EPW Forum

Dear Prof. Roxana Margine:

Thank you very much for your kindly answer.
But, I still guess that the item (1) in my previous message holds true.

Sincerely,

Hyungjun Lee

I see. Prof. Roxana Margine indicated that we have to set epbread to be "true" different from Samuel's suggestion.

In this case, I guess no problem, but I think that there are some duplicated processes which might waste computing time.
Also, to the best of my knowledge, we have to use the same number of pools when reading the previous *.epb* files while we can use the different number of pools when reading *.epmatwe* files, etc. since it is the merged file. This is just my guess. I absolutely don't know well the internal structure of EPW.

I mentioned that there might be some problem with missing information such as bg, etc. if we have to set epbread to be "false" and epwread to be "true..

Very sorry for inconveniences.

Sincerely,

Hyungjun Lee

Hello Hyungjun Lee,

The info about lattice parameters etc should be read from the "crystal.fmt" file.
I'm not 100% sure it works in the superconducting case. It will indeed work when reading the .epb but then you cannot use different number of cpu which
could indeed be limiting.

I definitely need to spend some time on the superconducting part of EPW !

Best,
Samuel

Dear Samuel:

Thank you very much for your kindly answer.

Sincerely,

Hyungjun Lee