Menu
▾
▴
Re: [Libmls0-developers] RE: libmls-0.3 released
|
From: Daniel S. <ds...@ho...> - 2004-06-23 15:09:07
|
Hi again! As always, one more thing. Daniel: it would be really nice if you could=20 provide a very short summary of what the different programs in you=20 lsputils actually does. "FRD to ZMA transformer" is not very useful information if you do not=20 know what frd and zma is. Of course I might look at the source, which I=20 will do anyway, and find out for myself, but a very short and quick=20 description will get me started a little quicker...:-) /Daniel Daniel Sundberg wrote: >> >>> I have two questions: what is FHT? Fast Hamahard Transform? >> >> >> >> Fast Hadamard Transform, I think. >> >> >>> Daniel: did libmls use FHT before and now uses FFT? >> >> >> >> Yes, I basically replaced the FHT transform routines from Anders=20 >> Torger with FFTW. >> I'm not sure if it would be more correct to call it DFT or FFT though,= =20 >> since it's not N^2 length. >> But FFTW must use some kind of FFT method anyway since it's so fast. >> >=20 > Well, I just realized I have studied this in a signal processing=20 > course... I actually have access to very detailed theory on this=20 > subject...:-) >=20 >> >> The above also means that the information in the README file is=20 >=20 > > longer correct. There is no mic calibration mode anymore, and > > there must be a loop from the output to the left (reference > > channel) and the right channel must be used for measurements. > > >=20 > If this is the case we really need some updated documentation, could yo= u=20 > please patch the README and we will make an updated release. >=20 >> And then to my attachment: >> I have attached the loudspeaker library that I'm working on. >> I plan to put it on savannah.gnu.org or sourceforge soon, but I would=20 >> like to have some comments on the code before I do that. >> >=20 > I will try to take a look at this and of course comment if I feel the=20 > need to... >=20 >> My goal with this library, is that it, together with libmls, should be >> complete enough to replace CALSOD, MLSSA etc. with open source tools. >> I also want to keep it as portable as possible. >> >=20 > It's all about the GUI today, you know...:-) >=20 > Regards >=20 > Daniel Sundberg >=20 >=20 >> The intended audience is: >> Developers of measurement or loudspeaker software like e.g. gspeakers. >> Advanced users that would like complete freedom in choosing design=20 >> methodology. >> >> So it is both a library, to satisfy the first group. >> And a set of command line utilities using the above library, to=20 >> satisfy the second group. >> >> To make it both as portable and as easily scriptable/customizable as=20 >> possible the command line utilities should only do one operation each=20 >> and the only thing requiring potentially hard to port or compile GUI=20 >> code should be the waveform viewer. >> >> Implemented at the moment is: >> *FRD/ZMA file I/O >> *Import of ddb speaker data base. >> As generated and used by: >> http://www.danbbs.dk/~ko/ubmodel.htm >> *Partial implementation of the loudspeaker model in: >> http://www.hal-pc.org/~bwhitejr/btsmodel.zip >> found here: >> http://www.geocities.com/loudspeakerguru/Speaker_Design_Equations.ht= ml >> *Standard filter generator >> *Ladder filter simulator with file I/O >> *FRD waveform viewer >> >> The waveform viewer can use plotutils or gnuplot and it can display=20 >> the results from mra/sra. >> It doesn't require plotutils to compile and it should be very easy to=20 >> port to other platforms or to embed in other applications. >> >> So most of the boring infrastructure things are at least usable and=20 >> lots of fun remains: >> Filter/box optimization, spice like simulator and other box types (tl,= =20 >> horn, etc.). >> >> When mra/sra has been updated with a bit more options (getopt) it=20 >> would be very easy to create a graphical front end that would just=20 >> launch mra and pass the results to resgraph (the waveform viewer). >> >> There are some example scripts in the script directory, >> both for using plotutils and gnuplot. >> >> Regards, >> Daniel W >> >> >>> From: Daniel Sundberg <ds...@ho...> >>> Date: 2004/06/22 ti AM 11:11:57 GMT+02:00 >>> To: lib...@li... >>> =C4mne: Re: [Libmls0-developers] RE: libmls-0.3 released >>> >>> Hi! >>> >>> I have two questions: what is FHT? Fast Hamahard Transform? >>> >>> Daniel: did libmls use FHT before and now uses FFT? I am afraid I=20 >>> have not been looking very close at the source code lately (mostly=20 >>> since I do not own a decent measurement microphone). >>> >>> Also, when some of you start to work on a sample application it might= =20 >>> be a good idea to create a separate module in the cvs or create a new= =20 >>> project. >>> >>> Regards >>> >>> Daniel Sundberg >>> >>> >>> dan...@br... wrote: >>> >>>> Hi Scott >>>> >>>> >>>> >>>>> A little while ago Daniel talked about replacing the FHT with=20 >>>>> another scheme to generate the impulse response, which appears to=20 >>>>> have been done. Can you confirm that it is as fast as the FHT? >>>> >>>> >>>> >>>> I don't know if it's as fast as FHT, it's certainly fast enough for=20 >>>> desktop computers. I have no idea how they do it, but fftw doesn't=20 >>>> seem to be that much slower for n^2-1 fft lengths than n^2 lengths. >>>> >>>> >>>> >>>>> Does the correlation still create a perfect unit response as it=20 >>>>> should with the MLS? >>>> >>>> >>>> >>>> I'm not sure what you mean with a perfect unit response. But I'm=20 >>>> sure the new scheme is much more accurate than the FHT. I have=20 >>>> compared the results from the sweep mode with the MLS mode and they=20 >>>> are the same. >>>> >>>> The problem with FHT MLS is that you assume that the soundcard=20 >>>> doesn't degrade the signal at all. It's in fact even more sensitive=20 >>>> to this than you might think. Even with an almost perfectly flat=20 >>>> freqency response you will still not get good results with the FHT.=20 >>>> I think the FHT should be viewed as a last resort, only to be used=20 >>>> when speed is top priority and not accuracy. >>>> Although some of the FHT problems could be solved with calibration I= =20 >>>> still don't think you will get as good results as when you correlate= =20 >>>> in the frequency domain. I'm not sure that the FHT errors are static= =20 >>>> enough to be effectivly removed with calibration. >>>> >>>> >>>> >>>>> I must get started on a test/demo app. >>>> >>>> >>>> >>>> I have almost finished a waveform viewer/library that can be used=20 >>>> from other appplications, so you might want to wait for that. >>>> I will make it public as soon as I figure out where to put it. >>>> >>>> >>>> >>>>> One final thing. If we have upgraded the project name to level 1=20 >>>>> then I think we need to upgrade the major revision number to 1 as=20 >>>>> well. Perhaps 1.0. >>>> >>>> >>>> >>>> I agree on this, although I don't think we should go 1.0 until the=20 >>>> library is stable. >>>> >>>> /Daniel >>>> >>>> >>>> >>>> >>>> ------------------------------------------------------- >>>> This SF.Net email is sponsored by The 2004 JavaOne(SM) Conference >>>> Learn from the experts at JavaOne(SM), Sun's Worldwide Java Develope= r >>>> Conference, June 28 - July 1 at the Moscone Center in San Francisco,= CA >>>> REGISTER AND SAVE! http://java.sun.com/javaone/sf Priority Code=20 >>>> NWMGYKND >>>> _______________________________________________ >>>> Libmls0-developers mailing list >>>> Lib...@li... >>>> https://lists.sourceforge.net/lists/listinfo/libmls0-developers >>>> >>>> >>> >>> >>> >>> ------------------------------------------------------- >>> This SF.Net email sponsored by Black Hat Briefings & Training. >>> Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital= =20 >>> self defense, top technical experts, no vendor pitches, unmatched=20 >>> networking opportunities. Visit www.blackhat.com >>> _______________________________________________ >>> Libmls0-developers mailing list >>> Lib...@li... >>> https://lists.sourceforge.net/lists/listinfo/libmls0-developers >>> >=20 >=20 >=20 > ------------------------------------------------------- > This SF.Net email sponsored by Black Hat Briefings & Training. > Attend Black Hat Briefings & Training, Las Vegas July 24-29 - digital=20 > self defense, top technical experts, no vendor pitches, unmatched=20 > networking opportunities. Visit www.blackhat.com > _______________________________________________ > Libmls0-developers mailing list > Lib...@li... > https://lists.sourceforge.net/lists/listinfo/libmls0-developers >=20 >=20 |
