double x0 = 0x1.0000000000001p0;
double y = 0x1p-53 * (1. - 0x1p-11);
double z1 = x0 + y;<p>or other examples taken from <a href="http://hal.archives-ouvertes.fr/docs/00/28/14/29/PDF/floating-point-article.pdf" rel="nofollow">http://hal.archives-ouvertes.fr/docs/00/28/14/29/PDF/floatin...</a> would have been more interesting than 0.1 + 0.2, for which every tested language seem to compute the proper IEEE 754 double precision result.