Re: Forward & Center Derivative Approximation Breakdown

[c.]

On 4 Feb 2014, at 08:18, Javier <address@hidden> wrote:

> Hello all,
> 
> Can anyone explain why my approximations break down at the iteration 15?  Is 
> this a memory issue? How does it really work? 
> Thanks.

What do you mean by "break down"?

If you are surprised that your difference 
formulas returns 0 for h = 1e-17 you shoud 
read this:

http://docs.oracle.com/cd/E19957-01/806-3568/ncg_goldberg.html

or simply try any of the following:

(1 + 1e-17) == 1
(1 + eps)   == 1
(1 + eps/2) == 1

HTH,
c.