I understand very reviews has 3 phases if this renders a study, and with respect to the phase it might read static data or arranged data fields etc, but can there be any benefit writing formulas in these phases? The finish result ought to always be exactly the same basically specify the phase or otherwise?

The finish result will not continually be exactly the same and omitting evaluation time functions can occasionally result in incorrect results. For instance, consider that you've a report that's arranged which you are attempting to calculate simple subtotals of these groups having a variable. Say you're resetting the variable within the group header and exhibiting it within the footer while upgrading it within the particulars section via:

numbervar subtotal;
subtotal := subtotal + {table.numericvalue}

Very will evaluate this formula while reading through records and before the records have been in a foreseeable, arranged, and sorted order. It isn't before the second pass, the whileprintingrecords pass, that records happen to be arranged and sorted. This is exactly why group selection occur in this pass.

Case an example, but you will find various ways to interrupt a study by omitting evaluation time functions, mostly getting related to using variables (a minimum of will be able to think about). Check out this resource for particulars on Crystal's multiple passes: The Multi-Pass Reporting Process of the Crystal Reports Engine