Page 3 of 3

Re: Abinit to yambo

Posted: Sat Jun 05, 2021 11:13 am
by FGH
Dear Davide,

Re: Abinit to yambo

Posted: Sat Jun 05, 2021 11:13 am
by FGH
Dear Davide,

Re: Abinit to yambo

Posted: Sat Jun 05, 2021 2:50 pm
by Daniele Varsano
Dear Farzaneh,
I can't see any problem.
If you inspect your outputs carefully you can see that the number of G's used in the screening (X G`s ) are the same in some of the files:

Code: Select all

o-10b_1Ry.qp.txt:# X G`s                                           :   9 [used]
o-10b_2Ry.qp.txt:# X G`s                                           :  27 [used]
o-10b_3Ry.qp.txt:# X G`s                                           :  27 [used]
o-10b_4Ry.qp.txt:# X G`s                                           :   51 [used]
o-10b_5Ry.qp.txt:# X G`s                                           :   65 [used]
o-10b_6Ry.qp.txt:# X G`s                                           :   65 [used]
o-10b_7Ry.qp.txt:# X G`s                                           :   89 [used]
o-10b_8Ry.qp.txt:# X G`s                                           :  137 [used]
o-10b_9Ry.qp.txt:# X G`s                                           :  137 [used]
This is because yambo consider a number of G such that the G-shell are closed.
In order overcame this issue I suggest you to assign in input the size of the matrix in RL instead of Ry e.g.

Code: Select all

NGsBlkXp= 200 RL 
Best,
Daniele

Re: Abinit to yambo

Posted: Sat Jun 05, 2021 3:41 pm
by FGH
Dear Daniele,

Ok, but as I should compare my abinit and yambo calculations with each other I must to choose in Ry. Is ther any other solution to solve this problem?

Re: Abinit to yambo

Posted: Sat Jun 05, 2021 3:49 pm
by Daniele Varsano
So, you can continue to use rydbergs with larger intervals in order to avoid repetitions.

Best,
Daniele

Re: Abinit to yambo

Posted: Wed Jun 09, 2021 9:24 am
by Laura Caputo
Dear developers,
Thanks for the new converter. As I understand, however, calculations already done with the old one should not be re-done again, since the results are correct but the number of G-vectors are higher than it should be. Right?