full-frequency calculation gives positive sigma

Submitted by rangel on Tue, 10/29/2013 - 13:46


Bug reports

I am testing the FF capabilities of BerkeleyGW .

The calculation in Sigma, gives me these errors several times:
l = 24 n1 = 24 E_l = 9.145 E_n1 = 8.496 wx = 51.041 E_max = 44.800 WARNING: The real frequency range is too small.

This is strange, since I am calculating sigma up to 50 eV
The parameters I set are:
The FF parameters I used are:
frequency_dependence 2
init_frequency_eval 0.0
delta_frequency_eval 0.1
number_frequency_eval 501

Then at the end, I have a positive imaginary Sigma.
The code gives me this message:
WARNING: You have a positive imaginary Sigma at k = 0.000000 0.000000 0.000000

Do you have any hint?