Mismatch input versus output

[dagosta]

Hi,

I am running a convergence test of some of the GW parameters as suggested in the tutorials for a bulk material. I build the many input files for the NGsBlkXp variable, and check that each of them sports the correct values. However, when I run yambo, for certain input files, I get the value of that variable changed. For example, in one case I am running the NGsBlkXp = 7 case, and get the output of the NGsBlkXp = 9 calculation. (I checked, they are indeed identical.) Although, the report files report to have read the correct input, but it returns the wrong output.

gw_ppa_90b_7ry.txt

See the (shortened) report file. Is there anything wrong I am doing? Is Yambo assigning a value to NGsBlkXp that is compatible with the other variables? This is probably not strongly affecting my test but I would like to know if this is an expected behaviour.

Best regards,
Roberto

[Daniele Varsano]

Ciao Roberto,
When assigning blocks in RL yambo find the number needed to close a shell of g vectors closest to the number indicated in input. This is done in order to not to break the symmetries of the system. Note that instead of RL you canalso assign the matrix size in ferma of energy cut-off indicating Ry, Ha, mRy etc instead of RL.
Best,
Daniele