Convergence tests

Having trouble compiling the Yambo source? Using an unusual architecture? Problems with the "configure" script? Problems in GPU architectures? This is the place to look.

Moderators: Davide Sangalli, andrea.ferretti, myrta gruning, andrea marini, Daniele Varsano, Conor Hogan, Nicola Spallanzani

Forum rules
If you have trouble compiling Yambo, please make sure to list:
(1) the compiler (vendor and release: e.g. intel 10.1)
(2) the architecture (e.g. 64-bit IBM SP5)
(3) if the problems occur compiling in serial/in parallel
(4) the version of Yambo (revision number/major release version)
(5) the relevant compiler error message
User avatar
Daniele Varsano
Posts: 3773
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: Convergence tests

Post by Daniele Varsano » Mon Jan 31, 2022 11:10 am

Dear Mitra,

you can use separately. While GTermKind performs very well and it is suggested, XTermKind is discouraged as in many case you won't gain much performance.
So:
GTermKind =BG and XTermKind=none,

is legitimate and suggested.

Best,
Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/

8813204602
Posts: 58
Joined: Tue Jun 15, 2021 5:53 pm

Re: Convergence tests

Post by 8813204602 » Fri Feb 04, 2022 10:19 am

Dear Daniele,

I have two questions about convergence tests:

1) I did the convergence tests and the GW calculations for a 2D material using the Dojo quasipotentials, now I want to do these calculations again using the new quasipotentials of the ONCV. Do I have to do all the convergence tests again from the beginning?
2) To perform GW calculations for members of a family, such as VCl2&VI2, should convergence tests be performed for each separately, or can the same convergence parameters obtained for one member be used for the rest of the family?

Thanks & Regards ,
Mitra Helmi
Mitra Helmi,Ph.D student
Ferdowsi university of Mashhad
Mashhad,Iran

User avatar
Daniele Varsano
Posts: 3773
Joined: Tue Mar 17, 2009 2:23 pm
Contact:

Re: Convergence tests

Post by Daniele Varsano » Fri Feb 04, 2022 2:27 pm

Dear Mitra,

1) They could be similar, but some tests I think is needed, in particular if they have different valence electrons (semi-core etc.).
A side note: Dojo uses non linear core corrections. Yambo is able to treat them activating the flag Usenlcc, but some problems can arise as nlcc are not considered e.g. in the Fock term, this is exactly the same issue why pseudo with nlcc are not suggested for hybrid functional in QE. So, making the long story short I would go for psp without nlcc as for instance ONCV sg15

2) The electronic structure differs, so separate tests are needed.

Best,

Daniele
Dr. Daniele Varsano
S3-CNR Institute of Nanoscience and MaX Center, Italy
MaX - Materials design at the Exascale
http://www.nano.cnr.it
http://www.max-centre.eu/

Post Reply