From Gerris

Revision as of 18:50, 24 October 2009; view current revision
←Older revision | Newer revision→
Jump to: navigation, search

GfsOutputErrorNorm is used to write the norms of the difference between a scalar field and a user-defined reference field.

The norms are written using the same formatting as for GfsOutputScalarNorm.

The syntax in parameter files is:

[ GfsOutputScalar ] {
  s = [ GfsFunction ]
  unbiased = 1
  v = Error
  w = [ GfsFunction ]
  relative = 0

where s is the function defining the reference field and v is used to store the difference between the scalar and reference fields (v can be left undefined in which case a temporary variable is used).

If unbiased is set to one (default is zero), the bias (i.e. mean value of the difference) is substracted before computing the norms.

The norms can be weighted using the optional parameter w.

If relative is set to one, the norms are divided by the corresponding norms of the reference solution.


Personal tools