[seqfan] Re: Implementing first 500 sequences in Java

Sean A. Irvine sairvin at xtra.co.nz
Fri Oct 29 21:04:31 CEST 2010


Java does has the BigInteger class, but I actually use my own
implementation for big numbers because BigInteger is not
particularly fast and is a little feature limited.  My big
number arithmetic is based mostly on Lenstra's LIP package.

My implementation has been extensively tested back against
BigInteger. Also, every sequence matches exactly the terms
in the OEIS.  Where it was doubtful I also verified against
the b-files where they exist. I'm sure there will be some
errors, but hopefully not too many.

For a few sequences, I have made use of Mikko Tommila's
Apfloat package for arbitrary precision floating-point.
(e.g. for computing iterated tangent and other trig
functions).

Sean.

On Fri, Oct 29, 2010 at 4:51 PM, N. J. A. Sloane <njas at research.att.com> 
wrote:
 > Does Java do exact arithmetic on large numbers?
 > Can the results of these programs be trusted?
 > This is not a criticism of your work, I am just asking!
 > Neil
 >
 >
 > _______________________________________________
 >
 > Seqfan Mailing list - http://list.seqfan.eu/
 >




More information about the SeqFan mailing list