Precision and Self-Consistency

The precision of a calculation is determined by

• The function sets (basis sets, levels of frozen core approximation,
  and fit sets for the computation of the Coulomb potential)
• Numerical integration settings in real space and in k-space
• Convergence criteria (for the SCF procedure and the geometry optimization)
• A few more items that are rather technical and usually irrelevant (these are not discussed here).

The fragments you attach determine, through the fragment files, the function sets. Since each fragment traces back to one or more Create runs, the employed data base files in the Create runs determine the finally employed function sets.

For convergence of the geometry optimization see the key GEOMETRY.

In this part we examine numerical integration and the SCF procedure.