This is the mail archive of the gcc@gcc.gnu.org mailing list for the GCC project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]

Re: Change definition of complex::norm


nbecker@fred.net writes:

 > We talked about this before, but it appears that in 3.0.2 this still
 > isn't changed.
 > 
 > There are clear disadvantages to the current definition of
 > complex::norm in terms of abs, not the least of which is that it won't work for
 > complex<int>, for example.   I can't imagine any drawback to defining
 > norm as
 > 
 > real(z) * real(z) + imag(z) * imag(z).

What's the current definition?

-- 
Harvey Stein
Bloomberg LP
hjstein@bloomberg.com


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]