<div>dear all,</div><div> in INPUT_LD1.html about ecutmin,ecutmax,decut we have :</div><div>"</div><div>Parameters (Ry) used for test with a basis set of spherical</div><div>Bessel functions j_l(qr) . The hamiltonian at fixed scf</div>
<div>potential is diagonalized for various values of ecut:</div><div>ecutmin, ecutmin+decut, ecutmin+2*decut ... up to ecutmax.</div><div>This yields an indication of convergence with the</div><div>corresponding plane-wave cutoff in solids, and shows</div>
<div>in an unambiguous way if there are "ghost" states</div><div>"</div><div><br></div><div>how does ld1.x say "there is a ghost state somewhere" ?</div><div><br></div><div>when testing a PAW with ld1.x I have the following error:</div>
<div><br></div><div>"</div><div> Cutoff (Ry) : 50.0</div><div> N = 1 N = 2 N = 3</div><div> E(L=0) = -3.2666 Ry -0.2491 Ry -0.0148 Ry</div><div> E(L=1) = -1.8963 Ry -0.0931 Ry 0.0105 Ry</div>
<div> E(L=2) = -0.0637 Ry 0.0203 Ry 0.0602 Ry</div><div><br></div><div> Cutoff (Ry) : 55.0</div><div> N = 1 N = 2 N = 3</div><div> E(L=0) = -3.2666 Ry -0.2491 Ry -0.0148 Ry</div>
<div> E(L=1) = -1.8963 Ry -0.0931 Ry 0.0105 Ry</div><div> E(L=2) = -0.0637 Ry 0.0203 Ry 0.0601 Ry</div><div><br></div><div> Cutoff (Ry) : 60.0</div><div> N = 1 N = 2 N = 3</div>
<div><br></div><div> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%</div><div> from rdiags : error # 73</div><div> failed to converge (factorization)</div><div> %%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%</div>
<div><br></div><div> stopping ...</div><div><br></div><div>"</div><div>what does it mean ?</div><div>thanks.</div><div><br></div>