The reasons a class of codes for solving ordinary differential equations might want to use an extremely small step size are investigated. For this class the likelihood of precision difficulties is ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results
Feedback