The compound energy model is a modification of the bond energy model.
It is based on the use of model parameters defined as Delta degrees U-
AB = degrees U-AB - a'degrees U-AA - a ''degrees U-BB, where a' and a
'' are stoichiometric coefficients in a compound A(a)'B-a '' and may b
e different. The bond energy model was originally defined for a' = a '
' = 1/2 and uses a model parameter v(AB) = E(AB) - E(AA)/2 - E(BB)/2.
Recently, Gates and Wenzl also extended the bond energy model to the c
ase a' not equal a '' but only under the condition that there are no b
onds inside a sublattice. It is now shown that their treatment is iden
tical to the compound energy model in the case of two sublattices. It
appears as a semantic question whether or not the method of solving th
e problem appearing when a' not equal a '' justifies the new name ''Co
mpound Energy Model'' or not. For higher order systems the treatment b
y Gates and Wenzl differs from the compound energy model in that it us
es less parameters. The crucial question is whether this can be justif
ied theoretically or should be regarded as an arbitrary choice of the
relation between the parameters in the compound energy model. The comp
ound energy model can be used for Monte Carlo simulations of short ran
ge order In systems with two sublattices when there are no bonds insid
e the sulattices or when all sites are equivalent as in AuCu3.