Prev: RPN/RPL Calculator implementations, list of, regular post [long, FAQ]
Next: Graphics programming commands?
From: Jacob Wall on 8 May 2010 12:11
On 07/05/2010 8:34 PM, PremiumBlend wrote:
> Your "shift" code works very well. In an example from
> the Wolf and Brinker text, the authors shift the axes
> so that the most southerly and westerly stations are
> equal to zero. The way the author presented it, I was
> lead to believe that insufficient precision and serious
> round-off errors occur unless double precision is used.

IMO any round-off error should be taken seriously, so had the authors
said something like "insignificant round-off error may occur", then most
of us wouldn't give it a second thought.

Also, with the very large area used previously, the error seems
insignificant as a percentage, however if one was calculating a complex
shape with a small area of say <50 m� and you had even 0.5 m�
discrepancy due to round-off, all of a sudden it becomes a little more
of a concern.

Jacob
First  |  Prev  | 
Pages: 1 2
Prev: RPN/RPL Calculator implementations, list of, regular post [long, FAQ]
Next: Graphics programming commands?