Re: Strange result with exponents...

[Hans Boldt <boldt@xxxxxxxxxx>]

jt wrote:
> ...
> | As a rough guideline, I'd simply recommend avoiding float numeric values
> | unless you know what you're doing. Perhaps a B (or better) grade
> | in senior
> | high school physics would be a good pre-req. ;-)
>
> Hans, I got an A in HS physics and read graduate level textbooks to do my
> final paper.  I'm NOT stupid, but I'm guessing you know me well enough to
> know I do (and say) some of THE most outrageously STUPID things, anyway.
> (If I knew why I did some stupid things, I wouldn't...;-)  ANYhoo...

For me, I got a 90% in my final HS physics class, good enough for best 
in the class. For some reason, I felt it necessary to take the tough 
physics course in 1st year university, and I consider myself very lucky 
to have passed that course with a 57%! Some of my friends in that class 
weren't so fortunate.

My point is just that float format was primarily intended for scientific 
calculations, and an understanding of the concept of precision is important.

Here's another practical example. Let's say you have two 16 digit float 
values that differ only in the 16th digit. What is the precision of the 
result of subtracting the two values? 16 digits? No. The following 
Python command line session illustrates:

>>> a=1.111111111111111
>>> b=1.111111111111112
>>> b-a
1.1102230246251565e-015

Although IEEE format represents a value with 17 digits, only the first 
is significant.

OK sure, that's an extreme example, but I think it's important to 
understand the issues involved. And as RPG acquires more and more 
enhancements, it becomes more and more important to understand these 
fundamental issues.

> ...

Anyways, I just recorded a suggestion in our "Suggestions for Future 
Enhancements" file for a decimal exponentiation operator. It certainly 
wouldn't be as fast as the current ** operator, but it would provide 
more digits. (Speed or accuracy - make your choice!)

Cheers! Hans