Comparison to Boost.Multiprecision

An easy question to ask is why this is a separate library from Boost.Multiprecision. There are several reasons why:

  • The goal is for this library to be extremely lightweight. The module weight of this library will be 5, whereas Boost.Multiprecision is 25

  • Fundamentally, the designs of the types are different. In Boost.Multiprecision, int128 and uint128 are incident to the arbitrary precision integer cpp_int, not a specifically designed type. In this library both int128_t and uint128_t are their own classes with individually implemented operators to maximize performance.

  • In Boost.Multiprecision all types are based on a high-level template number to allow interoperability with each other. Again, both int128_t and uint128_t are their own classes, and are designed to work with and act like the built-in integer types.

Edit this Page