index precision



Pol wrote:
> Why do steps at 1.8 and 1.9 are displayed with 16 digits, unlike the other
> steps?

Because 0.1 cannot be represented exactly as a floating-point number, 
and repeatedly adding 0.1 will introduce (more) roundoff errors.  This 
is also why maxima prints out i = - 1.3877787807814457E-16 instead of 0.
Extra digits are needed because they are needed to represent the number.

Maxima has rational numbers. Why not use them?  Or count from -1 to 20 
and divide i by 10.0.   The results will be more accurate, but will 
still have roundoff issues.

Ray