On Sun, Jun 02, 2013 at 11:14:24PM -0400, David Pickett wrote: > Perhaps I was not clear. Take 2: > > It seems to me that Sampo claimed that the omni from a double > capsule mic is not the same as the W signal obtained by processing > the ouputs of the four mics in the A-format tetrahedral array, and I > would agree. However, the former can at least produce an omni in > one (i.e. the horizontal) plane -- which i what I want. So the > question really is whether X and Y derived from the tetrahedral > array are different from those that would be obtained with two > horizontal figure of eight mics. (I dont think that they are, but I > am asking for confirmation or denial.)
Neither the W, X, Y from a tetra mic nor those obtained directly from omni or fig-8 capsules will be perfect, but apart from the imperfections which in both cases will show up mainly at high frequencies and which will be different, they are absolutely identical. There is no fundamental difference. A horizontal decoder does not use Z. Since W has the same gain in all vertical directions and X and Y don't, the result is that sources with non-zero elevation will be reproduced with reduced rV and rE, both of them becoming zero at the zenith and nadir. Note that a W' that is omni in the horizontal plane and that exactly matches the vertical pattern of X and Y would need to be of 'infinite' order, since it would have a discontinuity in the first derivative at the zenith and nadir. As a first approximation to such a W' you could combine W with the second order 'R' component. Any additional components will have to be from even orders, so you need at least 4th order to improve this. A decoder using such W', X, Y would attenuate sources with non-zero elevation instead of reproducing them with a reduced velocity vector. It's not at all clear if that would actually be an improvement. Ciao, -- FA A world of exhaustive, reliable metadata would be an utopia. It's also a pipe-dream, founded on self-delusion, nerd hubris and hysterically inflated market opportunities. (Cory Doctorow) _______________________________________________ Sursound mailing list [email protected] https://mail.music.vt.edu/mailman/listinfo/sursound
