NetNews Usenet Archive 1992 #27/NN_1992_27.iso/spool/comp/lang/fortran/4436

Text File | 1992-11-24 | 1.4 KB | 32 lines

Newsgroups: comp.lang.fortran
Path: sparky!uunet!cs.utexas.edu!swrinde!sdd.hp.com!saimiri.primate.wisc.edu!caen!hellgate.utah.edu!lanl!cochiti.lanl.gov!jlg
From: [e-mail] (J. Giles)
Subject: Re: a question on ATAN function
Message-ID: <[e-mail]>
Sender: [e-mail]
Organization: Los Alamos National Laboratory
References:  <[e-mail]>
Date: Tue, 24 Nov 1992 00:35:07 GMT
Lines: 20

In article <[e-mail]>, [e-mail] (Darran Edmundson) writes:
|> I'd like to pose a neophyte question on using the arctan function to
|> calculate the phase angle of a complex variable (i.e. \theta = atan(v/u)
|> where u and v are the real and complex parts of my imaginary number
|> respectively).  Is it sufficient to test for u=0 (i.e. \theta = \pm \pi)
|> to avoid problems?  What happens if the ratio exceeds range allowed
|> for doubles or reals?  How do you avoid this problem?
|>
|> (Now that I think about it, is it better to use arcsin or arccos?)

What you should use is ATAN2(AIMAG(Z), REAL(Z)).  You needn't even
test for the real part being zero.  If the second argument is zero
and the first argument is not, the answer will have absolute value
equal to PI/2 (the sign depending on the sign of the imaginary
part of Z).  In fact, this function should work for any pair of
floating point numbers provided they aren't *both* zero and neither
is out of range.

--
J. Giles