder (n=daniel@1-1-5-26a.hud.sth.bostream.se)
11:44:48 Join FOAD_ [0] (n=dok@dinah.blub.net)
11:48:42 Join mrkiko [0] (n=mrkiko@host169-98-static.34-88-b.business.telecomitalia.it)
11:49:25mrkikoHi all again! :)
11:49:41mrkikoAre there victims regarding my license question?
11:50:59 Quit Nico_P (Remote closed the connection)
11:52:06 Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net)
11:55:03 Nick fxb is now known as fxb__ (n=felixbru@h1252615.stratoserver.net)
11:56:32linuxstbpreglow: OK, fixed. "svn blame" isn't an accurate indicate of (C) holder...
11:58:17preglowsure, but hunting for copyright holders isn't what i like to spend my rockbox time doing
11:58:56preglowand if the code had been moved around enough for svn blame not to show it, i would have had to check out different revisions and check them
11:59:14 Nick idnar_ is now known as idnar (i=mithrand@unaffiliated/idnar)
12:00
12:00:04 Quit safetydan ("Leaving")
12:02:10 Quit FOAD (Read error: 110 (Connection timed out))
12:02:11 Nick FOAD_ is now known as FOAD (n=dok@dinah.blub.net)
12:06:54 Join rp- [0] (n=rp@193.154.222.107)
12:15:07mrkikoHow can I help us in hunting (C) holders?
12:15:46mrkikoI will help you even if I have to look over all those .C and .S files!
12:15:59GodEater_mrkiko: you're assuming we want to change the license - no such thing has been agreed.
12:16:35rasherI think the consensus right now is "it's not worth the trouble coming to a consensus". At least not yet.
12:16:55mrkikono
12:16:59Bagderthat's my view of the situation as well
12:17:17rasherRight now, the only problem is a few months of espeak improvements.
12:17:22GodEater_so there's no point chasing down (C) holders
12:17:34mrkikoI would help aniway: I know how difficult writing code can be, so I think all should be clear regarding licenses...
12:17:59Bagdermrkiko: did it sound "clear" previously?
12:18:28GodEater_there's never going to be anything clear about legalese. That's why there are lawyers.
12:18:29Bagderit's a muddy area, finding copyright holders will only partially solve that
12:19:24LloreanIf you found 100% of copyright holders (or at least enough that all other parts could be rewritten), you could of course re-license. But any effort less than that is pretty useless
12:20:06GodEater_and given we don't even know if we're going to relicense - it's a complete waste of time.
12:20:47mrkikoehehe...
12:21:24mrkikobut in some files, especially audio drivers I think, have absolutely no autor name in them.
12:21:46GodEater_so you can't track them down anyway...
12:22:17Bagderwell, finding the copyright holder is a good idea for noname-parts imho
12:22:18rasherThat's where svn history comes into play.
12:22:25linuxstbIt would at least be nice if we could come to some agreement about exactly what the current licensing is though... Am I alone in thinking section 9 applies to files with the standard Rockbox header?
12:22:42*linuxstb just wants a show of hands, and promises not to argue any more ;)
12:23:08Bagderlinuxstb: I'm leaning back and forth and I'm not really sure where I stand in this issue yet
12:23:38rasherI don't think I need to say anything here.
12:23:53LloreanI believe that section 9 does. We point to a file that says "if no version number is mentioned in the code", I think the fact that the file we're pointing to says that means that its own version number isn't the relevant factor.
12:24:13LloreanSo I guess my hand goes for "Currently we're v2 or later"
12:24:23linuxstbAnd I guess the second question is: "Ignoring the practical problems, would you like Rockbox to go to v3?"
12:24:35LloreanNo, so I can live with us not being v2 or later. :)
12:25:03linuxstbLlorean: It doesn't mention "later" - it's "any version ever published".
12:25:52LloreanThat's true
12:26:08rasherlinuxstb: I would not like v3, but my opinion on the matter doesn't weigh heavily, considering my tiny amount of contributions.
12:26:50Lloreanlinuxstb: I do find it a little doubtful someone will relicense under v1, but I suppose that under my reading they have that option.
12:27:05 Quit FunkyELF ("Leaving")
12:27:38GodEater_my opinion is the same as rasher's
12:30:34rp-is there a wiki page about this license topic, so you don't have to rediscuss this whole thing all the time? :)
12:30:47 Join MethoS- [0] (n=clemens@pD955D580.dip.t-dialin.net)
12:30:58GodEater_a wiki page can only represent the opinion of it's author...
12:32:38 Join davina [0] (n=davina@cpc1-sout6-0-0-cust616.sotn.cable.ntl.com)
12:36:00 Join Daolan [0] (n=neo-ryan@dslb-084-062-040-176.pools.arcor-ip.net)
12:39:35 Join Thundercloud [0] (n=thunderc@resnet11.nat.lancs.ac.uk)
12:42:36***Saving seen data "./dancer.seen"
12:44:08 Quit MethoS- (Remote closed the connection)
12:44:24 Join MethoS- [0] (n=clemens@pD955D580.dip.t-dialin.net)
12:45:37 Quit parafin|sleep ("So long and thanks for all the fish")
12:45:40 Join parafin [0] (i=parafin@paraf.in)
12:46:46 Part Daolan
12:47:00 Join PaulJam [0] (i=PaulJam_@vpn-3078.gwdg.de)
12:50:17 Join desowin [0] (n=desowin@hdp186.internetdsl.tpnet.pl)
13:00
13:05:07 Quit mrkiko (Remote closed the connection)
13:08:30preglowwould remove() be the safest function to use to delete a function in c?
13:08:54Zagordelete a function?
13:09:05linuxstbYou mean delete a file?
13:09:10preglowyes...
13:09:18preglowi surely did
13:11:21linuxstbWhat are the other options? Reading the man page, I can't see the difference between remove() and unlink() - it appears to just call unlink() or rmdir() depending on whether it's a file or dir...
13:11:43preglowunlink might look like a unix function
13:12:04preglowah, yeah, remove is a c89/99 function
13:12:13preglowunlink is a unix thing
13:12:16preglowremove() it is, then
13:12:33preglowi don't know if windows cares, but i'll use remove just in case
13:12:37linuxstbYes.
13:12:51linuxstbAs in, I agree that remove() is standard C, so is preferred.
13:14:26 Join goffa_ [0] (n=goffa@216.220.23.105)
13:15:07preglowspeex encodes music remarkably well...
13:15:27jmspeexpreglow: it depends a lot on the kind of music
13:15:42jmspeexsome encodes decently, some is catastrophic
13:15:49preglowjmspeex: yeah, some of the drums do sound a bit pitched at lower bitrates :)
13:16:10jmspeex(single instrument with "rich" sound tends to be OK)
13:16:24jmspeexmultiple "pure" instruments lead to catastropy
13:16:34preglowit still performs better than i'd expect
13:17:06jmspeextry some jazz and see it collapse!
13:17:21jmspeexespecially if there's complex piano
13:17:37 Quit goffa (Read error: 110 (Connection timed out))
13:18:04jmspeexOTOH, rock tends to do OK even with many instruments
13:27:51 Join goffa [0] (n=goffa@216.220.23.105)
13:28:31 Join barrywardell [0] (n=barrywar@host-194-46-236-147.dsl-ie.utvinternet.net)
13:32:34 Quit goffa_ (Read error: 110 (Connection timed out))
13:42:12 Join goffa_ [0] (n=goffa@216.220.23.105)
13:42:37 Quit desowin (Read error: 104 (Connection reset by peer))
13:42:49 Join CaptainSquid [0] (n=Miranda@proxy14.netz.sbs.de)
13:49:41 Quit goffa (Read error: 110 (Connection timed out))
13:54:16 Join ZincAlloy [0] (i=d9eeef90@gateway/web/cgi-irc/labb.contactor.se/x-34ebd22b02c56e67)
14:00
14:00:11 Join goffa [0] (n=goffa@216.220.23.105)
14:00:39ZincAlloy /msg nickserv set hide email on
14:01:06peturdone... now set password
14:02:02 Quit goffa_ (Read error: 110 (Connection timed out))
14:02:21ZincAlloyuhm.... I think I'll keep the old one for now ;)
14:05:46 Join desowin [0] (n=desowin@atheme/developer/desowin)
14:06:33 Quit barrywardell ()
14:06:42 Quit goffa (Remote closed the connection)
14:06:48 Join FOAD_ [0] (n=dok@dinah.blub.net)
14:12:04 Join Arathis [0] (n=doerk@p508A640B.dip.t-dialin.net)
14:13:31preglowargh, what do i have to pay for a graphics card driver that actually works :/
14:14:03jhMikeSsince when do you get charged for card drivers?
14:14:50peturuse windows? *ducks*
14:15:19*jhMikeS never has been charged for windows drivers
14:15:45peturI was responding to preglow....
14:16:22jhMikeSpay money for linux drivers?
14:17:11*petur gives up
14:17:18jhMikeSbeh, whatever...I'm just hitting a snag on something so I'll sit here bs-ing
14:17:26 Join goffa [0] (n=goffa@216.220.23.105)
14:17:48peturstill early morning?
14:17:59jhMikeSthat's redundant :)
14:18:33jhMikeSmorning = early = semi-functional
14:18:57peturknow the feeling ;)
14:19:26 Quit FOAD (Read error: 110 (Connection timed out))
14:19:26 Nick FOAD_ is now known as FOAD (n=dok@dinah.blub.net)
14:19:28 Quit Febs (Read error: 110 (Connection timed out))
14:20:40ZincAlloyjhMikeS, you need caffeine!
14:20:50preglowjhMikeS: haha, no, but it feels like i would be willing to pay for it
14:21:24*jhMikeS thinks about why "if" became the preferred keyword in stead of "when" as in "when (a == 1)". :p
14:21:48*jhMikeS has been hitting the caffeine but it takes until noon to kick in
14:27:31 Join kugel [0] (i=kugel@unaffiliated/kugel)
14:31:48preglowjhMikeS: well, perl has "unless (a == 1)" :>
14:32:40*jhMikeS do { consume_caffeine(); } while (it_is_so_damn_early);
14:34:29 Join TTThomas [0] (n=tblackwe@c-68-38-172-205.hsd1.de.comcast.net)
14:36:23 Quit goffa (Remote closed the connection)
14:36:43 Join goffa [0] (n=goffa@216.220.23.105)
14:42:41***Saving seen data "./dancer.seen"
14:47:38*linuxstb is reminded of http://www.cs.bgu.ac.il/~omri/Humor/verbose-c.html
14:59:02 Join Jag86 [0] (i=Raziel@host57-216-dynamic.58-82-r.retail.telecomitalia.it)
15:00
15:01:26 Quit Jag86 (Client Quit)
15:04:52 Quit goffa (Remote closed the connection)
15:05:30 Join goffa [0] (n=goffa@216.220.23.105)
15:06:52DarkStylistyou guys have the mod.codec download page?
15:07:32markunnope
15:08:12DarkStylistdamn
15:08:49markunDarkStylist: you still only have the same options as linuxstb gave you yesterday: 1) compiler rockbox yourself with the MOD patch 2) look in the forum if there is a 'unsupported' build with the MOD patch
15:09:07markuncompiler -> compile
15:09:31preglowjhMikeS: hrmph, looks like we still have clicking problems
15:19:23DarkStylista friend has 1 at home cutomly made for sansa build (official)
15:19:38 Join Daolan [0] (n=neo-ryan@dslb-084-062-040-176.pools.arcor-ip.net)
15:19:56jhMikeSpreglow: from whence?
15:20:07preglowjhMikeS: ml
15:20:14DarkStylistlooks like i have 2 wait 4 him
15:20:17preglowjhMikeS: now in talk file clips
15:20:47jhMikeSI saw a thing about .talk's getting truncated but that seems it's probably pcmbuffer related
15:22:23jhMikeSbesides, how would a .talk clip click if it's just one atomic stream?
15:23:10GodEater_DarkStylist: are you saying you've searched the unsupported builds forum and haven't found one which includes the patch you want ?
15:23:50preglowjhMikeS: i mailed and asked him to confirm it's click-free from the encoder with rbspeexdec now
15:25:08jhMikeSput up a request for anyone to post the particular clips somewhere. I need to be able to actually experience and analize the m
15:25:41*preglow visualizes someone analizing voice clips
15:26:23*jhMikeS reminds preglow it's still AM hours here
15:27:10preglowsure, doesn't make the prospect any less amusing :P
15:27:16 Quit goffa (Read error: 110 (Connection timed out))
15:27:20 Join goffa [0] (n=goffa@216.220.23.105)
15:28:38*jhMikeS gets confused nowadays between Brit and US spelling and ends up typing neither.
15:34:41 Join TMM [0] (n=hp@ip5650d1ab.direct-adsl.nl)
15:41:39 Join goffa_ [0] (n=goffa@216.220.23.105)
15:42:15 Join MethoS-- [0] (n=clemens@pD955F73A.dip.t-dialin.net)
15:47:29 Quit goffa (Read error: 110 (Connection timed out))
15:48:51 Join kubiix [0] (n=Miranda@adsl-pha21-156-158-212.bluetone.cz)
15:48:52 Quit kubiix (Client Quit)
15:50:51 Quit rp- ("leaving")
15:51:12 Join mf0102 [0] (n=michi@85.127.21.194)
15:53:48 Join Siku [0] (i=Siku@e81-197-68-206.elisa-laajakaista.fi)
15:58:21 Part LinusN
15:59:27 Quit MethoS- (Read error: 101 (Network is unreachable))
16:00
16:00:55 Quit Zagor ("Client exiting")
16:06:36 Quit CaptainSquid ("Miranda IM!")
16:07:05 Join scorche|w [0] (n=42c007b2@rockbox/administrator/scorche)
16:08:46DarkStylisti never found one for sansa with the mod patch sorry
16:08:57markun"I recently installed rockbox in less then 5 hours and it was very easy and nothing went wrong." :)
16:09:15scorche|wi do hope he means minutes
16:09:24markunI hope so too
16:09:41markunalthough 5 minutes is also less than 5 hours of course :)
16:09:42desowinand that includes compiling ;-)
16:10:34amiconn?
16:10:48desowinI mean, rockboxing my sansa took my less than 5 minutes including compiling rockbox
16:11:00desowinI wonder about that quote though
16:11:19markunamiconn: was reading some old forum post http://forums.rockbox.org/index.php?topic=13962.msg104608#msg104608
16:12:45 Nick fxb__ is now known as fxb (n=felixbru@h1252615.stratoserver.net)
16:15:38DarkStylistrockboxing my sansa was a few minutes too
16:15:47DarkStylistused the graphical installer :)
16:17:19DarkStylisti must confess mp3's on sansa firmware sucks compared to rockbox the sound can be much better configurated with rockbox
16:19:21amiconndesowin: Compiling for sansa (+make zip) would take ~12 minutes for me..
16:23:22DarkStylistthx for all help guys see ya
16:23:28 Part DarkStylist
16:32:25 Join Arathis2 [0] (n=doerk@p508A3C81.dip.t-dialin.net)
16:36:21 Join Crash91 [0] (n=evil91@41.232.212.187)
16:37:10 Quit MethoS-- (Remote closed the connection)
16:42:44***Saving seen data "./dancer.seen"
16:45:17 Quit J3TC- (Read error: 110 (Connection timed out))
16:50:00 Quit Arathis (Read error: 110 (Connection timed out))
16:50:03 Join HowdyDoody [0] (n=Administ@c-71-225-221-149.hsd1.pa.comcast.net)
16:50:11 Quit InHisName (Read error: 113 (No route to host))
17:00
17:00:44 Quit Jon-Kha (Remote closed the connection)
17:02:17 Join XavierGr [0] (n=xavier@ppp70-154.adsl.forthnet.gr)
17:02:54 Join Jon-Kha [0] (i=jon-kha@80-248-247-190.cust.suomicom.fi)
17:03:13 Quit TMM ("Ex-Chat")
17:05:05 Join kubiix [0] (n=Miranda@adsl-pha21-156-158-212.bluetone.cz)
17:10:49 Join jgarvey [0] (n=jgarvey@cpe-024-163-032-204.nc.res.rr.com)
17:11:11 Nick j0tt is now known as jott (n=j@unaffiliated/jott)
17:18:02 Quit ZincAlloy ("CGI:IRC (Ping timeout)")
17:21:03 Quit goffa_ (Remote closed the connection)
17:21:26 Join goffa [0] (n=goffa@216.220.23.105)
17:26:22 Join OlivierBorowski [0] (n=OlivierB@ANancy-157-1-11-6.w86-204.abo.wanadoo.fr)
17:30:04 Quit Crash91 ("Bye Bye!")
17:31:51 Quit GodEater_ (Read error: 110 (Connection timed out))
17:32:29 Join J3TC- [0] (n=jetc123@dhcp75-126.njit.edu)
17:41:05 Join bughunter2 [0] (n=Administ@ip565fbeaa.direct-adsl.nl)
17:42:32 Join GodEater_ [0] (n=bryan@rockbox/staff/GodEater)
17:47:12GodEater_anyone know who deleted the ipod classic thread from new ports ?
17:47:15pixelmaSoap: don't you mean the other Dominik...? ;)
17:47:47Soappixelma, should I? Bluebrother asked me to ask Domonoky...
17:48:07 Nick Arathis2 is now known as Arathis (n=doerk@p508A3C81.dip.t-dialin.net)
17:48:35 Quit goffa (Remote closed the connection)
17:48:40pixelmasorry, didn't know that
17:49:32Soapis the other dominik in the area as well?
17:49:40*Soap looks at the google map
17:50:04Soapyou're all the way up in Berlin, right?
17:50:19pixelmayes (almost)
17:50:25scorche|wSoap: dominik is bluebrother
17:50:46Soapduh - got confused for a minute.
17:50:52 Join perrikwp [0] (i=98216b42@gateway/web/cgi-irc/labb.contactor.se/x-233f32565ad4fe39)
17:54:07 Quit kugel (Read error: 104 (Connection reset by peer))
17:54:15 Quit J3TC- (Read error: 104 (Connection reset by peer))
17:54:18 Join goffa [0] (n=goffa@216.220.23.105)
17:54:55 Join J3TC- [0] (n=jetc123@dhcp75-126.njit.edu)
17:56:18 Quit perrikwp ("CGI:IRC (Ping timeout)")
18:00
18:00:56 Join Lear [0] (i=chatzill@rockbox/developer/lear)
18:05:25 Part pondlife ("Gone")
18:07:00 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
18:11:53 Join lazka [0] (n=lazka@84.119.31.204)
18:14:06 Quit XavierGr (Nick collision from services.)
18:14:09 Join XavierGr [0] (n=xavier@ppp260-88.adsl.forthnet.gr)
18:16:24 Join TMM [0] (n=hp@c5147518c.cable.wanadoo.nl)
18:19:24 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
18:19:57 Quit XavierGr (Nick collision from services.)
18:19:58 Join XavierGr_ [0] (n=xavier@ppp163-117.adsl.forthnet.gr)
18:20:10 Nick XavierGr_ is now known as XavierGr (n=xavier@ppp163-117.adsl.forthnet.gr)
18:21:12 Quit linuxstb (Read error: 110 (Connection timed out))
18:26:19 Join moos [0] (i=moos@m236.net81-66-159.noos.fr)
18:26:19 Quit J3TC- (Read error: 104 (Connection reset by peer))
18:26:45 Quit petur ("work->home")
18:28:02 Join kugel [0] (i=kugel@unaffiliated/kugel)
18:28:07preglowjhMikeS: could you add some enlightening comment in #8270 and close it if you're convinced you know it's not a bug?
18:29:38 Join rlpowell [0] (n=rlpowell@chain.digitalkingdom.org)
18:29:40*jhMikeS tends to think he's sheds dark on the subject
18:29:54 Join toffe82 [0] (n=chatzill@h-74-0-180-178.snvacaid.covad.net)
18:31:08 Join BigBambi [0] (n=Alex@rockbox/staff/BigBambi)
18:31:13 Quit lee-qid (Read error: 110 (Connection timed out))
18:31:38rlpowellCan anyone suggest a good choice to run rockbox on with a minimum of 8GiB? My ipod 4G just died. :(
18:32:54linuxstb_You can buy an 8GB Sansa E280, which also accepts Micro(HC)SD cards for memory expansion.
18:33:12linuxstb_This page has some useful info - http://www.rockbox.org/twiki/bin/view/Main/BuyersGuide
18:33:19rlpowellI read a review that said that USB syncing wasn't working on the E280.
18:33:38linuxstb_It's the same as the ipod - you need to reboot to the original firmware.
18:33:39rlpowellSo you had to load the original firmware to put any music on it, but the original firmware doesn't accept m3u playlists.
18:33:43rlpowellUmmm.
18:34:02rlpowellI didn't have to do any such thing on my ipod.
18:34:11linuxstb_Yes you did, you just didn't know it...
18:34:18rlpowellUmm. OK.
18:34:27linuxstb_Rockbox automatically rebooted to the original firmware's "emergency disk mode" for you.
18:34:40rlpowellAh.
18:35:17linuxstb_But the Sansa appears as a normal external drive (i.e. UMS), so you can copy any files to it, including m3u playlists.
18:35:40linuxstb_It doesn't matter that the original firmware can't read them (which the ipod's OF couldn't either).
18:36:03rlpowellhttp://www.amazon.com/review/R1V6G1ZRQJS92/ref=cm_cr_rdp_perm −− that's the review I'm talking about.
18:36:21 Join lee-qid [0] (n=liqid@p549674EE.dip.t-dialin.net)
18:36:28rlpowellWhere he says he has to do all his syncing with Windows Media Player, and he has to run a plugin to convert the playlists back to m3u to use them in rockbox.
18:37:04rlpowellPresumably MSC mode solves that problem, but: When the Sansa is in MSC mode, it takes literally 5-10 minutes to boot while it catalogs its content
18:37:15rlpowell(and he's not the only one who makes that long boot time complaint)
18:37:29LearI never see it...
18:37:52jhMikeSpreglow: I added some enlightenment. I can't say whether there's a bug for Nano however.
18:37:55LearEither because the bootloader stops it, or it never boots longer than to the language selection screen.
18:38:09Soapsubstitute seconds for minutes, rlpowell
18:38:10rlpowellLear: You have an e280 in MSC mode, and it boots fairly quickly? How much space are you using up?
18:38:28 Join bertrik [0] (n=Bertrik_@249-022-045-062.dynamic.caiway.nl)
18:38:49LearWell, there's some 150 MB free or so. Most of it in Vorbis though.
18:39:09rlpowellAh. I suppose it can't inventory those. :)
18:39:52linuxstb_rlpowell: The Rockbox bootloader disables the OF's database refresh
18:39:57LearThere are some MP3:s too though.
18:40:13rlpowellThanks for alll your help, all. I think I'll go ahead and buy one of those then; it was my first choice until I read that review.
18:40:32linuxstb_rlpowell: Also, a native USB mode for Rockbox is under active development, so hopefully you won't even need to reboot soon.
18:40:47rlpowelllinuxstb_: Huh. Sounds like that reviewer isn't using rockbox very well, since he apparently was avoiding MSC mode, so I'm not surprised.
18:41:00rlpowellYeah, saw that. How's that going? (the dev)
18:41:59linuxstb_I think it's close, but it's one of those things where the last 5% of the work takes 95% of the time... So it's hard to predict when it will be working.
18:42:30preglowjhMikeS: so one really shouldn't be able to see any difference in black level with the ycbcr clipping?
18:42:47***Saving seen data "./dancer.seen"
18:43:52rlpowellGreat; thanks again for all your help.
18:43:56jhMikeS16 = black and should convert to 0 in RGB space. anything below should clip
18:44:07preglowyes, exactly
18:45:17 Part bughunter2
18:46:48 Join bughunter2 [0] (n=Administ@ip565fbeaa.direct-adsl.nl)
18:46:55 Part bughunter2
18:48:42rlpowellOh, one more thing. Is the port at the bottom of the E280 the same as the one at the bottom of my 4G ipod?
18:48:57rlpowell(I doubt it, but I've got a bluetooth adapter I'd love to keep using....)
18:50:00jhMikeSI wouldn't do that
18:50:23rlpowellWouldn't do which?
18:50:37Soapif your Bluetooth adapter worked with Rockbox on your 4G iPod, I would be forever grateful if you would fill out the IpodAccessories wiki page.
18:51:07rlpowellIt *sort of* worked; I couldn't change channels, but the sound was fine.
18:51:18rlpowellHell, if it won't work with the sansa, I can just sell you the thing.
18:51:50rlpowell"the IpodAccessories wiki page" == URL?
18:51:55 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
18:52:01 Part pondlife ("Gone")
18:52:02rlpowellOh, nevermind; found it.
18:53:00 Join linuxstb__ [0] (n=linuxstb@i-83-67-212-170.freedom2surf.net)
18:53:35*rlpowell goes to lookup what the device actually *was*. :)
18:54:14rlpowellAs an aside, I *HEART* my HT280 bluetooth headphones.
18:54:41rlpowellTEN Technology naviPlay Bluetooth Stereo Adapter for iPod −− that was it; adding to the wiki now.
18:55:40 Quit kugel ("ChatZilla 0.9.79 [Firefox 2.0.0.8/2007100816]")
18:55:40rlpowellBug in your TWiki setup: hitting cancel at the login takes you to a page that says "{ScriptUrlPath}/view/TWiki/TWikiRegistration"
18:56:04 Nick linuxstb__ is now known as linuxstb (n=linuxstb@i-83-67-212-170.freedom2surf.net)
18:56:10 Quit linuxstb_ (Nick collision from services.)
18:57:49 Join Rincewind [0] (i=H0XIZkug@nat-wh-1.rz.uni-karlsruhe.de)
18:58:25 Join J3TC- [0] (n=jetc123@dhcp75-126.njit.edu)
18:58:28rlpowellSomeone give RobinPowell write access, please.
18:58:40rlpowell(otherwise I can't do what Soap told me to. :)
19:00
19:00:36linuxstbrlpowell: Done
19:00:39pixelma7 seconds too slow :)
19:00:39rlpowellThanks.
19:02:33preglowisn't .aac an extension you see for aac?
19:05:25 Join roxfan [0] (n=dunno@171.222-64-87.adsl-dyn.isp.belgacom.be)
19:05:46rlpowellSoap: http://www.rockbox.org/twiki/bin/view/Main/IpodAccessories −− last line. If such a thing is useful to you, I'd love to sell it. :D
19:06:57linuxstbpreglow: I think that normally means raw AAC (i.e. not in mp4)
19:07:15linuxstb(I could be wrong though...)
19:07:20 Join mud-rb [0] (n=mud@dialup-4.156.222.100.Dial1.Boston1.Level3.net)
19:07:28preglowlinuxstb: right
19:07:33 Join japc [0] (n=japc@194.65.5.235)
19:07:47 Join Domonoky [0] (n=Domonoky@f051066030.adsl.alicedsl.de)
19:09:34rlpowellThere aren't any rockbox-supported players with native bluetooth, are there?
19:09:57Domonokyno
19:10:33rlpowellDidn't think so; thanks.
19:11:12rlpowellRockbox is #1 on my requirements list, but I man can dream... :)
19:13:35 Join Suleeto [0] (n=jagrei@71.177.21.78)
19:13:43Suleetowow lots of people in here
19:14:15 Join linuxftw [0] (n=raul@71-223-154-181.phnx.qwest.net)
19:14:15Suleetoanyonw awake?
19:14:36pixelmanope, all bots :)
19:14:41linuxftwumm, does anyone know how to get linux on my ipod with rockbox working?
19:15:04linuxftwi anailated my apple msofts -_-
19:15:10preglowwow, i really had to read that nick twice
19:15:14*linuxstb too ;)
19:15:17 Join hcs [0] (n=agashlin@rockbox/contributor/hcs)
19:15:30scorche|wlinuxftw: you might want to try #ipodlinux
19:15:43linuxftwthey alllll afk lol.....
19:15:53 Quit roxfan[zzz] (Connection timed out)
19:16:01preglowlinuxftw: then what makes you think we're interested in helping you to put linux on your ipod?
19:16:09linuxftw-_- cuz ur nice?
19:16:15preglowwe're nice when it comes to rockbox
19:16:20preglowwe don't care too much about linux on the ipod
19:16:21linuxftwwell fine -_-
19:16:22 Join EspeonEefi [0] (i=espeonee@30-7-114.wireless.csail.mit.edu)
19:16:40mud-rbi think most rockbox users just use rockbox, i doubt many people know exactly
19:17:11 Nick linuxftw is now known as rockboxftw (n=raul@71-223-154-181.phnx.qwest.net)
19:17:16rockboxftwD:?
19:17:22linuxstbWelcome.
19:17:34scorche|wthat doesnt make ipodlinux on-topic here
19:17:37rockboxftwok can i install rockbox with a dead ipod?
19:17:47scorche|wdefine dead
19:17:49mud-rbdead in what way?
19:17:53rockboxftwi reformatted it all D:
19:17:57linuxstbNo, the install tools require a working ipod - i.e. one that can boot into the original firmware.
19:18:09rockboxftwoh noes..
19:18:11rockboxftwcrap..
19:18:15linuxstbWhat desktop OS do you use?
19:18:23rockboxftwlinux/windowsxp
19:18:34linuxstbThen see here - http://www.rockbox.org/twiki/bin/view/Main/IpodManualRestore
19:18:35scorche|wyou can try this though: http://www.rockbox.org/twiki/bin/view/Main/IpodManualRestore
19:18:44mud-rbhaha
19:18:49rockboxftwcreepy..
19:18:57rockboxftwboth u typed it at nearly same time
19:19:15scorche|wnot really...i typed it a while back, but CGI:IRC lags a decent bit
19:19:18rockboxftwdown with ipodlinux ?
19:20:32mud-rbi imagine there are also tools from apple to recover ipods, i really wouldn't know though
19:20:55rockboxftwbrb need get on my windows pc
19:21:21scorche|wmud-rb: i think they stopped distributing them outside of itunes
19:21:39mud-rbscorche|w: oh really? that stinks
19:22:01 Join Redbreva [0] (n=Miranda@host86-144-108-57.range86-144.btcentralplus.com)
19:22:12rlpowellSomebody else besides jabra *must* make 3.5mm jack bluetooth adapters...
19:25:16 Join Bagder_ [0] (n=daniel@1-1-5-26a.hud.sth.bostream.se)
19:29:29 Join Frazz [0] (n=Fraser@thelawsons.plus.com)
19:32:33preglowlinuxstb: do you know if we've tried using low-power sbr mode in faad?
19:34:49 Quit bertrik ("bye")
19:36:39preglowaccording to some sources, low power sbr mode should be about equal in complexity to lc aac at the equivalent sample rate
19:37:47hcsaccording to some sources we never landed on the moon
19:38:20rockboxftwbull
19:38:29preglowi surely never have
19:38:42hcsAre you sure?
19:39:04hcsBecause I may have compromising photographs of you on the moon.
19:39:15hcsAnd your denial will only make things worse.
19:39:20 Part rockboxftw
19:40:17preglowwell, i thought i never had...
19:40:21preglowbut now that i think about it
19:40:26preglowyeah, i distinctly remember landing on the moon
19:41:32*Domonoky also wants those drugs, preglow must have taken.. :-)
19:41:47*preglow strokes his 500 sheet stack of lsd
19:42:42scorche|w500?...jees..
19:42:44 Quit Bagder (Read error: 110 (Connection timed out))
19:42:59 Join austriancoder [0] (n=austrian@rockbox/developer/austriancoder)
19:43:41 Quit J3TC- (Read error: 104 (Connection reset by peer))
19:46:22 Join J3TC- [0] (n=jetc123@dhcp77-17.njit.edu)
19:46:44 Join pondlife [0] (n=Steve@rockbox/developer/pondlife)
19:46:52 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
19:49:20preglowscorche|w: btw, is it easy to reset a gigabeat s? :>
19:49:42 Quit japc (Read error: 104 (Connection reset by peer))
19:51:09scorche|was in power on/off reset, or resetting of the firmware?
19:51:56preglowas in resetting it when i've made it hang
19:52:04Suleetohey do you guys know why the rockbox code won't work on sigmatel chips yet? any word on development there
19:52:15preglowSuleeto: no development whatsoever
19:52:31preglowscorche|w: i just had to go hunting for a paperclip for my h120 and remembered hard reseting something is essential functionality for me :)
19:52:34scorche|wpreglow: there is a small "battery switch" at the bottom
19:52:37Suleetois it that the sigmatel chip is too new?
19:52:52scorche|wSuleeto: just that no one has done it..
19:53:25Suleetoif I understand correctly, all the newer Sansas use that... I know that Clip and Express do
19:53:44preglowSuleeto: all newer sansas use an austramicrosystems chip
19:53:46preglowno sigmatel
19:53:51preglowditto clip
19:54:16scorche|wSuleeto: they use http://www.austriamicrosystems.com/03products/products_detail/AS3525/description_AS3525.htm
19:54:24*Suleeto mutters to himself... *why did I think it was sigmatel...*
19:54:42preglowand development on that front is on its way
19:55:27 Join petur [0] (n=petur@rockbox/developer/petur)
19:55:35mud-rbso i have a question: what exactly is "iram"?
19:55:36 Join Xerion [0] (n=xerion@cp198589-d.landg1.lb.home.nl)
19:56:17preglowmud-rb: fast sram that is internal to a chip
19:56:59mud-rbpreglow: ah, thanks. i assumed it was something like that. do all of the players have it?
19:57:00 Quit Xerion (Client Quit)
19:57:12preglowmud-rb: most do, but not all in such quantities that we can use it the same way
19:57:37preglowmud-rb: where we have enough of it, we allocate it for all performance sensitive areas, where we don't have enough of it, we just use it for low-level stuff for such a thing is needed
19:57:51preglowlike in bootloaders
19:57:58preglowcurrently, i think only gigabeat f doesn't have much iram
19:58:05mud-rbcan plugins generally steal it if it's not being used for playback?
19:58:07preglowand the original archoses
19:58:12preglowmud-rb: sure
19:58:31 Join Zagor [0] (n=bjst@46.35.227.87.static.tab.siw.siwnet.net)
19:58:44mud-rbhaha i don't know how you guys don't go nuts getting rockbox to work on all of these players..."normal" architecture differences are bad enough
19:59:51 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
20:00
20:00:08scorche|walready nuts
20:00:17preglowwhat he said
20:00:26Learpreglow: Do you know how/if the low-power mode affects quality? (And no, I haven't tried it. Didn't know much about it.)
20:00:32preglowLear: it does, yes
20:00:34Suleetopreglow, the information on anythingbutipod says it's sigmatel: http://www.anythingbutipod.com/forum/showthread.php?t=17236
20:00:46preglowLear: but also significantly lowers computational power requirements
20:00:46 Part austriancoder ("Konversation terminated!")
20:01:00preglowLear: it's an official part of the aac sbr spec
20:01:21scorche|wSuleeto: that is the express though..
20:01:43Lloreanpreglow: I think I've been told the Gigabeat has as much iram as the Archoses (or some Archos).
20:01:54LloreanWhich is, like, 4k?
20:01:55preglowLlorean: and that is true
20:01:59preglowLlorean: something in that ballpark
20:02:14 Quit goffa (Read error: 104 (Connection reset by peer))
20:02:16preglowLear: i'll test it as soon as i get my synced faad up and going, which i'm working on now
20:02:19Suleetoi had thought express was newer.. has it been out for awhile?
20:02:23 Join webguest30 [0] (i=185da6ea@gateway/web/cgi-irc/labb.contactor.se/x-46b317f7082f9cd2)
20:02:52scorche|wi interpret "newer sansas" to mean the recent version 2
20:02:58Suleetothose guys seem to think that it can be ported, the biggest benefit being SDHC chips
20:03:08Suleetoahhhh and express is not v2 then i take it
20:03:39LloreanYeah, the "newer sansas" tend to mean "the v2 editions of the hardware we support, and their close brothers the m200 and clip"
20:04:19 Join spiorf [0] (n=spiorf@host202-206-dynamic.16-79-r.retail.telecomitalia.it)
20:04:49Suleetoand nobody has an interest in porting to an express then
20:05:02Suleetoi lack the knowhow or i would do it myself
20:05:25BigBambiSuleeto: Your best bet is to encourage enthusiasm on an owners forum
20:05:27scorche|wi havent seen anyone come forward with the willingness and the capability step forward to do the port, no
20:07:05Suleetobigbambi: there is an active thread on this with the earliest post dated in July and the most recent as of this last Sunday.
20:07:29SuleetoI am unclear myself as to what is involved in doing a port
20:07:47LloreanSuleeto: http://www.rockbox.org/twiki/bin/view/Main/NewPorts
20:07:55Lloreanhttp://www.rockbox.org/twiki/bin/view/Main/NewPort even
20:07:58scorche|wstrike an s
20:08:40BigBambiSuleeto: Sure, but what is there is the extent of the effort here. If that needs more people, you need to atract them from elsewhere
20:09:10Suleetoi dont understand
20:09:23 Join Robin0800 [0] (n=Robin080@cpc2-brig8-0-0-cust498.brig.cable.ntl.com)
20:09:32mud-rbyou're going to need help, and if rockbox devs don't care about your platform, you're going to have to try to get other people to
20:09:45ZagorSuleeto: it's quite simple: nobody here has an express
20:10:11BigBambiSuleeto: Ports are made by owners, no-one here has one, so you ned to get more owners with the requisite skills interested
20:10:26 Join Robin_Watts [0] (n=Javacat@91.84.89.194)
20:10:47Robin_WattsEvening all
20:10:57Robin_WattsWas wondering if preglow was about?
20:11:22Suleetoahhhhhh
20:11:33Suleetoi am reading the link, at the very least I can obtain more info
20:11:47BigBambiSuleeto: That is a good start
20:12:18scorche|wRobin_Watts: well, he was a bit ago..
20:12:39preglowdamn, this codec rapes the stack
20:12:52Robin_WattsAha.
20:12:58BigBambiVoila!
20:13:11scorche|wa wild preglow has appeared!
20:13:22Robin_Wattspreglow: I have an optimised version of your qmf_synth ARM code for the speex codec.
20:13:55preglowRobin_Watts: ahh, it's you
20:14:07preglowRobin_Watts: i've tried it, yes, and it actually decreased performance on my target :)
20:14:09Robin_WattsI sent a copy to the speex dev list, but other than getting a message back saying that it was held for moderation, I haven't heard anything back (they are probably busy)
20:14:18*Robin_Watts boggles.
20:14:26Robin_WattsWhat target?
20:14:28preglowRobin_Watts: at least for one mode, heh
20:14:32preglowRobin_Watts: arm7tdmi, ipod
20:14:40preglowRobin_Watts: you seem to be on something arm9 or greater
20:14:46preglowsince you care about load dependencies
20:15:12 Part rlpowell
20:15:31Robin_WattsThe code is written so that it will benefit on things like SA and Xscale cos of the load dependencies, but I can't think of anything I did offhand that should have *slowed* it on any ARM...
20:16:26preglowRobin_Watts: well, the order 8 case of iir_mem16, for example, you stack a register to be able to free one register for the interleaved loads
20:16:38preglowRobin_Watts: the interleaved loads won't hurt arm7, but the extra stack operation will
20:17:02Robin_Wattsbut I should have gained in that routine by using MLA instead of a MUL then a SUB...
20:17:12Robin_Wattsso overall I'd still expect to win.
20:17:14preglowRobin_Watts: mla uses exactly as many cycles as one mul + one sub
20:17:30preglowmla uses mul cycles + 1, sub uses one cycle
20:18:43preglowRobin_Watts: but that should go in anyway... i feel kind of stupid for not seeing that
20:20:15*Robin_Watts boggles again...
20:20:31preglowRobin_Watts: anyway, only the order 8 case of iir_mem16 was slower, and that by a small amount, but enough to make wb decoding slower than it already was, and i'm just pondering if i should care about it
20:22:23Robin_WattsI'm surprised at your claim for MLA timings.
20:22:40preglowwhy? i'm 99% certain it is correct
20:22:55Robin_WattsI've been using ARMs for nigh on 20 years now (since the ARM2 first came out), and I've never twigged that.
20:24:04preglowwell, that's what the arm reference docs say, and it does seem to be confirmed by my performance figures
20:24:10Robin_WattsI built an ARM in an FPGA as a uni project, and MLA is just to be a MUL where you preset the register that you repeatedly shift and add to a value, rather than 0.
20:25:24preglowyou're sure that what arm themselves do?
20:25:33Robin_WattsNo :)
20:25:47Robin_WattsWhich ARM ref docs please?
20:26:20Robin_Watts(I don't distrust you. I'll just still reeling from having a long time preconception shattered)
20:26:33preglowi'm looking for a ref as we speak
20:26:52amiconnhttp://infocenter.arm.com/help/index.jsp?topic=/com.arm.doc.ddi0210c/Chdbedbg.html
20:27:42amiconnmla takes one cycle more than mul, and (s|u)mlal one cycle more than (s|u)mull on arm7tdmi
20:28:19Robin_WattsIndeed.
20:28:22 Join JdGordon [0] (i=836b0065@gateway/web/cgi-irc/labb.contactor.se/x-79f54d62945f5930)
20:28:38Robin_WattsOn SA they are the same (according to the SA instruction timing note)
20:28:40JdGordonhey all
20:29:05Robin_WattsI guess I'd just assumed that previous ARMs were the same.
20:29:12Robin_WattsWell, you learn something new every day!
20:29:15preglow:)
20:29:16amiconnEven within the same architecture version, timing can be different
20:29:33Robin_Wattsamiconn: Yes, I appreciate that.
20:29:35amiconnarm920t and arm7tdmi are both armv4t, but have rather different timing
20:29:39pixelmamoos: that doesn't look right too...
20:29:43 Quit JdGordon (Client Quit)
20:29:43preglowRobin_Watts: but anyway, using mla means less code than mul + sub, so that's going in anyway
20:30:08 Join JdGordon [0] (i=836b0048@gateway/web/cgi-irc/ircatwork.com/x-a7a2393c0baba733)
20:30:19 Quit EspeonEefi ("さよなら")
20:30:23Robin_WattsIt's a shame that that code clips to +/- 7fff.
20:30:36moospixelma:sorry, just commited the right thing
20:30:47Robin_WattsI can clip to the more usual +7fff, -8000 faster.
20:31:36preglowRobin_Watts: yeah, i actually do clip to +7fff..-8000 in my coldfire version of this same code, though, in qmf_synth
20:31:39preglowRobin_Watts: and it works fine
20:32:08Robin_Wattspreglow: The qmf_synth code should have been faster, regardless of processor though ?
20:32:26preglowRobin_Watts: it was
20:33:54preglowRobin_Watts: so feel free to optimize the clipping in qmf_synth to 7fff..-8000 if you want
20:35:16Robin_WattsMVN r9,#&8000 MOV r5,r10,LSR #15 TEQ r5,r5,ASR#31 EORNE r10,r9,r5,ASR #31
20:35:45preglowlooks like the trick we use in dsp.c
20:35:49Robin_Wattsoops, ASR, not LSR. Clips r10 to 15 bits + sign.
20:37:21*Robin_Watts fails to find dsp.c
20:37:50preglow if ((int16_t)sample != sample)
20:37:50preglow sample = 0x7fff ^ (sample >> 31);
20:38:03preglowit's in apps/
20:38:17Robin_Wattsah, right.
20:39:21preglowwell, that shaves off 5 cycles and is more compact
20:40:06*preglow prefers saturating integer arithmetic :/
20:40:21 Quit J3TC- (Read error: 110 (Connection timed out))
20:41:54Robin_Wattsit applies in order_10 and order8 too ?
20:42:02amiconnpreglow: Pah, that would be boring ;)
20:42:43preglowRobin_Watts: those are called too far in the front of the signal chain, if there's a single sample negate happening after that on a -8000 value, things will blow up
20:42:48***Saving seen data "./dancer.seen"
20:42:56preglowRobin_Watts: qmf_synth is at the very end of the signal chain
20:42:58Robin_Wattsfair enough.
20:43:35preglowmov r9, #0x7f00 \n orr r9, r90, #0xff
20:43:38preglowwhat the hell was i thinking
20:43:58Robin_WattsYou did have a @ This can be optimised further in there to cover yourself :)
20:44:12preglowi haven't programmed arm for more than two years, sporadically, heh
20:44:23preglowso i sometimes just ignore glaringly obvious stuff
20:44:30amiconnpreglow: mvn r9, #0x8000 is not identical to that afaics
20:44:42preglowit's not, nmo
20:44:43Robin_WattsUsing the MVN 0x8000 one would have mucked up the -ve test.
20:44:48amiconnThis will set r9 to 0xffff7fff
20:45:06 Quit jhulst ("Konversation terminated!")
20:45:30preglowyes, yes it will
20:45:42 Quit hcs ("Leaving.")
20:45:43preglowtoo much coldfire for me
20:46:02*Robin_Watts hasn't ever used coldfire.
20:47:02preglowit's nice, has fast multiply accumulate, saturating arithmetic, parallel loads, etc
20:47:11preglowbut otherwise looks very much like 68k
20:49:43 Quit ompaul (Read error: 113 (No route to host))
20:51:28preglowRobin_Watts: would there be a way to combine that clipping code with interleavign outputs into top/lower part of registers? i want to replace those four strh's with a stmia...
20:52:45Robin_WattsCan't see how.
20:53:24Robin_WattsAnd STMIAs may be slower on some later ARMs, I think.
20:53:41preglownot on arm7
20:53:48preglowyou pretty much always win on arm7 if you use stm
20:53:50preglowand ldm
20:54:15Robin_Wattsyes, but later ARMs you start to pay a penalty I think.
20:54:36Robin_WattsWhich always annoys me, as the multiple load/stores are lovely.
20:54:36preglow4 strh = 12 cycles, 2 element stmia = 4 cycles, afaik
20:54:48preglowwhich is quite a boost...
20:55:14preglowRobin_Watts: why would stm be slower, really? that just sounds silly
20:55:48Robin_WattsOn the SA they just internally decode LDMs and STMs to STRs.
20:55:53preglowgenious...
20:56:00Robin_WattsOh, it's worse than that.
20:56:00 Quit Nico_P (Read error: 104 (Connection reset by peer))
20:56:29 Join EspeonEefi [0] (i=espeonee@STRATTON-FIVE-FIFTEEN.MIT.EDU)
20:56:35Robin_WattsIf you do STM<conditioncode>IA... they decode to n x STR<conditioncode>...
20:56:52preglowfrom what arm revision on?
20:56:54Robin_WattsSo it doesn't just take a single cycle if the condition code fails.
20:56:55preglowthat's bloody braindead
20:57:03Robin_WattsThat was in the SA. Blame Intel.
20:57:16preglowoh, so arm's cores don't do that?
20:57:16Robin_WattsI believe it's also like that in the Xscale, but I could be wrong.
20:57:26amiconnThat should still not be slower than using n x str directly
20:57:29Robin_WattsARMs cores are probably better behaved.
20:57:37amiconnAnd it saves a few instructions
20:57:50Robin_WattsI *think* there are cases where it is slower - but I can't recall them offhand.
20:58:41amiconnpreglow: On arm9, using ldm/stm is also better, even if it does not directly save cycles
20:58:41Robin_Wattsmaybe it's slower for single register STMs or double ones or something.
20:58:54preglowamiconn: why?
20:59:02amiconnUsing them makes it easier to avoid pipeline stall (at least for ldm)
20:59:06preglowRobin_Watts: well, doing a single register stm isn't exactly clever
20:59:41amiconnAnd you save (n-1) instructions
20:59:46Robin_Wattspreglow: it's tempting to stack/unstack things using STMFD/LDMFD just cos it 'looks right'.
21:00
21:02:58*amiconn wonders what moos did in r15881
21:03:15amiconnThe net difference between r15880 and r15882 is just 2 lines...
21:03:20Robin_WattsIf your figures are right... then doing MOV r0,r0,LSL #16 MOV r1,r1,LSL #16 ORR r0,r0,r1,LSR #16 to mix 2 regs would take 3 cycles. Do that again, and it's another 3. Then 4 for the store is still better than 12.
21:03:44preglowRobin_Watts: indeed
21:04:36preglowRobin_Watts: but that would be slower on arm9 and up
21:04:44Robin_WattsAre your figures for cached or uncached stores?
21:04:56moosamiconn: yup sorry, commited the wrong file, I wanted just correct a little mistake made by the author of the patch.
21:04:57preglowRobin_Watts: cached, 3, uncached, more :)
21:05:13preglowperhaps we really should rename the current one to filters_arm7.S and make a filters_arm9.S
21:05:15amiconnThat's why I enabled the arm asm optimisations for the ape filters for arm7 only...
21:05:17Robin_WattsThe write buffer should merge STRH's to successive addresses, I thought.
21:05:32preglowbut then again, this stuff doesn't just apply to arm9, bah
21:05:38preglowsometimes arm do annoy
21:06:35 Join ompaul [0] (n=ompaul@gnewsense/friend/ompaul)
21:08:30 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
21:10:19*preglow has working aac again
21:12:20preglowLear: i get compiler errors when i compile in low power sbr...
21:13:13 Quit Robin0800 (Read error: 110 (Connection timed out))
21:13:30preglowthanks to some asm, it seems
21:13:36 Join pradin [0] (n=pradin@vpnwl-228-42.net.rpi.edu)
21:15:44pradinI'm currently working on a modification for the text_editor? I have it tested under the UISimulator but I'd like to see it running on my Sansa e200. What are the steps necessary to move a custom firmware build to the Sansa? Perhaps there is a wiki page already describe this procedure?
21:16:20Robin_Watts(Interesting. Looks like the ARM7 didn't have different timings for MUL and MLA. Only the 7M.)
21:17:07preglowLear: haha, it even uses a macro/function not included in the source :)
21:17:28preglowSBR_SQRT_Q2()
21:18:10 Quit Rincewind ("bye")
21:19:09 Join MethoS- [0] (n=clemens@pD955F73A.dip.t-dialin.net)
21:19:29 Join linuxstb_ [0] (n=linuxstb@rockbox/developer/linuxstb)
21:24:55 Join bertrik [0] (n=Bertrik_@249-022-045-062.dynamic.caiway.nl)
21:29:20bertrikdoes the sansa e200 actually have an RTC backup battery?
21:30:25 Quit stewball`ghost (Read error: 110 (Connection timed out))
21:30:41Learpreglow: Mentioned on hydrogenaudio 2.5 years ago. :)
21:30:42Lloreanbertrik: The easiest way to find out would be to set the RTC, unplug the main battery, and see what happens. ;)
21:30:52preglowLear: yeah... i just mailed the faad guy
21:30:52 Quit linuxstb (Read error: 113 (No route to host))
21:30:59 Join entropic [0] (n=grey@mal2.stopavoiding.us)
21:31:04 Join J3TC- [0] (n=jetc123@dhcp77-17.njit.edu)
21:31:10bertrikLlorean: yeah, but maybe there's a supercap somewhere too
21:31:19preglowLear: anyway, there's still a gcc bug going on
21:31:22preglowso i can't test
21:31:29preglowperhaps it's gone with newer gccs
21:31:45bertrikI wonder because during charge, the BVDD value seems too high, while the RTCSUP value seems about right
21:31:58preglowLear: 40% realtime sbr in full mode...
21:32:00preglowfor h120
21:32:03preglowwoopee
21:32:50LearYep, disabling SBR/PS on non-gigabeats does make some sense...
21:34:27preglow137% for a 128kbps, not exactly impressive
21:34:29amiconnoh silliness....
21:34:49amiconnLooks like the cf ape predictor can be optimised further
21:35:52 Quit TTThomas ("Lost terminal")
21:38:18 Part Daolan
21:38:38 Join Daolan [0] (n=neo-ryan@dslb-084-062-040-176.pools.arcor-ip.net)
21:38:41 Quit linuxstb_ ("Leaving")
21:39:29 Nick Bagder_ is now known as Bagder (n=daniel@1-1-5-26a.hud.sth.bostream.se)
21:41:50Learpreglow: That's on coldfire? Seems a little bit slower than what's in svn.
21:42:01preglowLear: that's on coldfire
21:43:17 Quit TMM ("Ex-Chat")
21:47:59 Join linuxstb [0] (n=chatzill@rockbox/developer/linuxstb)
21:48:45 Quit amiconn (Nick collision from services.)
21:48:53 Join amiconn [0] (n=jens@rockbox/developer/amiconn)
21:48:58 Quit Frazz (Read error: 110 (Connection timed out))
21:49:04preglowLear: yeah, it does seem slightly slower, i wonder if i've removed some iram or something
21:49:41preglowuses 35kb as it is
21:50:11preglowalmost all of it in windows and mdct stuff
21:50:43LearI tested a 170 kbps file a while ago, and got ~140%.
21:50:52preglowtesting svn for the same file now
21:50:56preglow152% realtime
21:50:57 Part pradin
21:51:12preglowstill not very far from useless
21:51:14*preglow kicks faad
21:52:56linuxstbDoes faad support a lot more profiles than the other decoders (namely the helix one)?
21:53:31preglowyes
21:53:42preglowbut that's rather a moot point
21:53:47pregloweven the main profile never sees use
21:53:49LearI think it does more, but only PS is really interesting (and what we have enabled in Rockbox).
21:53:55preglowand when i say never, i mean it, i've never seen a main profile file
21:54:00preglowand they're all disabled in rockbox right now
21:54:05preglowLear: and sbr
21:54:18linuxstbI'm just wonderign if a "faad lite" library would be easier to optimise - i.e. delete everything we don't want.
21:54:19LearYes, but Helix does SBR, doesn't it?
21:54:33preglowlinuxstb: i don't think so, i think faad's memory usage is fundamentally broken
21:54:36preglowbut i might be wrong
21:54:38preglowLear: yup
21:54:42preglowbut helix has that license thing
21:54:46LearI know.
21:55:00preglowi'd love to use helix
21:55:20linuxstbI'm still not sure exactly what the issue is with the helix license...
21:56:05 Quit JdGordon ("CGI:IRC")
21:56:10preglowi sure as hell don't know
21:56:18preglowwould love to find out we don't need to worry about it...
21:56:33 Quit desowin ("use linux")
21:56:37Zagorlinuxstb: it requires that derivative works be licensed under the terms of the RPSL, and mandates that any litigation take place in Seattle, Washington.
21:56:47preglowi love that last part
21:57:03Zagoryeah, it's a gem
21:59:05LearHm, from license page at helixcommunity.org: "Under the terms of the GPL, RPSL or the RCSL, you may license the Helix DNA for either Research & Development Use and/or Commercial Use". Don't really know what to make out of it yet...
21:59:15linuxstbZagor: It also talks about software modules developed independently of the Real source code being distributable under a "compatible license". So that's not good enough?
21:59:26scorche|wwell, the client is under the GPL...it is just the server
21:59:41LloreanLear: If that means exactly what it says, you can license Helix under the GPL for Commercial use.
22:00
22:00:06Zagorlinuxstb: doesn't that still mean the helix code must be RPSL?
22:00:21LearBut not non-commercial? Sounds odd...
22:00:22linuxstbZagor: Yes, but doesn't that happen already, with for example BSD code we use?
22:00:35Zagorthe difference is that bsd is gpl compatible
22:00:44scorche|wi assume we are just talking about the server code?
22:00:55 Join Frazz [0] (n=Fraser@thelawsons.plus.com)
22:01:53linuxstbscorche|w: We're talking codecs, which are under the RPSL or RCSL IIRC.
22:01:56LloreanLear: Do they define what they mean by "Research & Development Use" because I'd certainly say Rockbox is at least "Development use" ;)
22:02:46*preglow gets his hopes up
22:02:57 Quit scorche|w ("CGI:IRC")
22:03:13 Join scorche|w [0] (n=42c007b2@rockbox/administrator/scorche)
22:03:52linuxstbLear: What page are you reading? The AAC codec license info is here - https://datatype.helixcommunity.org/2005/aacfixptdec
22:04:03scorche|wlinuxstb: ah...i was thinking that was under the client's licent for some odd reason...
22:04:08LloreanLear: But it looks like that's just a list of what their licenses are, meaning "They may be one, all, or any combination of these, you have to look to see which"
22:06:26Learlinuxstb: On the licensing page, linked from helixcommunity.org.
22:08:07Nico_Pwhat about this: https://helixcommunity.org/content/complicense ?
22:08:10preglowi'm tempted to just go ahead and make it into a rockbox codec and just not redistribute it, just to see how efficient it is
22:08:25 Join przemhb [0] (n=przemhb@fan115.internetdsl.tpnet.pl)
22:08:35Bagderpreglow: you can distribute the patch that makes it into one ;-)
22:08:36przemhbhi
22:08:45preglowBagder: heh, true
22:08:49 Join rockboxftw [0] (n=raul@71-223-148-25.phnx.qwest.net)
22:08:56*Bagder recalls how lame started...
22:08:58linuxstbLear: IIUC, the "DNA components" don't include the codecs.
22:09:08rockboxftwhow can i get the bootloader on rockbox?
22:09:26james01copy it to the drive
22:09:29LloreanNico_P: Basically that says "You can attach GPL code to RPSL code, but you can't attach RPSL code to GPL code, because you can't change the license on the RPSL code to the GPL"
22:09:31linuxstbrockboxftw: By following the install instructions in the manual.
22:09:48Nico_PLlorean: ah ok. doesn't that settle the question?
22:09:51rockboxftwi'm looking at a completley diff loader..
22:09:58preglowoh, sweet lord, it just requires 30kb ram for decoding lc
22:10:09linuxstbpreglow: Stop looking.... ;)
22:10:12james01dd if=loaderthatyoumade of=/dev/sda
22:10:20LloreanNico_P: If it's RPSL I'd say it pretty clearly answers it, yeah.
22:10:20ZagorNico_P: see the footnote on that page
22:10:30Soaprlpowell (if you read the logs) thank you very much for filling out IpodAccessories
22:10:31Zagor"Any attempt to apply non RPSL license terms, including without limitation the GPL, to Covered Code is expressly forbidden."
22:10:43Nico_PZagor: that doesn't mean much to me
22:11:23Nico_Pbut Llorean made it clearer :)
22:11:23linuxstbZagor: But does the GPL force all code to be GPL'd?
22:11:30Bagderno
22:11:40Bagderjust not impose any further restrictions
22:11:46Bagderwhich I believe the RPSL does
22:12:02scorche|wrockboxftw: the rockbox bootloader is the only one supported here
22:12:06przemhbcould someone take a look on polish language file patch? it is very simple patch - it fixes translation of a LANG_ALARM_WAKEUP_SCREEN
22:13:03linuxstbjames01: All that will do is trash the partition table...
22:13:11 Quit petur ("switching")
22:13:21 Join petur [0] (n=petur@rockbox/developer/petur)
22:14:04przemhbFS #8205
22:14:09 Join jhulst [0] (n=jhulst@unaffiliated/jhulst)
22:14:11ZagorI don't suppose anyone has jtag on a pp target?
22:14:33linuxstbI think barrywardell?
22:14:42 Part webguest30
22:14:46scorche|wZagor: the sansa devboard?
22:14:48linuxstbAnd isn't there a Sansa dev board somewhere?
22:15:09mooslinuxstb: he have for H10 no?
22:15:09 Quit Lear ("ChatZilla 0.9.79 [Firefox 2.0.0.11/2007112718]")
22:15:28linuxstbmoos: I think so, yes.
22:15:49scorche|wlinuxstb: i think it is gathering dust in some corner of Bagder's desk ;)
22:15:52BagderZagor: http://daniel.haxx.se/sansa/e200-devboard.html ...
22:15:59Bagderlots of dust on that...
22:16:00james01then its sda1 or somethin
22:16:02preglowhigh time that gets used
22:16:04Nico_Plinuxstb, preglow: http://lists.helixcommunity.org/pipermail/open-licensing/2005-March.txt the mail from robla at real.com is interesting
22:16:06james01I dunno read the manual
22:16:33ZagorBagder: I might find use for that
22:16:36rockboxftwdoes the ipodpatcher bootloader work for rockbox?
22:16:57BagderZagor: you're welcome to play with it as much as you like!
22:17:22 Join TMM [0] (n=hp@ip565b35da.direct-adsl.nl)
22:17:30scorche|wrockboxftw: our version includes the rockbox bootloader
22:17:40scorche|wZagor: please do :)
22:17:57Nico_Pmaybe we could send a mail to real
22:18:05ZagorI get a nasty freeze when enabling DEBUG on target builds.
22:18:16rockboxftwrlly... how do i access it
22:18:16Zagorduring boot, before the logo
22:18:39Lloreanjames01: It's strongly suggested that you not tell people to do things unless you're sure they're the right thing to do... Especially when they're directions like the one you gave that would corrupt the iPod.
22:19:14Lloreanrockboxftw: See the manual install instructions in the manual, and they tell how to use iPodpatcher to install the Rockbox bootloader. It's our tool, after all.
22:19:18rockboxftwis there a key combo for it?
22:19:20james01not like its terribly hard to fix
22:19:38Lloreanrockboxftw: Key combo for what? Ipodpatcher is a program you run on the computer to install a bootloader...
22:19:52moosNico_P: nothing to lost in trying...
22:19:52james01isnt there a site that hosts all the untouched mbr's
22:20:12rockboxftwi know, i'm talking about the boot menu
22:20:30Lloreanrockboxftw: The Rockbox bootloader doesn't have a boot menu. If you want help with some other bootloader, ask whoever is providing it.
22:20:44Nico_Pif we do mail real, which I think we should, we should write directly to Rob Lanphier. He seems to be the open source guy
22:20:44 Quit Frazz (Read error: 110 (Connection timed out))
22:20:58pixelmaBagder: isn't that the thing on top of teh Tower? :)
22:21:11Lloreanjames01: Which of course is perfectly easy for someone who doesn't know the dd method of installing a bootloader to do. Please, just don't tell people to do stuff unless you check it out first.
22:21:16Nico_Poh crap he doesn't seem to work at real anymore
22:21:17Bagderthe crown jewel!
22:21:19Bagder:-)
22:22:36moosan angel on top of the christmas fir (that's the period :)
22:23:17scorche|wa device that has been screaming out to be used
22:23:38preglowlinuxstb: stripping away unneeded parts from faad shouldn't be too hard, though, but would of course not make syncing any easier
22:24:26rockboxftwis it possible to make rockbox only play music
22:24:35rockboxftwand get rid of all the extras other then the "database"
22:24:50Lloreanrockboxftw: Why?
22:24:59rockboxftwmy sansa e200
22:25:02rockboxftwscreen broken
22:25:05rockboxftwcan't see anyting
22:25:13rockboxftwwould be nice to hit middle a few times and play musix
22:25:18LloreanUse the spoken interface to get things set up.
22:25:37LloreanAnd the "Play" button resumes playback, so once you have a playlist of all music it's rather easy to just use auto-resume when you turn on, or just tap play
22:25:42LloreanAll of which is described in the manual
22:26:21rockboxftwok, thanks do all the mp3players u guys support have same interface?
22:26:42scorche|wdepends what you mean by "interface"
22:26:48LloreanRockbox has a fundamentally identical interface on each of them, visually speaking. The controls are always a bit different because there's variation of hardware
22:26:50rockboxftwrockbox interface..
22:27:02rockboxftwok cool thnx
22:31:40Nico_PBagder, preglow, linuxstb: licensing@helixcommunity.org is the address for licensing questions... I'd send them one if I knew what to say
22:32:10rockboxftwhave good day everyone and thanks for standing my newbie ness..
22:32:20rockboxftwwhen comes to ipod mod or moding crap i suck..
22:32:56 Part rockboxftw
22:36:03 Quit Siku ()
22:39:02moosNico_P: maybe let's one of the trium vira people made it ?
22:39:47moosI mean swedishes of course ;)
22:41:17Zagorwhat do we want to know? "hi can you please change your license?"
22:41:37mooshehe :)
22:41:57moosclarifications at least, no?
22:42:12Zagorclarification of what?
22:42:16Nico_PZagor: most of what I read seems to indicate they are pretty open to the GPL, but that AAC has patent issues in the US. Maybe they'd be willing to let us use the decoder under the GPL as rockbox is based in sweden
22:42:24Nico_PI think it's worth trying
22:42:46moosNico_P: nothing to lost indeed
22:42:49***Saving seen data "./dancer.seen"
22:43:14Zagorif they let us license it under GPL the cat will be out of the bag, and then everyone in the world will get it under gpl. they know that full well.
22:43:36moosthat's true, open gate
22:43:50Nico_Pmaybe they don't mind. asking for at least clarifications can't hurt
22:44:06Nico_Pmost of the helix code seems to be available under the GPL
22:44:20Zagorasking for dual licensing is the only thing that makes sense in my mind
22:44:21LloreanMaybe ask them something like "We have a GPL licensed program linked to various audio codecs, all of which so far are GPL license. We can't see any legal way to include RPSL licensed codecs, but at the same time acknowledge that your AAC decoder is probably the most suited to our needs, and were wondering if there was a known solution to this problem?
22:47:20 Join safetydan [0] (n=safetyda@rockbox/developer/safetydan)
22:47:25ZagorLlorean: sure, go ahead. your english is better than mine.
22:47:33LloreanWhat's the email address?
22:47:46scorche|w<Nico_P> Bagder, preglow, linuxstb: licensing@helixcommunity.org is the address for licensing questions... I'd send them one if I knew what to say
22:47:47Lloreanlicensing@blah.org?
22:47:49LloreanOkay
22:48:05scorche|wme realises he just highlighted 4 people =/
22:48:54linuxstbLlorean: Not all our codecs are GPL'd though - some are LGPL, some BSD...
22:50:07Lloreanlinuxstb: I've changed it to "licensed in a manner compatible with our GPL code"?
22:50:44Lloreanhttp://pastebin.ca/806739 look good?
22:50:51linuxstbLlorean: I think the main point is that Rockbox itself is GPL'd and we don't consider codecs as separate entities, they're closely linked to the core.
22:51:52LloreanMaybe changed "linked to various audio codecs" to "with various audio codecs compiled into it"?
22:52:04linuxstbLlorean: I think that email is fine as is.
22:52:19LloreanAlright then.
22:52:28LloreanI'll send it on, and hope they get back to us.
22:52:50linuxstbYou could give the Rockbox URL though.
22:53:01LloreanI just added that in. :)
22:53:07LloreanI was thinking the same thing
22:53:37LloreanShe's off
22:53:52BigBambiGod bless all who sail in her
22:54:05Nico_Pcool :)
22:54:56ZagorBagder: what is REALBIN in apps/Makefile? the comment "this is not needed to get built when doing debug builds" looks worrying
22:55:13Zagoryet I do get a mi4 built
22:56:29Domonokyis there a function in the plugin api to get the current position of the playing audio file ?
22:57:20Nico_PDomonoky: audio_current_track()->elapsed
22:57:33BagderZagor: I can't see why a debug build would build an mi4...
22:57:44DomonokyNico_P: nice, thanxs..
22:57:46Bagderwith that makefile code I mean
22:58:39ZagorBagder: well, it does
22:59:07Zagormaybe because I run "make bin" explicitly?
22:59:29Bagderah yes
22:59:44Bagderthat builds the BINARY explicitly
23:00
23:01:44Zagorcould something else in the makefile be to blame for my failing DEBUG target build?
23:02:09ZagorI'm trying to activate DEBUGF over usb
23:02:33BagderI don't think I've ever tried a DEBUG build...
23:02:47*ender` yawns
23:02:54 Join mud-rb_ [0] (n=mud@dialup-4.156.9.60.Dial1.Boston1.Level3.net)
23:03:00Zagorno they're quite tightly coupled to the sim. I've had to change a couple of source files just to make it compile.
23:03:10 Part przemhb
23:04:40 Quit Domonoky (Read error: 104 (Connection reset by peer))
23:04:41scorche|wender`: you know...all i ever see you do is yawn...
23:04:54ender`well, that's what i do most of the time :)
23:05:09*Nico_P thought ender` might be a bot
23:05:24*ender` 'd like to see a yawning bot :)
23:05:56*Llorean goes to whip up yawnbot 0.1
23:06:29scorche|wi was thinking seriously about making a borkborkbot at one point in time for this channel
23:06:43ender`that reminds me, i did write a yawnbot to replace me on a DC hub a few years ago when i went skiing for a week
23:08:56 Quit bertrik ("bye")
23:09:44*preglow wonders if he should commit synced libfaad
23:11:21 Quit Nico_P (Remote closed the connection)
23:12:32linuxstbpreglow: Why wouldn't you?
23:13:15 Quit J3TC- (Read error: 110 (Connection timed out))
23:13:24Zagorwhat it STUBOFFSET used for?
23:13:25Zagoris
23:13:35linuxstbgdb stub?
23:13:59Bagdersounds like a fair guess
23:14:12linuxstbDon't ask me any follow-up questions though... ;)
23:14:14Zagorah, that could really cause confusion
23:14:24Bagdera gdb stub will be a nice addition for you Zagor ;-)
23:14:38Zagortomorrow :-P
23:14:51*Bagder looks to the sky and dreams of gdbing on target...
23:14:51 Join Nico_P [0] (n=nicolas@rockbox/developer/NicoP)
23:15:06 Join BRi7Xtra [0] (n=bri7x@c-24-127-157-38.hsd1.pa.comcast.net)
23:15:34Zagorthose were the days
23:15:51Zagorah, yeah that was the problem. now it boots nicely.
23:16:22linuxstbZagor: What are you working on?
23:16:29BRi7Xtrahey folks, anyone familiar with the gigabeat f port of rockbox? i have these headphones with a faulty connector (and also, the gigabeat jack is quite faulty at this moment as well)... basically what's happening is every time it jiggles, the song pauses.. .i'm aware of the pause on headphone unplug feature, and i've turned it completely off, but for some reason it's still doing it... any ideas?
23:16:32Zagorlinuxstb: debugf over usb
23:16:38Nico_Pwhat's required for on target gdb? serial communication?
23:17:10BagderNico_P: yes
23:17:37Nico_Pisn't that (almost?) working in the usb stack?
23:17:54Zagorlookie lookie
23:18:06Zagorflush_fat()
23:18:10ZagorSectors written: 1
23:18:18preglowlinuxstb: good question, i think i'll just kill some warnings, then
23:18:19Bagderoooh
23:18:22Zagorsome string cutoffs though
23:18:49preglowlinuxstb: and btw, i think the better approach is to go as far with keeping the code unaltered as we can, even to the effect of disabling warnings in gcc
23:19:03preglowlinuxstb: but i don't think that's the way we've usually done it
23:19:27preglowi see there are a ton of int -> uint changes in our svn faad which i haven't applied yet
23:19:36ZagorNico_P: doing it over usb is a chicken and egg problem. you'd need the target running before you can start it from gdb :)
23:19:56linuxstbpreglow: Have you noticed any useful improvements?
23:19:56Nico_Pah
23:20:21preglowlinuxstb: well, it's a wee bit faster, but that might just be my optimizations
23:20:36 Quit mf0102 ("Verlassend")
23:21:26 Nick parafin is now known as parafin|away (i=parafin@paraf.in)
23:21:39preglowlinuxstb: apart from that i don't really have much experience with aac, and not many test files
23:24:15 Quit mud-rb (Read error: 110 (Connection timed out))
23:24:49linuxstbpreglow: If I had the time and nothing else to do, I would be tempted to use libfaad as the basis for a new AAC decoder - i.e. strip out all the code we don't use and restructure it to be more efficient.
23:25:13LloreanSoC proposal?
23:25:16jottafter how many ticks (approx.) yield should be called within a plugin, to ensure smooth audio playback and how big is the actual overhead (when calling it too often)?
23:25:23LloreanGah, SoC needs to not mean "System on a Chip" as well
23:26:04preglowlinuxstb: well, me too, but i don't know if i can be bothered
23:26:19preglowi should start coding stuff i'll actually use, and then i'd be more tempted to fix up tremor
23:27:16linuxstbpreglow: But I think it's worth seeking - if someone ever did that, it would be nice to start with the latest libfaad.
23:27:49 Quit mud-rb_ (Read error: 110 (Connection timed out))
23:28:00linuxstbs/seeking/syncing/
23:28:10 Join mud-rb_ [0] (n=mud@dialup-4.156.9.79.Dial1.Boston1.Level3.net)
23:28:49 Join karashata [0] (n=karashat@207.61.208.119)
23:29:24preglowyeah
23:29:28preglowit spews errors, though
23:29:32preglowwarnings
23:34:21 Join Robin0800 [0] (n=Robin080@cpc2-brig8-0-0-cust498.brig.cable.ntl.com)
23:35:43LloreanZagor: Real has responded
23:35:51preglowthat was quick...
23:35:53LloreanYes
23:36:13LloreanHe's familiar with our project and would like to set up a time to talk about this over the phone.
23:36:27 Quit Arathis ("Bye, bye")
23:36:29preglowwith whom?
23:36:42Zagorinteresting
23:37:10LloreanScott Nelson, "Dir Bus Dev IP Lic"
23:37:28 Nick james01 is now known as HeroinHero (n=james@EV-ESR-74-215-26-69.fuse.net)
23:37:53preglowi mean, with which rockbox person. you?
23:38:13toffe82BRi7Xtra: this function doesn't disable the remote , so if you have a bad contact on the jack it can simulate a pause or othre function of the remote
23:38:22LloreanAt the moment, Me, because I'm the one who contacted him, but I'm quite certain I could redirect him to anyone who might be better suited than I.
23:38:55preglowLlorean: well, i can't think of any qualifications that would be needed that you don't have
23:39:30preglowso unless you sound incredibly hostile and rude over the phone or something, it should be fine :)
23:40:09ZagorLlorean: yeah, you talk with him. and if either of you wants to bring a swede into the conversation I'm available
23:40:21linuxstbIt sounds promising though - at least it's not an immediate "no, you won't be able to use it".
23:40:45preglowi can definitely live with people wanting to speak to us
23:40:53LloreanAlright, I'll set up a phone meeting and see where we stand, then if things get over my head one way or another suggest someone else I think he might better talk to.
23:41:43LloreanOne question though, what's the important distinction with the way our plugins and/or are linked that forces them to be GPL compatible, unlike programs that use an OS API?
23:41:53Lloreanplugins and/or codecs are linked, rather
23:42:01 Quit OlivierBorowski (Remote closed the connection)
23:42:36preglowoh, don't ask
23:43:04preglowwe call host functions through struct pointers, linux programs use interrupts
23:43:21preglowours is slightly more hardwired than the other
23:43:27preglowbut i don't know where one draws the line
23:44:50 Quit jhulst ("Konversation terminated!")
23:45:36 Quit Suleeto ()
23:45:44BRi7Xtratoffe82, oh no... for some reason i thought it was that, because sometimes it would change the track and i was all like "wtf?"... is there any way to disable the remote functions then? hmm, wait.. does a remote even come with gigabeat?
23:46:08*Robin_Watts waves
23:46:10 Part Robin_Watts
23:46:24linuxstbBRi7Xtra: You could compile your own version of Rockbox, and disable support for the remote.
23:46:32markunBRi7Xtra: mine came with a remote
23:47:02BRi7Xtrahmm, if that's the case i'll plug my headphones into the remote and just do it like that
23:47:29Lloreanpreglow: Well it was more, "is there a hard line we know we're on one side of, or am I going to have to say 'We're not sure of a license-specific line we're crossing, and have decided for ourselves that we consider this linking code in the GPL sense'"?
23:47:57BRi7Xtraheheh, when i first got this, i tore open the box, went right for the essentials (gigabeat, usb cable, power cable, dock [not even the latter two the second time around]), turned it on, laughed at the crappy existing firmware, then loaded up rockbox
23:48:07BRi7Xtrai'll check the box though
23:49:15preglowLlorean: someone more knowledgable than me should answer this
23:49:20*preglow trips Zagor
23:49:48LloreanI'm just worried that how we link codecs will come up, and what to know, more or less, our position there.
23:50:03ZagorLlorean: the linux kernel people have washed this issue inside out, and came to the conclusion that since kernel modules (codecs and plugins) are compiled for specific versions, they are derived works.
23:50:03 Quit HowdyDoody (Read error: 104 (Connection reset by peer))
23:50:05preglowwell, it's a valid concern
23:50:12 Join HowdyDoody [0] (n=Administ@c-71-225-221-149.hsd1.pa.comcast.net)
23:50:30preglowZagor: and that surely applies to us as well, most of the time
23:50:44Zagoryes, we're pretty much in the same situation as they are
23:50:49LloreanOkay
23:50:51LloreanThat's what I needed to know
23:50:54LloreanThanks
23:51:05amiconnZagor: What I don't understand at all is where to draw the line?
23:51:07 Join fenugrec [0] (n=ABC@206.167.203.215)
23:51:15preglowwell, that's a very clear line, if you ask me
23:51:17 Quit ompaul (Client Quit)
23:51:23amiconnWhy is it allowed to run gpl incompatible applications on linux?
23:51:44Zagorbecause they are not version specific
23:52:05toffe82BRi7Xtra: there was the remote only with the F40 , not the f10 or f20
23:52:07amiconnhmmm
23:52:36BRi7Xtrayeah i've got the f40
23:52:46Zagorit _is_ a gray area though
23:53:00amiconnThing is: Right now we have a rockbox core, which is gpl'd, and plugins need to be gpl'd too because they use the api blah blah
23:53:12LloreanSo if we standardized our API to the point that codecs could be compiled independently, it'd be different?
23:53:18preglowLlorean: aye
23:53:25preglowif we standardized the interface and never broke it
23:53:30LloreanOf course, that's back to relocatable plugins
23:53:31amiconnBut afaiu, introducing a "license compatibility glue layer", e.g. lgpl'd, would fix that
23:53:41preglowamiconn: that's libc in linux, afaik
23:54:25preglowLlorean: yeah, the fact that our plugins are statically linked enhances the point of them being derived works
23:54:32amiconnWhy is that? The interaction still happens between the core and the plugin, it's just passed through the compatibility layer. There's no visible difference for the user
23:54:34Zagoramiconn: no it doesn't. that just makes the source code clean. the compiled binary is still a derivate though.
23:55:20preglowamiconn: which reminds me, any showstoppers for relocating plugins/codecs to start of ram, or is it just that noone has done it yet?
23:55:56amiconnZagor: In what way is linux different? The kernel is gpl'd, so no matter how many layers are between the kernel and the application, one could consider a linux application derivative work
23:57:21 Quit MethoS- ("Konversation terminated!")
23:57:22amiconnpreglow: No one has done it, and I'm not sure about the best way to handle the partial relocation
23:57:37preglowamiconn: partial relocation?
23:58:48amiconnWell, either the bootloader would need to know that the binary must be loaded at an offset, or the binary needs to relocate itself
23:58:55Zagoramiconn: the syscall interface is designed precisely to allow applications to run without including any operating system code. the plugin (or kernel module) interface is not like that.

Previous day | Next day