Most current languages don't 'handle' floating point numbers themselves, they simply use the CPU instructions available to work with them.

The only way around that is to use a library that supports higher-precision floating point through emulation. Then agian, that still doesn't change the theoretical background in any way.