[Wien] Iterative diagonalization in SCF cycles

[msoumeli at physics.auth.gr]

Dear Wien2k users,

I am running wien version WIEN2k_09.2. I am studying semiconductors  
(nitrides) and using TELNES.2. During the SCF cycles I often come up  
to QTL B value problems.
I have tried changing RmtKmax, k points, E max, Energy cut-off. I am  
also doing core hole approaches and my problem is that the scf cycles  
take many weeks in order to converge or others do not converge at all.  
when i tried iterative diagonalization in a case  that had previously  
failed, it converged in a day.

My question is if iterative diagonalization in scf cycle affects the  
accuracy? Are there any restrictions in its use? Will it give the same  
results with full diagonalization or not?

Thank you in advance.

M.Soumelidou