Transferred momenta grid building error

[will]

Dear experts,
My system is monolayer WSe2. When I used a 12*12*1 shifted k-grid, the initialization setup was stopped on the point of building the transferred momenta grid. And when
a Gamma-centered k-grid is chosen, everything is ok.
Noticed that other users have asked the same question as me
http://www.yambo-code.org/forum/viewtopic.php?t=1761,
I uncommented the tag "NoDiagSC" in the input file.
But the problem remains. I uploaded the input and report files. Expect your help!

Best
Will

[Daniele Varsano]

Dear Xiaowei,
the report is not attached and it will be useful.
Anyway, have you tried to delete the ./SAVE/ndb.kindx and run the setup again using the NoDiagSC flag?

Best,
Daniele

[will]

Dear Daniele,
Thanks for your quick reply! I'm sorry that I've forgotten to attach the report file.

Anyway, have you tried to delete the ./SAVE/ndb.kindx and run the setup again using the NoDiagSC flag?

Yes, I did. But it didn't work.

Best
Xiaowei

[Daniele Varsano]

Dear Xiaowei,
this is the log file and not the report.

Best,
Daniele

[will]

I have edited my last post. Please check the attachment files.
Thanks!
Xiaowei

[Daniele Varsano]

Dear Xiaowei,
in this case, there is not much to do.
It happens that in some cases QE adds some extra k points that make the q grid as non-uniform.
The last try you can do is to delete the database ndb.kindx and try to run the setup with these lines only in input:

Code: Select all

setup
NoDiagSC

i.e. removing all the other keywords in input.

Next, if it fails and you really need a shifted grid you can try to generate a shifted grid having the same q points of the one obtained with the gamma centred sampling. I do not know if this fits your needs.

Here you can find the procedure on how to proceed.

Best,
Daniele

[will]

Dear Daniele,
Thanks a lot! The problem remains if I just keep the two tags "setup" & "NodiagSC".
Anyway, I will try the instruction you shared! Hope it works!

Best
Xiaowei