**** BEGIN LOGGING AT Thu Apr 07 02:59:59 2011 Apr 07 03:02:48 howdy Apr 07 03:20:44 morning Apr 07 03:51:50 hi, Apr 07 03:52:15 just read the "how to get started" from meego Wiki.... Apr 07 03:52:42 how do I gain account for OBS system? any body can help? Apr 07 03:56:30 joseph__: community obs? Apr 07 03:56:47 i think yes Apr 07 03:57:13 you need to ping lbt about it Apr 07 03:58:18 hum.. thanks timoph Apr 07 03:58:26 I'd guess he's still sleeping. It's bit early in europe Apr 07 03:59:22 ok, maybe later! Apr 07 05:13:41 hello? Apr 07 05:14:56 hi Apr 07 05:21:33 whats the deal with using meego on tablets? Apr 07 05:22:06 "it works" Apr 07 05:22:27 is there a premade image for it Apr 07 05:22:55 like there is for netbooks and handsets etc Apr 07 05:23:09 check out the website. Apr 07 05:23:13 i did Apr 07 05:23:15 theres a live image you can download Apr 07 05:23:25 donde? Apr 07 05:23:43 cuz im in the downloads section and see nothing Apr 07 05:25:47 http://repo.meego.com/MeeGo/builds/trunk/1.1.99.1.20110405.3/images/meego-handset-armv7l-n900/ http://repo.meego.com/MeeGo/builds/trunk/1.1.99.1.20110405.3/images/meego-tablet-ia32-pinetrail/ http://repo.meego.com/MeeGo/builds/trunk/1.1.99.1.20110405.3/images/meego-netbook-ia32/meego-netbook-ia32-1.1.99.1.20110405.3.img Apr 07 05:34:40 Or even http://meego.com/downloads/releases/1.2/meego-tablet-developer-preview Apr 07 05:41:53 Stskeeps : hi :) Apr 07 07:01:40 Packaging Error: Command 'rrpmbuild -bb meegoimengine.spec' failed.Exit code: 2 Apr 07 07:01:53 rrpm? Apr 07 07:02:34 when trying to build "meego-inputmethodengine" from Qt-creator Apr 07 07:03:13 Qt-creator invoked in Ubuntu10.04 by using meego-sdk-qtcreator Apr 07 07:21:57 anyone able to build any packages/application on qt-creator and launch them in qemu successfully Apr 07 07:36:49 morning Apr 07 07:41:38 morning Apr 07 07:53:30 good night! Apr 07 07:54:38 anyone wanna place bets on the number of additional rounds of ping pong remaining for https://bugs.meego.com/show_activity.cgi?id=15207 ? :) Apr 07 07:56:45 impressive :-( Apr 07 07:59:59 andre__: it's best that it happen w/ bugs filed by someone like me Apr 07 08:00:04 instead of an actual random person Apr 07 08:00:11 that'd be really frustrating for that person Apr 07 08:16:01 alterego, ping Apr 07 09:46:48 HI is there some way by which I can genrate a ks file for a particular meego image Apr 07 09:49:03 when I boot my livesub image for meego1.1 IVI X86 i just get a balck screen saying Boot: can somebody suggest wat may be the sisue Apr 07 09:51:52 Venemo: yeah? Apr 07 09:58:11 Hiiii Apr 07 10:03:16 hello. is there an updated installable tablet UX image for the S10-3t? Apr 07 10:04:00 on repo.meego.com maybe Apr 07 10:04:51 wouldn't it be at http://repo.meego.com/MeeGo/builds/trunk/latest/images/meego-tablet-ia32-pinetrail/ ? Apr 07 10:05:15 Guest98320, i think you hit the nail on the head. but does that work on the s10-3t? Apr 07 10:05:40 have there been changes to mic2 recently .. am getting errors after i updated Apr 07 10:07:16 arfoll: I don't know, I've been building pinetrail images and they've worked on the lenovo at least Apr 07 10:08:04 w00t_, thanks i guess i have to try, its just my internet is sooo slow Apr 07 10:08:32 :-) Apr 07 10:26:38 http://pastebin.com/YJmNm5Fi Apr 07 10:26:54 My .rpm does not contain the .so file Apr 07 10:28:15 But install command shows, .so file is created and placed in "rrpmbuild/instroot/usr/lib/libmeegoimengine.so" Apr 07 10:28:42 can someone point out what is going wrong Apr 07 10:29:18 mikhas: ^ maybe you have a clue Apr 07 10:33:17 hey ..i could see; all sub dir has separate rrpmbuild directory Apr 07 10:33:40 and outer rrpmbuild is used to create the .rpm file Apr 07 10:33:47 which contains nothing Apr 07 10:35:22 so can some one point out how to club all these "rrpmbuild" so that the .rpm which is created out of "rrpmbuild" contains all required .h and binaries Apr 07 10:52:55 bkalinga: a) .h and .so go into -devel subpackage b) most, if not all of the people seriously working on the distribution either write spec files manually, or use spectacle, so I don't expect that you'll find many people capable of helping with your problem Apr 07 10:55:00 where is this -devel subpackage located Apr 07 10:55:14 see b) :) Apr 07 10:55:27 i see only one .rpm file in the outer most directory Apr 07 10:55:51 where are there are multiple rrpm folders in the sub directories Apr 07 10:56:18 *where as there are multiple rrpm folders in the sub directories Apr 07 10:57:20 so you mean..everything can be controlled though modifying the spec file? Apr 07 10:58:02 of course. the spec file contains all the information for building a rpm package Apr 07 10:58:07 when I boot my livesub image for meego1.1 IVI X86 i just get a balck screen saying Boot: can somebody suggest wat may be the issue Apr 07 11:05:40 w00t_ what? Apr 07 11:05:52 ugh, .rpm related questions Apr 07 11:06:09 I left RPM behind in 2003, for good I had hoped Apr 07 12:32:35 rustylynch, we just decided to remove reactionmap functionality from meego-keyboard, in MeeGo 1.2 Apr 07 12:32:39 happy now? ;-) Apr 07 12:32:58 reasoning: it is pretty much non-functional on MeeGo 1.2 anyway Apr 07 12:33:03 hehehe Apr 07 12:34:39 was that the immersion plugin place? Apr 07 12:36:40 for the vibra? yeah Apr 07 12:38:41 ah Apr 07 12:39:13 Stskeeps, if someone could create some pressure that such backend is needed in MeeGo ... ;-) Apr 07 12:40:24 hopefully meego can do away with framework abuse.. ie frameworks that only really work if certain closed source bits are there.. Apr 07 13:19:03 alterego, do you plan to enter the coding competition? Apr 07 13:51:29 this is rather annoying, and I don't exactly know where to start slicing Apr 07 13:52:33 Bostik: Aim for the heart. Apr 07 13:52:35 given the latest 1.1.90 snapshot (which contains qt-4.7.2), I'm trying to port the qt.spec towards future; it is a bit tricky when even trying to build the provided .src.rpm against the very .spec found in the package fails Apr 07 13:54:54 Bostik: so... rpmbuild --rebuild qt-4.7.2-x.y.src.rpm fails to build from source? Apr 07 13:55:00 build error occurs in src/dbus/qdbusintegrator.cpp and is a simple error from invalid cast but as far as I can see, it's not from any of the patches Apr 07 13:55:49 gabrbedd: so far I've done the pre-steps manually - unpacked .src.rpm, put sources+patches+icons in rpmbuild/SOURCES/ and qt.spec in rpmbuild/SPECS Apr 07 13:56:01 but just to be sure, I'll give that exact command a run Apr 07 13:56:43 Bostik: Well that SRPM was created by OBS... so it's the result of a successful compile. Apr 07 13:57:04 indeed, and that's why I'm scratching my head Apr 07 13:57:10 Bostik: I think the logs are available so that you can maybe see what's different between your build system and theirs. Apr 07 13:57:25 Bostik: are you building on MeeGo... or in a MeeGo chroot? Apr 07 13:58:32 gabrbedd: inside a meego-only VM Apr 07 13:58:47 so it's on meego, certainly Apr 07 14:00:14 Bostik: perhaps paste the error messages? Apr 07 14:01:55 the actual message is "error: invalid conversion from 'const void*' to 'void*'", so it might be some kind of default build flag which is passed to g++ in OBS but not here Apr 07 14:02:32 I'll put the full error in pastebin so you can see the context Apr 07 14:02:46 Bostik: also need to paste the code it refers to. Apr 07 14:03:19 oh yes, getting that too :) Apr 07 14:04:02 Bostik: My guess is that it's probably a version mismatch in a 3rd party lib or some kind of compiler switch. Apr 07 14:04:22 Bostik: ...or you have an "optional dependency" installed that wasn't installed in OBS -- thereby activating the feature. Apr 07 14:07:47 http://pastebin.com/H723f0aS Apr 07 14:09:21 and exactly same error from "rpm --rebuild" too Apr 07 14:14:32 Bostik: I don't see your compiler flags in that log. Those are important. Apr 07 14:15:19 Bostik: And what arch are you compiling for? Apr 07 14:16:05 hmmmmm, that rings a bell... Apr 07 14:16:23 just a minute, it could be something to do with arch specification Apr 07 14:28:07 also seems to be a case of my VM environment not setting any explicit flags at all Apr 07 14:28:13 that would explain quite a lot Apr 07 14:29:19 Bostik: In my build environment, I found that I had to set up RPM's build flags.... /etc/rpmrc or something. Apr 07 14:30:18 gabrbedd: thanks, that's one more place to look at then Apr 07 14:31:38 Bostik: I dunno if that'll fix your problem... but it's a start. Be sure to also check the build log for Qt from OBS for more clues... Apr 07 14:31:58 oh yes, I have that one already open for comparison purposes Apr 07 14:32:55 ok, I won't paste the link, then. :-) Apr 07 15:07:11 Venemo: probably yeah, Apr 07 15:08:01 alterego, and what's your app going to be? :) Apr 07 15:08:50 Not sure yet :) Apr 07 15:09:15 I'm thinking of doing a Qml based Guitar tuner and chord dictionary Apr 07 15:09:27 alterego, I'm also wanting to code something, but haven't got a good idea yet :) Apr 07 15:09:34 If I have time maybe some recording functionality. Apr 07 15:09:42 alterego: not going to work on the dialer anymore? Apr 07 15:10:00 w00t_: not for the coding competition no :P Apr 07 15:10:17 w00t_: but yes, of course I am going to continue the dialer :) Apr 07 15:13:41 * timeless_mbp pokes dneary Apr 07 15:15:47 alterego: Were you thinking of doing it dial-style... or a strobe tuner? (i.e. I'm interested...) Apr 07 15:19:31 pulse dial is the way to go, dial tones are so 1990s Apr 07 15:48:36 gabrbedd: dial style, though might make a couple of "themes" Apr 07 15:50:32 alterego: If you want to collaborate... lemme know. Apr 07 15:51:09 gabrbedd: cool :) Apr 07 15:51:52 I actually implemented my own fft module already as I was going to do it a while ago for maemo. Apr 07 15:52:17 But never really got anywhere other than displaying the most dominant tone on screen in hz Apr 07 15:53:06 Don't need fft for that, I guess Apr 07 15:54:00 RST38h: seemed the best solution to me, what would you suggest? Apr 07 15:54:28 I use a simple algorithm that counts 0-crossings Apr 07 15:54:48 Got very specific input though Apr 07 15:58:15 alterego: Yeah, you have some options other than fftw... also I wonder how well fftw will work on an ARM. Apr 07 15:58:33 gabrbedd: mine worked pretty well Apr 07 15:58:41 alterego: FWIW, fmit uses fftw... gtkguitartune is doing something else. Apr 07 15:59:33 * gabrbedd is NOT lobbying against fftw Apr 07 16:00:27 :) Apr 07 16:00:50 I just really enjoyed learning about fourier and it seemed like the most elegant solution. Apr 07 16:01:11 I was thinking of implementing it in glsl too :D Apr 07 16:03:46 how do i craete initrd in meego? Apr 07 16:03:58 I was just reading the wikipedia article on strobe tuners... and they work like RST38h was suggesting... switching the strobe light at zero crossovers. Apr 07 16:04:28 naquad: meego doesn't do initrd, so if you want to do that then you're charting new territory. Apr 07 16:04:46 gabrbedd: yeah, I remember reading about the too. Apr 07 16:05:04 I might write a quick implementation of one of those and see which works best. Apr 07 16:05:10 naquad: Instead of initrd, comple stuff you need in-kernel instead of as modules. Apr 07 16:05:52 gabrbedd, i want to create pxe boot, my idea: nbd-client (network block device client) in initrd mounting remote disk as /mnt and doing pivot_chroot there, usual boot process continues Apr 07 16:10:42 hildon-input Apr 07 16:10:48 err. Apr 07 16:24:04 * arjan wonders if there's a telepathy person here Apr 07 16:24:52 * tomeu bets there is Apr 07 16:27:46 vgrade: were you able to see anything interesting with the archos9-tablet image? Apr 07 17:10:33 vgrade: hi...i was browsing youre folder to try out the latest n900 image, and came across joggler images...i also happen to have one of those, what image is best? Apr 07 17:35:35 hi Apr 07 17:37:03 I've heard that amd has joined Meego. Does that mean that Meego 1.2 will include radeon driver ? Apr 07 17:42:30 vlj: Yes, they joined last fall/winter at the MeeGo conference -- but since then I (personally) have seen no action in MeeGo regarding any AMD CPU's or GPU's. Apr 07 17:43:05 vlj: I don't know if they've allocated any engineering resources to MeeGo -- but my guess is that they have not. Apr 07 17:44:45 gabrbedd: in what consist their participation if they dont allocate resources on it ?? Apr 07 17:45:12 ask AMD :) Apr 07 17:45:20 vlj: That's a good question! :-) Apr 07 17:45:44 berndhs: +1 Apr 07 17:46:14 they dont work on it behind closed doors ? Apr 07 17:46:47 vlj: FWIW, Intel has invested lots of resources so that people can ship top-notch, high-performance intel products that run MeeGo. Apr 07 17:47:16 vlj: So, if AMD wants to do the same for their chips... I would think they need to do the same (and tell people about it). Apr 07 17:47:53 vlj: But whatever happens behind closed doors.... my palantir is out for repairs Apr 07 17:49:00 gabrbedd: Apr 07 17:50:01 this is sad if there nothing from amd Apr 07 17:50:55 it looked promising to have both chip maker behind a good performing os Apr 07 17:51:07 brb Apr 07 17:51:09 vlj: Why? Seems likes it's only sad for AMD. Apr 07 17:51:34 ...unless MeeGo flops -- then it's /good/ for AMD. Apr 07 18:10:06 alterego: if you want to make the tuner interesting, have it be able to tune all strings at once ( like http://www.tcelectronic.com/polytune.asp ) Apr 07 18:10:38 npm: alterego: now /that/ would require fft Apr 07 18:11:17 well you could also use discrete cosine transforms Apr 07 18:11:28 or wavelets Apr 07 18:15:15 DCT is still O(N log N) :-) Apr 07 18:16:21 are you planning on running this on an abacus or something? Apr 07 18:17:12 npm: Even worse... ARM ! Apr 07 18:17:16 (just kidding) Apr 07 18:17:29 * RST38h wonders if ARM SoCs can be called this way Apr 07 18:17:39 doh... man that's almost as much of a slag-off as the guy from AMD yesteday at the linuxcollabconf Apr 07 18:18:20 something about ARM is already in the data center.. in all the disk drives.. and that's where it should stay... Apr 07 18:19:05 somebody shoulda reminded him about MeeGo and AMD Apr 07 18:20:52 I'm just kidding... ARM is really cool. I just like to rag on their floating point, is all. Apr 07 18:21:27 aren't you supposed to recode everything for NEON? Apr 07 18:21:48 Doesn't compiler do it for you? Apr 07 18:22:18 And doesn't long (12 steps?) pipeline for Neon instructions make them kinda useless on A8? Apr 07 18:22:34 Even NEON isn't as powerful as Atom for FLOPS Apr 07 18:22:37 then why http://wiki.meego.com/ARM/NEON Apr 07 18:23:03 npm:what about it? Apr 07 18:23:06 And Atom is even kinda wimpy compared to a Core or Core2 Apr 07 18:23:33 About the same as the first Core, no? Apr 07 18:23:37 existence proof of need for recoding (see patch) Apr 07 18:23:50 At least according to the perf test results... Apr 07 18:24:17 hmm "Project not found: "devel:hardfp:prep:Trunk:Testing:neon"" Apr 07 18:24:23 npm: These are highly optimized pieces of media code that have been in assembly anyway Apr 07 18:24:24 non-existence proof ? Apr 07 18:24:40 npm: For generic math code, your compiler should do it for you =) Apr 07 18:24:49 but that's essentially what gabrbedd was talking about w/r/t FFTs Apr 07 18:25:20 tha'ts the kind of code you may need to recode to take advantage of neon arch, e.g. optimum bit widths for operations etc Apr 07 18:25:36 yea Apr 07 18:26:18 and i assume the "bad floating point" comment was w/r/t running linux audio engines that use floating point as "mixer" Apr 07 18:27:21 (which i'm not sure is a good idea sonically, but don't want to get yet another "digital summing" != "analog summing" flamewar Apr 07 18:27:46 Does not matter, if your integers allow for good resolution Apr 07 18:27:50 npm: No, it has to do with FFT's (which are O(N log N)) to transform a signal to the frequency domain. Apr 07 18:28:09 And U32 will probably be enough for most cases Apr 07 18:28:54 As far as mixing... using floats just plain makes everything easier. But in the end you can do all the same stuff with integers -- it's just more work. Apr 07 18:29:02 ...more work on the programmer. Apr 07 18:29:03 gabrbedd: and thus the math used to perform that FFT, which optimizes out different if you have floating point versus fixed Apr 07 18:29:23 not reallymore work Apr 07 18:29:49 RST38h: when adding floats, I never have to check for overflow. Apr 07 18:29:57 Same with ints Apr 07 18:30:00 I don't even have to clip them. Apr 07 18:30:05 Same Apr 07 18:30:09 How? Apr 07 18:30:17 http://www.underbit.com/products/mad/ Because MAD uses integer computation rather than floating point, it is well suited for architectures without a floating point unit. All calculations are performed with a 32-bit fixed-point integer representation. Apr 07 18:30:23 You are adding 16bit ints into a 32bit int Apr 07 18:30:44 You will have to add 65536 different sound sources before it overflows. Apr 07 18:30:55 Have you got that many? Apr 07 18:31:47 ok... and so I need to use 64-bit ints to sum my 32-bit data? Apr 07 18:31:58 usually Apr 07 18:32:23 take a look at the bit widths of pro mixing boards Apr 07 18:32:31 more or less Apr 07 18:32:41 except that you rarely get 32bit audio samples :) Apr 07 18:32:53 or for example the digital mixer in the RME's .. how wide? Apr 07 18:33:06 Right... but I still think d[i] += s[i] * gain Apr 07 18:33:13 is easier than.... Apr 07 18:33:29 gabr: Ok, 256 sources by using individual 0..255 gains Apr 07 18:33:47 gabr: 256 is still a freaking lot of audio sources though. You usually get atmost 16. Apr 07 18:34:04 or the envy24's 36 bit wide mixer for 10 sources: http://code.google.com/p/mudita24/source/browse/trunk/mudita24/README Apr 07 18:34:52 http://nielsmayer.com/npm/envy24mixer-architecture.png Apr 07 18:34:56 int64_t t foreach{ t += s[i] * gain_num / gain_den; } d[i] = t; Apr 07 18:35:12 ...or something like that. Isn't that what you'd need to do? Apr 07 18:35:23 garb: One suggestion Apr 07 18:35:29 Oh wait... I have to bit shift the t or something... Apr 07 18:35:33 garb: Divisions are really bad for your performance Apr 07 18:35:39 anyone here on mibbit? Apr 07 18:35:48 AVOID. Apr 07 18:36:03 RST38h: so how do you scale? Apr 07 18:36:09 RST38h: you work w/ 32 bit "frames" if you have 24 bit audio Apr 07 18:36:11 You can always shift t Apr 07 18:36:25 * gabrbedd only does floating point DSP.... Apr 07 18:36:41 Shifting t is the same thing as scaling the whole thingdown Apr 07 18:36:49 RST38h: So then I'm always changing the volume by -3db Apr 07 18:37:09 (or +3dB) Apr 07 18:37:09 Well, depends on how much you WANT to change it Apr 07 18:37:28 You do not HAVE to shift Apr 07 18:37:51 You can even divide rather than shift, but do divide t rather than individual components Apr 07 18:38:03 This applies to floating point numbers as well Apr 07 18:38:33 But in FP, you don't have to divide to get a fractional gain. Apr 07 18:38:51 Anyway... I know it can all be done in int... it's just there's more pitfalls. Apr 07 18:39:28 RST38h: done that way, all your fadeouts would have "zipper noise" ... that's where floating point is very helpful Apr 07 18:41:52 Maybe, but I am not dealing with high-precision audio anyway, only doing CD-quality Apr 07 18:42:59 npm: BTW I just realized my response was partially nagative about DCT and wavelet. Sorry. Those are good suggestions. :-) Apr 07 18:44:02 esp. for tuning five strings at once... and not computing imaginary plane stuff that got thrown out at the A/D converter Apr 07 18:45:10 RST38h: Well, in pro audio apps it's all floating point... which is a relaxing environment as far as that goes. :-) Apr 07 18:45:17 RST38h: what's are you doing that's CD quality. IMHO, given mp3's these days, i'd be happy with actually getting CD quality Apr 07 18:45:50 gabrbedd: in linux pro audio apps using jack.... Apr 07 18:46:17 ...and jack is a 32-bit floating point api. Apr 07 18:47:17 But even the Windows and OS X apps are using FP PCM, too. Heck, I think my Delta 1010 does 24-bit FP (but my memory may be failing there) Apr 07 18:47:34 i'm not sure that applies to those doing mixing inside an FPGA Apr 07 18:47:59 delta 1010 uses a 36 bit digital mixer Apr 07 18:48:33 RME uses 48 or 64 Apr 07 18:49:27 and sometimes i think it sounds cleaner using the envy24 digital mixer than using jack Apr 07 18:49:42 is that the sound of floating point versus fixed? Apr 07 18:49:53 the truncation dither? Apr 07 18:50:53 but i'm in a minority and usually get flamed mercilessly by fons on LAU for even mentioning it :-) Apr 07 18:50:55 npm: IIRC, while the data you send to a 1010 is 32-bit in length, it ignores the low byte Apr 07 18:51:18 yep, it's easier to waste 8 bits Apr 07 18:51:38 As for mixing quality.... can't say I've noticed (or looked for) any difference. Apr 07 18:52:37 However, when using HW monitoring -- I often notice a difference. Sometimes I think it's just gain... but other times I think it's all the A/D and D/A conversions. Dunno. Apr 07 18:53:44 npm: mixing 1-16 fm / wave / melodic audio channels Apr 07 18:54:03 npm: from emulated hardware like OPL2/OPL3 chips etc Apr 07 18:54:27 can you emulate my yamaha db50xg ? :-) Apr 07 18:55:20 What is it? A bunch of OPL2s? Apr 07 18:56:44 npm: BTW, when doing the +/- 3dB volume change thing.... I've heard that you can minimize the zipper effect by only changing volume at zero crossovers. Apr 07 18:56:57 Ah I know what it is. Not really a sound card. Just a wavetable library Apr 07 18:56:59 http://www.dancetech.com/item.cfm?threadid=328 Apr 07 18:57:59 sounds like http://nielsmayer.com/DjColtraneRexx-qchord-salsa.flac Apr 07 18:59:41 in a more widely used format http://nielsmayer.com/npm/Skyline_Herbie_Hancock_on_analog-modded-dynex+13.50db.ogg Apr 07 19:01:44 RST38h: very wrong about "just a wavetable library" ... http://qxgedit.sourceforge.net/qxgedit-index.html Apr 07 19:01:55 it's a full on synth, controllable by linux Apr 07 19:02:24 ok, so you can change instrument settings Apr 07 19:02:39 Anything starting with OPL2 will give you the same power Apr 07 19:03:02 Some less mainstream chips like YM2608/2610 too Apr 07 19:03:04 OPL2 is just an fm synth Apr 07 19:03:19 Yes.And it still lets you change these things. Apr 07 19:03:29 the db50/60 and all of yam's later synth uses AWM which FM's samples Apr 07 19:03:42 Yeah, much better of course Apr 07 19:03:53 But attack/decay stuff is the same Apr 07 19:04:11 and lets you setup all matter of routings to FM the FM etc. Apr 07 19:04:30 anyways, have got a plane to catch (not that it is going anywhere, but still) Apr 07 19:05:01 npm: Yea. You normall do not get to set the sample in those low-end YM chips, but you do get to connect and configure the generators Apr 07 19:05:01 http://nielsmayer.com/npm/Intelligent-Summer-on-analog-modded-dynex+12db.ogg is a good showoff of the sounds Apr 07 19:05:40 Some (2608?) do allow some limited waveform generstion, usually used for percussion and sound effects (see Sega Genesis) Apr 07 19:06:36 anyways, me go airplane now bye bye Apr 07 19:06:48 the db60xg lets you DSP the sample Apr 07 19:06:50 RST38h: Have a great trip! Apr 07 19:06:59 i have yet to figure out how to hook up the input tho Apr 07 19:07:19 RST38h: bon voyage Apr 07 19:10:52 npm: Sounds liike fun :-) Apr 07 19:13:27 it's a low quality input. what i wanted to do is drive the i2c directly and likewise pull the i2c digital feed out of their DSP and avoid extra A/D/D/A's Apr 07 19:15:10 basically it's the same chips as in http://www.sonicstate.com/synth/yamaha_mu10/ Apr 07 19:18:19 I think 5 string simutaneous tuning is a bit beyond my capability to be honest ;) Apr 07 19:18:26 And the N900s :D Apr 07 19:19:27 I just want a reliable tuner implementation. The chord dict will be where most of the work goes. Apr 07 19:19:42 I want it to all look really pretty, be responsive etc. Apr 07 19:19:57 And maybe work on turning it into a mini guitar oriented studio. Apr 07 19:20:19 But for the sake of the coding competition, it'll just be a chord dict and a tuner. Apr 07 19:21:05 My project ideas has loads of stuff to add in the future, like turning the N900 into a mini effects and amp for electric guitars Apr 07 20:00:28 npm: Just install the pinetrail kernel and don't worry about it. It'll "do the right thing" and your old kernel won't be overwritten (unless the package names are the same) Apr 07 20:10:39 gabrbedd: thanks for the reassurance. Apr 07 20:11:19 so does that '38 kernel include snd_hrtimer ?? Apr 07 20:12:39 and will i be able to choose the kernel i run at boot Apr 07 20:17:01 npm: Dunno... I usually just compile my own. Yes, you can choose kernel at boot time on an x86 machine. (Hit [TAB] when you get to the extlinux boot loader screen) Apr 07 20:17:11 What kernel are you currently running? Apr 07 21:46:00 TSCHAKeee, evening, I don't have A9 an my Jogger is at home. Are you still getting the flashing? Apr 07 21:46:22 yes Apr 07 21:48:01 TSCHAKeee, has there been any EMGD updates posted on meego-commits? we could try raising a bug report and see where that goes Apr 07 21:48:14 didn't see any Apr 07 21:48:17 okay. Apr 07 23:19:06 * CosmoHill goes to bed **** ENDING LOGGING AT Fri Apr 08 02:59:58 2011