A simple math library for template metaprogramming in c++. The main purpose of this library was to create a compile-time floating point data type as an exercise in template metaprogramming.

However, just like the run-time type, operations are not guarantied to give exact values because of size limitations in c++ data-types as well as the nature of rational numbers.

• Floating point representation using rational numbers
• The basic four operators implemented (+, -, *, /)
• All comparison operators implemented (=, !=, <, <=, >, >=)
• sqrt() function implemented

• Better precision
  • Add seperate member for whole integers