Page 1 of 1

Yambo gets incorrect k-point sampling from KSS file

Posted: Mon Dec 06, 2010 3:06 am
by Guangfu Luo
Dear friends,

Actually, I mentioned this matter and shortly discussed it with Dr. Daniele Varsano in another irrelevant topic. Here I'd like to start a new topic right for this issue.

I found this problem while testing the BS spectrum of Graphene. I checked the k-points Yambo used in r_optics_bse_em1s_bss and found some of them are different from the k-points that KSS file contains (I confirmed this by using the cut3d tool in Abinit). The different k-points in Abinit and Yambo are attached for your inspection, together with the test-purpose input files for Abinit and Yambo (they can be finished in less than 5 minutes).

This problem can cause some bad results, because Yambo misses some high-symmetry points.

Could you please enlighten me?

Re: Yambo gets incorrect k-point sampling from KSS file

Posted: Mon Dec 06, 2010 8:43 am
by Daniele Varsano
Dear Guangfu Luo,

I don't think Yambo the k-point are different, they are just in
different coordinate systems: cartesian vs reciprocal lattice vectors.
The k-point is yambo report are reported in [iku].

Here it's reported hour reciprocal lattice.

Code: Select all

  Reciprocal Lattice(RL) unit cell [iku]
  B1 = 1.00     0.500     -.500E-15
  B2 = 0.00      1.00     -.500E-15
  B3 = 0.000000  0.000000  1.000000
Hope it helps,

Daniele

Re: Yambo gets incorrect k-point sampling from KSS file

Posted: Tue Dec 07, 2010 10:17 am
by Guangfu Luo
Dear Daniele,

Thanks so much for your reminder, and I am much relieved. (I am going to confirm the relationship). By the way, I just realized that this issue had been discussed before (viewtopic.php?f=14&t=113).

Sincerely,
Guangfu Luo