The problem isn't how the numbers are stored. It is that when a number

is stored as an integer, LB, seems to think you want to do "infinite

precision" math on it, which can take a horrendous amount of time if

there are a lot of digits. That issue mainly arises with powers, which

can build up a lot of digits quickly. In many cases, you know whether

you are dealing with such numbers. But if you are working with

user-supplied values, anything is possible.

The point is that there are cases where the programmer needs to be

sure that floating point math is performed. This is normally done by

using a floating point data type, but in LB it needs to be done a

little more indirectly. That was the purpose of the Power function I

posted.

Sam W.

--- In libertybasic@yahoogroups.com, "rod_bird001" <rodbird@...> wrote:

>

> Just to say that Liberty will store a variable as a float whenever it

> becomes one and as an integer whenever it becomes one. Simple

> straightforward strategy.

>

> I'm not into maths at the level you are, I do struggle to understand

> how the strategy prevents maths operations.

>