Abstract
A computational algorithm is described which quickly rounds large fractions into a fixed-length multiprecision floating-slash representation, using single-precision operations where possible. An easily calculated criterion for terminating the rounding process is given.

This publication has 3 references indexed in Scilit: