The relationship between K-point and QP states

[202321070091]

Dear all,
Before performing Yambo calculations, I tested two different K-point grids, 20×20×1 and 36×36×1. I attached the input and output files for both self-consistent and non-self-consistent QE calculations. After conducting YAMBO processing, I obtained the r_setup file. I noticed an issue with the reported K and Q points in this file. For instance, in the r_setup file for the 20×20×1 grid, the IBZ Q-points are shown as 44, matching the IBZ K-points at 44, and the QP states are 1 44, which seems normal. However, in the 36×36×1 grid results, the IBZ Q-points are 127, identical to the IBZ K-points, yet the QP states are 1 127. I am seeking clarification on this discrepancy; could you advise which scenario is correct?
Yours sincerely, WangQi

[Daniele Varsano]

Dear WangQi,

sorry, but I have not got what problem you are facing. In the setup yambo assign indexes to various quantity.
What is reported in the last part of the report (r_setup) is the indexing for the QP wrt the k points, [WR./SAVE//ndb.kindx].
They are correctly as much as the k point sampling.

Best,
Daniele

[202321070091]

Dear Daniele，
Thank you for your patient reply. I apologize that my previous explanation was unclear. What I meant to say is that when using a grid of 36×36×1, the K/Q-points units shows a total of 50 Q/K points in the r_setup file, which does not match the 127 QP states. However, when using a 20×20×1 k-point grid, these two numbers are the same. So, I would like to ask if the number of Q/K points needs to be consistent with the number of QP states.
Yours sincerely, WangQi.

[Daniele Varsano]

Dear WangQi,

that's just a formatting reason, please read the line just after your red rectangle

Best,

Daniele