On 11/9/2009 7:30:17 PM, ElSid wrote:
>I do not think this is a bug, so I am posting here.
It isn't. It's a round-off error.
>Is there a way of having Mathcad move the constant to the denominator instead of calculating the value and truncating to the set number of decimal places?
It's a convention that numbers whith decimal dot are "at machine precission" rationals, or "realcons", or other names, depending the language, including for example RPN from Hewlett-Packard.
Actually, it's a short hand for declaring float point values. You can input 1.08 as 108/100 to avoid truncation errors. The reciprocal is true: to input values like 2 as floating point write 2. or 2.0.
Im mcad/maple you can write convert(expr,rational) to have all values as exact ones, and I remember a bug in mcad14 related with truncation and laplace, but searching in the forum seems broken.
Regards. Alvaro.
PD: Time ago I discover that users with a long experience using microsoft word find death symbolics comfortable.