On Convergence Rates of Root-Seeking Methods

[emece67]

(03-18-2018 08:34 PM)rprosperi Wrote:  

(03-18-2018 07:18 PM)emece67 Wrote:  Oops, it was my fault. There were no billion digits, but 1 billion bits, this is ~300 million decimal digits. I screwed it up when switching to gmpy.

In any case, the program does not store them on disk. After each run it only reports the number of iterations needed, the time spent and the number of correct digits in the result. When running at the maximum precision of ~300 million digits it takes ~2.5 GB of RAM.

No problem, that is still a LOT of digits.

However, if the result is not stored, how do you know the results are correct?

Comparison against the sqrt function inside the mpmath library and also by squaring.