Path: utzoo!attcan!uunet!mcvax!nikhefh!t68
From: t68@nikhefh.hep.nl (Jos Vermaseren)
Newsgroups: comp.lang.fortran
Subject: Fortran vs C for computations
Message-ID: <535@nikhefh.hep.nl>
Date: 5 Sep 88 08:37:27 GMT
Organization: Nikhef-H, Amsterdam (the Netherlands).
Lines: 23



There is another much more annoying feature of C that restricts its floating
point handling to the 'playing only' level.
In the second edition of Kernighan and Ritchie it is explained that the
standard response of a floating point operation (multiplication, sqrt etc)
to an irregularity is returning special values.

So
	float x = -2.0;
	z = sqrt(x);

gives back some value. Amazingly enough many UNIX vendors like this philosophy,
so even the UNIX fortran compilers often prefer it this way.
If I have to put all kinds of ifs around each floating point operation I'll
never do computations in C.
The better thing would be to have a parameter that can be set by the user
that determines whether an overflow or sqrt of a negative number should be
fatal (with intelligent errormeassage please). Most fortran compilers have
this option.

J.Vermaseren
T68@nikhefh.hep.nl