I'm somewhat surprised that antibase gives up at 14 decimal digits of precision. Consider #:!. . Fit would help document the occurrence of an algorithmic choice.
9!:14'' j602/2008-03-03/16:45 10 ((] # [) #: o.@(^ i.)) 16 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3.14159 0 0 0 0 0 0 0 0 0 0 0 0 0 0 3 1.41593 0 0 0 0 0 0 0 0 0 0 0 0 0 3 1 4.15927 0 0 0 0 0 0 0 0 0 0 0 0 3 1 4 1.59265 0 0 0 0 0 0 0 0 0 0 0 3 1 4 1 5.92654 0 0 0 0 0 0 0 0 0 0 3 1 4 1 5 9.26536 0 0 0 0 0 0 0 0 0 3 1 4 1 5 9 2.65359 0 0 0 0 0 0 0 0 3 1 4 1 5 9 2 6.5359 0 0 0 0 0 0 0 3 1 4 1 5 9 2 6 5.35898 0 0 0 0 0 0 3 1 4 1 5 9 2 6 5 3.58979 0 0 0 0 0 3 1 4 1 5 9 2 6 5 3 5.89793 0 0 0 0 3 1 4 1 5 9 2 6 5 3 5 8.97931 0 0 0 3 1 4 1 5 9 2 6 5 3 5 8 9.79297 0 0 3 1 4 1 5 9 2 6 5 3 5 8 9 7.92969 0 3 1 4 1 5 9 2 6 5 3 5 8 9 7.92969 0 3 1 4 1 5 9 2 6 5 3 5 8 9 7.92969 0 0 ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
