Next: 3.1.4 Accurate Functions
Up: 3.1 Floating Point
Previous: 3.1.2 Constant Functions
Consider the n-ary function ,
with implementations of and provided.
Consider the m-ary function ,
along with n m-ary functions
which are exact over .
If
then
and
An implementation of invokes
with arguments given by exact implementations
of ; an implementation of similarly uses
.
Examples include:
The following examples demonstrate another method
for computing common constant functions:
This method is usually preferable to the high precision method
outlined in the previous subsection, as it is more portable
and easily implemented.
This method is, of course, restricted
to those functions and constants which may be directly constructed
from the provided operators.
Next: 3.1.4 Accurate Functions
Up: 3.1 Floating Point
Previous: 3.1.2 Constant Functions