Abstract
Extrapolation methods for the solution of the initial-value problem for a system of ordinary differential equations (ODEs) advance the integration one step by combining the results of a number of subintegrations carried out with a simple method and fixed step size. The choice of step size in the subintegrations influences the efficiency of the resulting formula. In a first investigation it was shown how to quantify this influence. Here this influence is studied when the code is attempting to select the optimal order at each step. A novel result about this optimal order allows the two most important experimental developments to be fully explained.

This publication has 0 references indexed in Scilit: