Campbell CR6 Series, Руководство оператора

Страница: 221 / 690

Section 7. Installation
221
When a number is assigned to a variable declared As Double , it carries through to
the variable with single precision unless it is appended with R . Consider the
following examples:
Double- Precision Math: (Double = Double) Results in Double
Dim ND As Double 'Double precision declaration
BeginProg
Scan ()
ND = 1R + 253546R 'Double precision math
ND = 253547 'Double precision result
NextScan
EndProg
Double-Precision Math: (Single = Double) Results in Single
Dim NS 'Single precision declaration
BeginProg
Scan ()
NS = 1R + 253546R 'Double precision math
NS = 253500 'Single precision result (because NS is declared
single precision)
NextScan
EndProg
Double Precision Math: (Double = Single) Results in Single
Dim ND As Double 'Double precision declaration
BeginProg
Scan ()
ND = 1 + 254546R 'Single precision math
ND = 253500 'Single precision result (because 1 is single
precision)
NextScan
EndProg
By the way, if you need to force a value to single precision, or at least be assured
that it is single precision, append it with an F .
A few operations are performed as double precision by default. These are
AddPrecise()
,
Average()
,
AvgRun()
,
AvgSpa()
,
CovSpa()
,
MovePrecise()
,
RMSSpa() , StdDev() , StdDevSpa() , Totalize() , and TotRun() . Double
precision has a 32-bit extension of the mantissa, resulting in 56 bits of precision.
Floating-point arithmetic is common in many electronic, computational systems,
but it has pitfalls high-level programmers should be aware of. Several sources
discuss floating-point arithmetic thoroughly. One readily available source is the
topic Floating Point at www.wikipedia.org. In summary, CR6 programmers
should consider at least the following:
• Floating-point numbers do not perfectly mimic real numbers.
•
Floating-point arithmetic does not perfectly mimic true arithmetic.