how to compute high precision decimals > 38

Question

 

 

 

I want to be able to compute very high precision decimals (> 38 digits).

I've tried looking into Speed Crunch source code, in c++, but it is too complicated.