ENCUT AND KPOINTS questions

[ufo991]

A pseudo-potential file of vasp has two energy-cutoffs (ENMIN and ENMAX) for the cutoff energy of plane wave. The vasp code does not guarantee the correct results are obtained if the energy-cutoff of plane-wave is smaller than ENIMN or larger than ENMAX. The large ENCUT make plane-wave base-set more completed but loss the accurate of pseudo-potential. There are exceptions that need increasing ENCUT such as in the calculations of stress tensor and the smooth P-V curves, but the ENCUT is still required no larger than 1.2-1.3 ENMAX to assure the correctness of pseudo-potential (See vasp manual) .

For orthorhombic structure, the k-points ixjxk is consistent with the shape of cell in real-space. The k-points mesh is proportional to 1/a,1/b,1/c.

I saw it from someone else. Is it right?

[admin]

the results may be inaccurate if the the cutoff is too small, but increasing the cutoff may ALWAYS improve the results
(you must have misunderstood the manual ENCUT = 1.3* ENMAX is a good compromise to get accurate results
with a reasonably increased basis set size)