On 01/05/2011 15:28, Fons Adriaensen wrote: ..
I'd like to take with some grains of salt. I wonder, how would Harpex deal with a choir covering almost a semicircle and signing unisono (e.g. Gregorian). Converting this to the sum of two plane waves just doesn't make sense. This is actually something that works well in 1st order, but still works a lot better in 3rd (tried it using 5 panned spot mics).
That's a particuarly interesting example, especially on account of the unison content, and I would be very interested to try such a test myself if the opportunity ever presented itself. Unless the voices are deliberately separated physically, one would generally not want to casually notice and locate an individual voice - it would imply the voices were not vocally well matched - or quite simply you are too close to them. A general sense of the width and depth of the group, sure; but "needle-sharp" localization? I would find that (if obvious) distracting, a bit clinical.
Similarly in an orchestral string section. Their job is to blend - to create one wide and deep mega-"voice". if the back desks can be picked out individually in any obvious way, it means they are not properly playing together (we would call such players "passengers").
So I will ask my question in a slightly different way - is the notion of "too-precise" localization absurd? How much is too much?
Richard Dobson _______________________________________________ Sursound mailing list [email protected] https://mail.music.vt.edu/mailman/listinfo/sursound
