CCL: energy sharply changed after CASPT2



 Sent to CCL by: "Pengchen  Ma" [mapengchen[-]iccas.ac.cn]
 Dear all,
 Recently I used MOLCAS to do a caculation,but the energy gap are lagerly
 different between RASSCF and CASPT2.Besides,DFT calculation result is the same
 with RASSCF. In consideration this result,I wonder if I choosed a wrong active
 space or some input error I ingnored?or other reasons?
 Here is the input file of CASPT2.
  &CASPT2
  Title
  a-1
  Multistate
  1 1
  frozen
  32
  IPEAshift
  0.25
  Imaginary
  0.1
  Noprop
  End of Input
 Any suggestion and information will be appreciated,thank you
 Pengchen Ma