Hello Neil!

Replying to a message of Neil Heller to Bill Birrell:

 BB>> That was why j notation had to be invented.

 NH> j notation?  What's that?

Imaginary numbers - although mathemeticians usually use i rather than j, in
my experience.  I used both in university, depending on whether I was in a
course offered from the Math faculty, or the Engineering faculty...

 NH> As originally stated, the code you posted went something like:

 NH> for (int n=0; n < MAXINT; n++)
 NH>      printf("d\n", n * n);

(minor typo - forgetting the % ;->)

 NH> When n > sqrt(MAXINT) the result of n*n would be negative.

And this is neither math nor engineering - it's computer science.  :-)

More accurately, the result of n*n would appear negative.  And that's not
entirely true, either.  The result of n*n _may_ _appear_ negative after the
high-bits were dropped.  But nothing says that they _will_ be negative - it
all depends on whatever high bit is left.  In all likelihood, some will
appear negative, some will not.  Either way, they're wrong once n >
sqrt(MAXINT), as you pointed out.

Darin

---