On Wed, Feb 29, 2012 at 07:51:28PM -0800, Aaron Heller wrote: > Good ideas. In fact, for the 3rd-order solution for the CCRMA array > (352 parameters) I had to switch to a local optimizer, PRAXIS, with > initial values from the pseudoinverse modified by the rE-max per order > gains to get it to converge. This is what one would do for a regular > array. 1st-order (88 parameters) was fine with a global optimizer > (CRS).
This isn't a particularly difficult one. The ambdec config I supplied for it was found just using standard methods and a little bit of manual tweaking. The condition number for the pseudo-inverse is just 1.45, which is quite 'relaxed'. Main errors are direction differences between rV and rE, but only in elevation, and even these are limited to 5 degrees or so. But bonlinear optimisation would surely be able to improve this. 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
