Page 1 of 1

Initialization with verbosity in yambo-4.4.0

Posted: Mon Dec 23, 2019 7:23 am
by xueshanxihua
Dear developers,

I found something strange in the initialization step of yambo-4.4.0.

Usaully, we simply use the command "yambo -i" to generate the input file, and run "yambo" to complete the initialization. In this case, a block named "QptCoord" doesn't appear in the generated input file, but from the "r_setup" file, it's found to be empty:
%QptCoord
%
However, if I generate the input file with verbosity, i.e., "yambo -i -V all", the default value of this block becomes
%QptCoord # [KPT] [iku] Q-points coordinates (compatibility)
0.000000| 0.000000| 0.000000|
%
And in the initialization step it leads to additional informations as
<---> [RL indx] Q-grid:User defined / from ndb.kindx.
<---> [RL indx] Q BZ pts are /= from X grid BS pts. Gamma point only.

I can still complete the initialization in both cases. But when I try to generate the input file for a real MBPT calculation, I find some paramemters/blocks are missing if the verbosity is used in the initialization. For example, attached below are two input files generated by "yambo -r -o eh -k sex -b -y d -V par", from the "SAVE" folder initialized with and without (wo) verbosity, respectively. Clearly, many basic parameters are missing in the verbosity case.

I'm not sure whether it's a bug or I just get something wrong in compilation, and any help will be gratefully appreciated.

Thank you very much,

Re: Initialization with verbosity in yambo-4.4.0

Posted: Mon Dec 23, 2019 8:50 am
by Daniele Varsano
Dear Zeyu Jiang,
Note that in the second case your setup builds databases as gamma points only, the grid is not seen as uniform and this prevents to go on with the calculations (build-up of the screening). The correct way to proceed is to leave the input file as it is without defining any %QptCoord, otherwise, you should need to define all the q points of your sampling. This part of the input is meant to initialize databases for extra k points compatible with the q grid of a previous calculation. For what I can see it can generate confusion so we will reshape it.
Thanks for reporting,

Best,

Daniele