Date: Wed, 6 Dec 1995 16:06:31 +0300 From: "Alexander V. Lukyanov" To: djgpp AT sun DOT soe DOT clarkson DOT edu Cc: gip AT firefox DOT ino DOT it Subject: Re: Possible Bug in V2 ( -O and libm) Remember that calculating with floating point has some limited precision. In that case, subtracting two large numbers can give a relatively small error, such as 1.0 --- Alexander Lukyanov lav AT video DOT yars DOT free DOT net