[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Implementor support for the binary interchange formats

Since binary128 supports 34-bit integers EXACTLY, there should be zero
precision loss for traditional COBOL usage of scaled decimal (e.g. by
handling currency in micropennies).

This applies even when legal requirements specify a given decimal precision,
e.g. IRS requirements to round fractions to 4 digits after the decimal point,
or European Union currency conversion rules.  Surely COBOL has been handling
this well for years...

This reminds me of a question I asked earlier:  would COBOL exploit the fact
that DFP permits decimal fixed-point arithmetic to be emulated without the
need to remember the scale separately?  My guess was NO, given the statement
that COBOL would assume all floating-point (including decimal) to be normal,
and the fact that this would inhibit easy switching between arithmetics.

But, as William Klein pointed out:
1) Arithmetic mode PRIMARILY deals with "intermediate results".

So operations like  (7 / 4 ) * 4  could be affected.

---Sent: 2011-02-28 18:35:14 UTC

754 | revision | FAQ | references | list archive