ABSTRACT
A computer can store only a finite amount of information, i.e., substantially less compared to what is mathematically possible. We must therefore specify the domains of computed numbers and how we have to calculate using these numbers. The following view supports fixed-point numbers for describing computer internal models. Since we want to achieve a well-defined fixed-point precision, it is often necessary to use a large number of digits.
Let us consider domains of numbers describing the representation of geometric objects and their calculation. Representation numbers often differ from numbers used for computing the representation. At first we have to start from the desired representation domains because they are the base of the computation.
