We'll the third soundmosaic test finally finished after computing for 3.5+ days.

http://noneinc.com/SWxTPM-4DWHUXGA/soundmosaic/sw4d-test-003.wav

Some of the tones are a little stronger then the --partition quick test-001 version. The problem is the sample base, it's too limited. Next round of tests will be with a larger piece of the movie, working toward the full movie audio being digested.

One of the things learned in this third test is that it's probably best to not continue working at 4800khz. Half of quarter resolution may work enough to show the theory can be pulled off.