Double grid for BSE

Run-time issues concerning Yambo that are not covered in the above forums.

Moderators: myrta gruning, andrea marini, Daniele Varsano, Conor Hogan

Post Reply
kevin
Posts: 2
Joined: Sat Oct 20, 2018 6:31 pm

Double grid for BSE

Post by kevin » Mon Oct 22, 2018 5:25 pm

Dear developers,

I am trying to use the double grid method to plot the BSE spectra of Tellurene. I manage to generate the ndb.Double_Grid (mesh of 4001 k points) file inside the SAVE directory to run with the interpolation. However, i am not sure if just only this file is enough (ndb.Double_Grid) or do i need some extra files from my previous calculation for BSE with converged parameters (e.g The screened coulomb interaction W, kernel, em1s, em1d?). If so, could you tell me which names of the files i need to look into the SAVE directory of my converged calculation?.

Another question quick questions are:
Does Double grid BSE only works with inversion?.
Does Double grid BSE accepts the QP corrections from GW calculation or it is mandatory to use the scizor operator?.

Best regards,
Kevin

Palummo Maurizia
Posts: 15
Joined: Wed Apr 01, 2009 8:10 am
Location: Dipartimento di Fisica Universita' Tor Vergata Roma
Contact:

Re: Double grid for BSE

Post by Palummo Maurizia » Tue Oct 23, 2018 9:04 am

Dear Kevin

yes ndb.Double_Grid must be in the SAVE directory and then you can use it for a BSE calculation using only the inversion solver

In principle you can use it also to calculate the inverse screening to build up the kernel and also for the GW calculation, but I never tested how it works

Yes you can use the ndb.QP

Maurizia
Dipartimento di Fisica
Universita' 'Tor Vergata'
Via della Ricerca Scientifica I
tel.++39-06-72594894
fax.++39-06-2023507
http://www.fisica.uniroma2.it/~cmtheo-group/

ETSF European Theoretical Spectroscopy Facility www.etsf.eu

User avatar
claudio
Posts: 448
Joined: Tue Mar 31, 2009 11:33 pm
Contact:

Re: Double grid for BSE

Post by claudio » Tue Oct 23, 2018 12:31 pm

Dear Kevin

when you have the double grid I advice you to use it for all the yambo calculations GW and BSE

best
Claudio
Claudio Attaccalite
[CNRS/ Aix-Marseille Université/ CINaM laborarory / TSN department
Campus de Luminy – Case 913
13288 MARSEILLE Cedex 09
web site: http://www.attaccalite.com

kevin
Posts: 2
Joined: Sat Oct 20, 2018 6:31 pm

Re: Double grid for BSE

Post by kevin » Fri Oct 26, 2018 4:35 pm

Dear Palummo and Claudio,

Thanks for your answers!,

regards,
Kevin

haseebphysics1
Posts: 169
Joined: Sat Aug 17, 2019 2:48 pm

Re: Double grid for BSE

Post by haseebphysics1 » Sun Mar 15, 2020 9:04 pm

Dear Yambo developers,

I am having an issue with BSE double-grid method. I am following the following tutorial:

http://www.attaccalite.com/speed-up-die ... ith-yambo/

The steps which I have done are as follows:
1: DFT (scf and nscf) with coarse k-points grid.
2: BSE spectra using inversion solver for the above k-mesh.
3: In the same directory, i.e same outdir and prefix.save folder, removing previous DFT wavefunctions only, I did another scf and nscf with dense k-grid.
4: Now, I ran p2y -w, then
5: ypp -m
Now, when I run ypp -m, it says,


Where is the SAVE?

yambo: cannot access CORE database (SAVE/*db1 and/or SAVE/*wf)


But actually ns.db1 and ns.wf is inside the SAVE folder! And I have make sure that code was compiled with netcdf. In fact ypp seems to be working with another save database but not with this double-grid database!!

Can you please identify the mistake that I am doing in these steps?

Thanking you,
Haseeb Ahmad
MS - Physics,
LUMS - Pakistan

User avatar
claudio
Posts: 448
Joined: Tue Mar 31, 2009 11:33 pm
Contact:

Re: Double grid for BSE

Post by claudio » Mon Mar 16, 2020 9:03 pm

Dear haseebphysics1

when you do "ypp -m" for the first time you should specify the folder
where is the folder with many-kpoints, something like

Code: Select all

kpts_map                       # [R] Fine to coarse grid Map
BZ_DbGd_Nk=0                   # BZ Number of points in the double grid
SkipCheck                      # Skip internal consistency check on k-point grids
#noBZExpand                    # Keep the Double Grid k-points in the IBZ without using the symmetries
#NoWeights                     #  Do not print points weight
%DbGd_DB1_paths                # Paths to the SAVE folders with the Double Grid DB1
"./SHIFT/bn.save/" |
%
and then you should run

Code: Select all

ypp
in the

Code: Select all

bn.save/
folder where you have the original calculation
with the 8x8 k-points grid (in this example).

Let me know if it works.

best
Claudio
Claudio Attaccalite
[CNRS/ Aix-Marseille Université/ CINaM laborarory / TSN department
Campus de Luminy – Case 913
13288 MARSEILLE Cedex 09
web site: http://www.attaccalite.com

haseebphysics1
Posts: 169
Joined: Sat Aug 17, 2019 2:48 pm

Re: Double grid for BSE

Post by haseebphysics1 » Mon Mar 16, 2020 10:14 pm

Dear Claudio, thanks for your reply!

I have understood the procedure now, and I am doing it again since I have distorted the database now! I will let you know in case of any problems.

Meanwhile can we please clarify the following things:

Q1: What does this variable means?

BZ_DbGd_Nk=0 # BZ Number of points in the double grid

Is it okay to comment it and hence not using it!

Q2: Does the double grid contains the course-grid points + dense grid points (in BZ)! Am I correct?

Q3: In step 6 of your tutorial, should I run the bse calculation again using the same bse input file with ndb.doublegrid in the SAVE folder and will it automatically use the modified k-points (double k-grid)?

Thanking and wishing you good health,
Haseeb Ahmad
MS - Physics,
LUMS - Pakistan

User avatar
claudio
Posts: 448
Joined: Tue Mar 31, 2009 11:33 pm
Contact:

Re: Double grid for BSE

Post by claudio » Mon Mar 16, 2020 11:39 pm

Dear Haseeb Ahmad
haseebphysics1 wrote: Mon Mar 16, 2020 10:14 pm
Q1: What does this variable means?
BZ_DbGd_Nk=0 # BZ Number of points in the double grid

Is it okay to comment it and hence not using it!
I do not remember this variable, other developers will answer to this question
Q2: Does the double grid contains the course-grid points + dense grid points (in BZ)! Am I correct?
Yes
Q3: In step 6 of your tutorial, should I run the bse calculation again using the same bse input file with ndb.doublegrid in the SAVE folder and will it automatically use the modified k-points (double k-grid)?
yes, if the caculation does not start, just remove the ndb.BS* databases

best regards
Claudio
Claudio Attaccalite
[CNRS/ Aix-Marseille Université/ CINaM laborarory / TSN department
Campus de Luminy – Case 913
13288 MARSEILLE Cedex 09
web site: http://www.attaccalite.com

sitangshu
Posts: 175
Joined: Thu Jan 05, 2017 8:08 am

Re: Double grid for BSE

Post by sitangshu » Mon Oct 31, 2022 6:18 am

Dear Yamboers,

I already have the ndb.Double_Grid file in my SAVE directory, how to do a GW calculation? Is it sufficient to run the standard GW input script with this SAVE folder?

Regards,
Sitangshu

claudio wrote: Tue Oct 23, 2018 12:31 pm Dear Kevin

when you have the double grid I advice you to use it for all the yambo calculations GW and BSE

best
Claudio
Sitangshu Bhattacharya
Indian Institute of Information Technology-Allahabad
India
Web-page: http://profile.iiita.ac.in/sitangshu/
Institute: http://www.iiita.ac.in/

User avatar
claudio
Posts: 448
Joined: Tue Mar 31, 2009 11:33 pm
Contact:

Re: Double grid for BSE

Post by claudio » Wed Nov 02, 2022 10:05 am

Yes, it is sufficient to run a standard GW input,
in the report you will indication about the use of the double-grid
Claudio Attaccalite
[CNRS/ Aix-Marseille Université/ CINaM laborarory / TSN department
Campus de Luminy – Case 913
13288 MARSEILLE Cedex 09
web site: http://www.attaccalite.com

Post Reply