<div>Dear users,</div>
<div> </div>
<div> </div>
<div>The processes to use the 'force theorem' are</div>
<div><em></em> </div>
<div><em>(i) converge calculation without s-o<br></em><i>(ii) using converged non s-o calculation run lapwso, lapw2up/dn for<br></i><i> M along symmetry directions. Anisotropy is then given by differences<br></i><i> of [:SUMup + :SUMdn] for M along say [001] and [100].<br>
</i></div>
<div><i></i> </div>
<div><i>I want to make sure if my understanding is correct.</i></div>
<div><i>My understanding is (1) do spin polarization calculation first, (2) do LS-coupling calculation with specific spin directions. (3) take the sum of eigenvalues and compare them.</i></div>
<div><i></i> </div>
<div><i>Then the difference of the sum of eigenvalues are in Ryd units?</i></div>
<div><i>And I used k-points of 10,000 for this test run. I want to make sure how many k-points "usuallly" needed for tetragonal two atoms structures.</i></div>
<div><i></i> </div>
<div><em>Thank you in advance.</em></div>
<div><em></em> </div>
<div><i></i> </div>
<div><i>All my best,</i></div>
<div><i>Jihoon Park</i></div>