The keyword shift of CASPT2 calculations

[wangbw]

Dear Molpro users,

To improve the convergence behavior of CASPT2 calculations, I used the keyword-shift after RS2C, but using the different shift values, it got different total energies and "Energies without level shift correction" in the output file. Which energy can I use or is there any choose condition about the shift value selection?  

Thanks.

WangBw

Chemistry department
Peking University
Beijing, China