Molcas Forum

shuoshuo

Member

From: Beijing Normal University

Registered: 2017-10-18

Posts: 35

Frequency calculation takes too much time when using &MCKINLEY

Dear Molcas experts,

I use molcas to calculate a transition state. The frequency is provided at the end of the calculation.

&SLAPAF
  FindTS
  TSConstraints
    A = Bond H1 O2
  Values
    A = 1.63 Angstrom
  End of TSConstraints

Timing: Wall=1502.86 User=1355.73 System=121.65

Meanwhile I use the optimized structure to do the frequency calculation.

&MCKINLEY
  Perturbation
  Hessian

Timing: Wall=10266.42 User=9550.99 System=627.86

I found the latter used much more time. I hope the frequency calculation can be as fast as the first example. Also, I don’t care about accuracy.

Ignacio

Administrator

From: Uppsala

Registered: 2015-11-03

Posts: 1,085

Re: Frequency calculation takes too much time when using &MCKINLEY

The first example doesn't give you a frequency calculation. It gives you some estimated frequencies from the approximate Hessian, which could be very wrong.