We discuss some linear acceleration methods for alternating series which ar
e in theory and in practice much better than that of Euler-Van Wijngaarden.
One of the algorithms, for instance, allows one to calculate Sigma(-1)(k)a
(k) with an error of about 17.93(-n) from the first n terms for a wide clas
s of sequences {a(k)}. Such methods are useful for high precision calculati
ons frequently appearing in number theory.