On Wed, Feb 29, 2012 at 07:16:18PM +0000, Paul Power wrote: > I am a currently looking into Ambisonic periphonic systems. > This is my first post so please go easy on me. I am currently using a 16 > speaker rig which includes a cube and an octagon set up. I understand that > this > is not a regular shape so is not optimal. > > The question i have at the moment is that i am trying to > implement a 2nd order encoder/decoder in Matlab, however at the > moment I am running into problems. > > I am using the method described in Hollerwegers paper “Periphonic > Sound Spatialization in Multiuser Virtual Environments” using the SN3D > equations, utilising the pseudo inverse method to derive my decoding matrix. I > currently have a 1st and 3rd order encoder/decoder giving > sensible answers with respect to Gerzons diametric decoder theorem. The > problem > with the 2nd order decoder is that the final gains given do not give > negative phase for the diametrically opposed speaker. I have also tried using > a > regular shaped speaker set up, but this still does not give me negative phase > in the diametrically opposed speaker.
Your setup does not support periphonic 3rd order, so the results you get for that should be taken with some big grains of salt. Regarding 2nd order, the driving functions for a regular layout would be 2nd order hypercardioids which have a positive back lobe. There is *no* reason to expect that the opposing speaker should be in antiphase. You are applying first-order theory to higher order, and that doesn't work. Ciao, -- FA Vor uns liegt ein weites Tal, die Sonne scheint - ein Glitzerstrahl. _______________________________________________ Sursound mailing list Sursound@music.vt.edu https://mail.music.vt.edu/mailman/listinfo/sursound
