RH>Quoting (part of) a msg from you of 08 Nov 1994:

RH> RB> I need to do a distance calculation between two points
RH> RB> (x0,z0) and (x1,z1).

RH>If you are going to use the mentioned Newton Raphson method (and why not),
RH>then it helps if the first appromimation is as close as
RH>possible with a simple calculation.

Thanks, using the dimensions of the triangle to calculate limits is
an observation I should use.

RH>And as reaction to the arguments about floating-point support being
RH>'standard' in OS/2, I would say: If the co-ordinates and
RH>the result have to be integer-values, you need 3 type
RH>conversions for every calculation. This might be more
RH>resource consuming than the integer approximation itself!
RH>You would have to write a small 'benchmark'-program to verify this.

I'm actually using fixed point mathematics with 16 bits for the
fractional part.  This is MUCH faster than floating point even with a
coprocessor, especially for addition and subtraction, and gives me the
precision I require.  No type conversion required.

Rob.
___
 X SLMR 2.1a X This tagline is umop apisdn

--- Maximus/2 2.01wb