Re: [cc65] Floating point support?

Sidney Cadot dnia 28 wrz 2004 o 18:56 +0200 napisal:
> But if you're calculating the average cost of N shopping items?

The next step might be comparing that average to something with == :).

> Yes it can. E.g. moving up from single to double precision (6 to 15 
> decimal digits!) gets many realistic problems out of a zone where 
> numerical roundoff introduces so much noise that all information is 
> lost.

I can't discuss with 'many realistic problems' argument :). Increasing
precision just delays appearance of the problem (sometimes delays enough),
it doesn't solve it. Change the scale and it will appear again.
Most problems with FP can be easily avoided once you realize that they
are almost never exact and remember about the involved error.

ytm

ps. How about putting brackets in various places to change the order of
    calculations?

Take a look at the last problem from
http://www.cs.mtu.edu/~shene/COURSES/cs3621/NOTES/overview/reals.html
On my computer (with doubles): gcc -O2 -lm assoc.c -o assoc; ./assoc 56 56
gives results that differ more than 0.025.

----------------------------------------------------------------------
To unsubscribe from the list send mail to majordomo@musoftware.de with
the string "unsubscribe cc65" in the body(!) of the mail.