Thank you. Much appreciated. - martin
On Tue, 26 Mar 2019 at 22:36, Fons Adriaensen <f...@linuxaudio.org> wrote: > On Tue, Mar 26, 2019 at 09:59:45PM +0000, Martin Dupras wrote: > > > However, no one has answered my question. I will ask again. > > > > Can someone point me to any published resources that explain the > principles > > and theoretical equations required to in theory convert from 2nd order > > A-format to B-format? I can find resources for 1st order but not for 2nd > > order. > > First, there is no such thing as a standard 2nd order A-format. All > higher order AMB mics I know of use different capsule geometries. > Also some are on a solid sphere, some are open, and that changes > things as well. > > Secondly, for second and higher order things become much more > complex. One of the reasons for this is that normal capsules > don't have much 2nd or higher order components in their polar > pattern. Whatever there is are imperfections or the result of > diffraction on the mic body. That means you can only obtain > the higher order components by exploiting the finite distance > between the capsules and any diffraction, and things become > very frequency dependent as a result of this. > > Basically this a system inversion problem, basic matrix maths. > > The theoretical polar patterns of the mics, or IR measurements > in a number of directions, define a matrix. Inverting this matrix > and then multiplying the result with another one defining the > B-format polar patterns, you obtain the A/B conversion matrix. > > This sound simple, but in practice it isn't. The inversion will > be frequency dependent and ill-conditioned at lower frequencies. > The theoretically 'perfect' result won't be usable, it may require > extreme gains and amplify noise and small errors in the original > data out of proportion. > > You need to deal with this in a practical way. This is were > the 'secret sauce' is, it is a careful balancing act between > conflicting requirements. There are some mathematical tricks > to help with this but that would lead us very far. The simple > ones (presented in some papers) don't work well in practice. > > The basic theory you'll find in several papers on higher order > Ambisonic theory. Warning: this becomes very mathematical, and > there is no other way. > > Ciao, > > -- > FA > _______________________________________________ > Sursound mailing list > Sursound@music.vt.edu > https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe here, > edit account or options, view archives and so on. > -------------- next part -------------- An HTML attachment was scrubbed... URL: <https://mail.music.vt.edu/mailman/private/sursound/attachments/20190326/83075168/attachment.html> _______________________________________________ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound - unsubscribe here, edit account or options, view archives and so on.
