Conclusion
June 9, 1996
55
6.0 Conclusion
XStar has certain characteristics that influence which methods should be used to
solve the N-body problem. They include:
*
The long "history" of locations that XStar generates
*
The real-time display requirements of XStar
*
The requirement of having high enough accuracy so that the worst of the
deviations do not occur, but there is no requirement that the accuracy must be
extremely high.
*
Smooth paths are the norm
*
For practical reasons, only a few to a few dozen stars can be display on the
screen without it becoming cluttered.
These characteristics lead to corresponding conclusions:
*
Multistep methods are very practical
*
Variable step size methods are much harder to implement
*
In practice, it appears that a method with an order of at least 4 is needed
*
Simple Adam-Bashford methods work well
*
Complex force function evaluation methods are not required
These conclusions combine to form the final conclusion that the 7
th
order Adam-
Bashford method is the best method for most of the practical accuracy range. For very low
accuracy, the taylor3 method is the best, and for the very high accuracy, the gpemce8 is the
best.
Previous page    Home    Table of Contents    Next page
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30
31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57
This document is best viewed as n-body.pdf because the translation to html was buggy.