I have found many of the Fourier transforms of Mathcad 15 to be in error where there were not in error on versions 13 and 14

[JamesGilliam]

I have founrd that Mathcad 15 is rife with errors on finding the Fourier Transform of various functions. For example the Heavyside Step function is incorrect. Interesting enough, I don't find the same errors on version 13.

Jim Gilliam

[PhilipOakley]

I think not....

However they have changed though. The choice of i and -i for forward and backward transforms, and for the forward and inverse scaling factors is entirely arbitrary and is bound up in one's expectations and the local conventions. (i.e. it is an endless source of discussion)

In the 'More Like This' tab, next to your post, it linked to http://communities.ptc.com/message/56135#56135 which discussed the problems. Obviously you couldn't have seen that until after you had posted!

Philip

[mzeftel]

There are some differences in Fourier results between Mathcad 13, and Mathcad 14 and 15, because we used a new symbolic engine from Mathcad 14 on.

The ones that I am aware of are:

1. The fourier transforms of certain functions resembling 1/t produce the negative of the (imaginary) result that they produced in 13. It is a different form of the results. If you want the same results as in Mathcad 13, then the simple workaround is to take the fourier transform of f(-t). This returns the same result as the Mathcad 13 fourier transform of f(t). It is a matter of whether the counter-clockwise (or clock-wise) rotation in the complex plane is assumed to be positive or negative.

2. There are similar issues with the Fourier transform of the Heaviside function.

If you have other results that you feel have issues, please post a worksheet here. \

Mona