...
Exclusions: offers a list of entry points to exclude from calculation linkbase validation.
Calculation 1.1 rounding: the rounding mode used for validation with Calculation 1.1. Only enabled when “Calculation Linkbase Validation Mode” is “Enabled” or “Auto”. There are two modes to choose from: "round-to-nearest" and "truncation" mode. We recommend to use "round-to-nearest" for your reports.
While not yet mandatory for ESEF, the new implementation of the calculation linkbase in version 1.1 can be chosen in the Settings.
You can choose between "Auto", "Enabled" and "Disabled". Only "real" calculation errors are displayed, no rounding issues are displayed.
Auto: includes the list of exclusions (taxonomies where no calculation linkbase validation will be applied):
Enabled: calculation linkbase validation always enabled (regardles of selected taxonomy).
Disabled: calculation linkbase validation always disabled.
Calculation Linkbase Validation Severity Error Threshold (in percent): when selected, a percent value can be set to define a threshold up to which calculation linkbase inconsistencies are still shown as WARNING instead of ERROR. Only selectable when “Calculation Linkbase Validation Mode” is “XBRL 2.1”.
In order to distinguish rounding inconsistencies from real calculation issues (that usually indicate tagging issues), the AMANA XBRL Engine allows to enable a Calculation Linkbase Validation Severity Error Threshold, which must be configured in the XBRL (user based) settings. If the threshold is activated and set to e.g. 1%, it means that the all calculation inconsistencies where the delta is smaller than 1% of the rounded reported value are deemed to be WARNINGS, while all inconsistencies with differences bigger then the threshold are deemed to be ERRORS. If the threshold is not set, all calculation inconsistencies are signalled with a WARNING severity.
Further reading: https://www.xbrl.org/esef-errors-and-common-pitfalls-8-rounding-and-calculations/
...