This "evolution" in
model atmosphere calculation has been driven by the enormous increase
in the quality of observational data (CCD spectroscopy), the
vast increase of computation power and the progress
in the field of atomic data needed to calculate especially the
absorption properties of the stellar atmospheric environment.
Whilst "simple" grey models are only rarely used today
there are in principal two ways to treat the problem of absorption
in the atmospheric layers of a star. The first one is based on
Opacity Distribution Functions(ODF) and treats bound-free absorptions
based on labels. The second one is the Opacity Sampling(OS) approach
that tries to treat all absorption features in a statistically "complete"
way. Whilst the first approach allows fast computations of even
small-meshed grids of models, the second one is much more time
consuming, but more flexible in the sense, that individual
element mixtures (adopted to the star analyzed) can be used.
In principal the second approach allows to account for
deviations from the assumption of local thermal equilibrium (LTE)
as is partly done by the