The Vogel-Fulcher-Tammann (VFT), Avramov and Milchev (AM) as well as Mauro,
Yue, Ellison, Gupta and Allan (MYEGA) functions of viscous flow are analysed
when the compositionally independent high temperature viscosity limit is
introduced instead of the compositionally dependent parameter ??. Two
different approaches are adopted. In the first approach, it is assumed that
each model should have its own (average) hightemperature viscosity parameter
??. In that case, ?? is different for each of these three models. In the
second approach, it is assumed that the high-temperature viscosity is a truly
universal value, independent of the model. In this case, the parameter ??
would be the same and would have the same value: log ?? = ?1.93 dPa?s for all
three models. 3D diagrams can successfully predict the difference in
behaviour of viscous functions when average or universal high temperature
limit is applied in calculations. The values of the AM functions depend, to a
greater extent, on whether the average or the universal value for ?? is used
which is not the case with the VFT model. Our tests and values of standard
error of estimate (SEE) show that there are no general rules whether the
average or universal high temperature viscosity limit should be applied to
get the best agreement with the experimental functions.