[Wien] Band gap calculation

[Gavin Abo]

Difference might happen if it not converged [1].  Did you do a k-point 
convergence test [2]?

It might depend on the gap you talking about, refer to [3].  If indirect 
gap [4], I don't recall how the algorithm that gives the value for the 
:GAP for scf* works.  It might be that it is a limitation of the 
algorithm.  Someone else might know or remember the algorithm 
mechanism.  There is also the source code that you could study yourself 
to determine why.

It might also depend on whether you used a shifted or non-shifted k-mesh 
[5,6].

[1] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg13466.html
[2] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg15599.html
[3] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg15445.html
[4] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg19404.html
[5] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg05699.html
[6] 
https://www.mail-archive.com/wien@zeus.theochem.tuwien.ac.at/msg18040.html

On 12/22/2019 10:12 PM, Peeyush kumar kamlesh wrote:
> Dear WIEN2k users!
> Is it ok, if we get different values of band gap from scf2 file and 
> spaghetti calculations?
> Thank you!