Grappling with ∞: High Precision Math in Perl 6

By Cal
Date: Friday, 25 August 2017 15:45
Duration: 20 minutes
Target audience: Intermediate
Language: English
Tags: algorithms datatypes perl6 rakudo

Arbitrary precision data-types fail spectacularly when asked to store transcendental numbers (such as π or 𝑒). Continued fractions help (especially if they are generalized continued fractions), but computers are not very good at working with infinity. Then, once an internal representation is settled upon, there is the issue of computing trigonometric and exponential functions to high levels of precision. This talk is about the trade-offs made on the path to having a workable implementation of high precision math in Perl 6.