I agree that using floating point is an issue.  Even using BCD or
packed decimal (2 decimal digits per byte) has problems.  But I think
that BCD or packed decimal is the way to go for accounting.  Float is
great for engineering or scientific calculations, but rounding issues
caused me to want to use either BCD, packed decimal, or integers only
for accounting.  At that, still need to check how math is done in your
libraries if you are writing your own package (yes, I have been bit
more often that I like to admit).

... Jack
_______________________________________________
Discuss mailing list
Discuss@blu.org
http://lists.blu.org/mailman/listinfo/discuss

Reply via email to