Re: Floating Point Issues

In article <00dd5f63-496a-4cd9-a040-af5ddc6b62e1@xxxxxxxxxxxxxxxxxxxxxxxxxxx>,
user923005 <dcorbit@xxxxxxxxx> wrote:
On Jan 2, 7:55 am, Bill Woessner <woess...@xxxxxxxxx> wrote:
I need to evaluate an expression of the form AB-CD and I only need the
fractional portion of the result.  Unfortunately, A and B are of the
order 10^6, C is of the order 10^9 and D is of the order 10^-1.  So
when I go to form my sum, which is of the order 10^12, I only get a
few digits of precision in the fractional portion

If you can use C++, then I think that double-double will answer very
nicely:
http://www.cs.berkeley.edu/~yozo/

Some Fortran compilers offer quadruple precision.

-- John Harper, School of Mathematics, Statistics and Computer Science,
Victoria University, PO Box 600, Wellington 6140, New Zealand
e-mail john.harper@xxxxxxxxx phone (+64)(4)463 5662 fax (+64)(4)463 5045
.