Re: Continued Fractions

Paul: I've experimented with up 5 intervals using Viggo Brun's,
Selmer's, and Nils Pipping's multidimension Euclidean Algorithm methods
and have found the Brun's seems to be the most accurate. Selmer's
oscillates a lot and becomes very unstable with more factors. See the
Mills Biblio for references or email me if you don't have them.

I've not tried either the Jacobi or J. Murray Barbour's "mixed
expansion." David Feldman, btw, is much more of an expert in this area
than I am.

--John