Sender:	nate AT cartsys DOT com
Message-ID:	<37603171.97E09047@cartsys.com>
Date:	Thu, 10 Jun 1999 14:43:13 -0700
From:	Nate Eldredge <nate AT cartsys DOT com>
X-Mailer:	Mozilla 4.08 [en] (X11; I; Linux 2.2.10 i586)
MIME-Version:	1.0
To:	djgpp-workers AT delorie DOT com
Subject:	Re: libm sources from cyberoptics
References:	<Pine DOT SUN DOT 3 DOT 91 DOT 990609131510 DOT 19031C-100000 AT is> <37600018 DOT F31667EB AT cyberoptics DOT com>
Reply-To:	djgpp-workers AT delorie DOT com

Eric Rudd wrote:
> 
> Eli Zaretskii wrote:
> 
> > On Wed, 9 Jun 1999, Martin Stromberg wrote:
> >
> > > _If_ lim(x^0) = 1,  then I would say 0^0 should be defined as 1,
> > >      x->0-
> >
> > Perhaps it would be a good idea to see what other libraries do in that
> > case.
> 
> Borland C++ v3.1: pow(0,0) = 1, domain error
> Watcom C++ v10.0: pow(0,0) = 1, domain error, "argument too large" [?]
> Microsoft C++: pow(0,0) = 1, no error return
> Linux gcc: pow(0,0) = 1, no error return
> hp 15C calculator: Error 0
> Matlab: 0^0 = 1, no error return
> Formal mathematics: limit is undefined
> 
> -Eric Rudd

IRIX 5.3 (cc): pow(0,0) = 1.000000, errno = 0 (Error 0) [i.e. no error]
SunOS 5.6 (gcc): pow(0,0) = 1.000000, errno = 0 (Error 0)
HP 48G calculator: 1
Sharp EL-531L calculator: Error 2
HP 32S calculator: "Invalid y^x"

IMHO, we should define 0^0 = 1 since that's what the ANSI amendment or
whatever it was says.  It seems likely to be approved.  Furthermore, I
don't feel that ANSI is totally off the wall on this (unlike, perhaps,
their views on trigraphs).  There certainly seems to be some
mathematical justification and support for 0^0=1, as for instance in the
FAQ I mentioned.
-- 

Nate Eldredge
nate AT cartsys DOT com

- Raw text -