Definition
The determination of a specific accuracy decides which number format is used internally for calculating key figures. The degree of accuracy is determined in the InfoObject maintenance for each key figure and in the query monitor for all query key figures.
Use
When converting decimal figures with numbers following the decimal point into binary floating-point numbers, unavoidable rounding errors occur. This is because many decimals (for example, 0.1) correspond to periodic fractions, and are therefore unable to be displayed exactly.
If you choose Calculate with Packed Numbers in the query monitor, packed numbers (ABAP data type P, with nine decimal places) are used as the internal number format for all query key figures instead of binary decimals. Rounding errors are avoided with commercial calculations with this fixed-point number, because in doing so no conversion of numbers into the dual system takes place.
Choose this option for those application areas in which a high accuracy is required (for example, for queries in the accounting and cost accounting areas).
Because calculating with packed numbers causes a high processing status overall, we recommend not using this number format for all query key figures, but rather only for those key figures for which a very high degree of accuracy is required.
You have to make this setting in the InfoObject maintenance for the respective key figure: