From: rogeeff (rogeeff_at_[hidden])
Date: 2001-12-19 01:43:44
> Gennadiy's floating_point_comparison looks to me what most testers
need.
> (Though I would express the tolerance in terms of number of
epsilons as
> some of operations might be expected to cancel
Do you mean that some operation can have exact result?
Currently I propose to express tolerance in terms of *number of
rounding errors*. Could you clarify you proposition?
>
> Paul
Gennadiy.