Restarting CCSD from reduced amplitude range

[jvi...@gmail.com]

Dear Dirac Experts,

is it possible to restart CCSD computation (4c) from another CCSD computation on the same molecule with the same AO and MO basis set, but with smaller/different .ACTIVE amplitude range?

I thought that it might resolve some problems with the convergence if there would be initial guess for amplitudes from already converged (or almost) computation with just subrange of amplitudes (these are quicker and converge easily, but of course, less accurate).

Restarting CCSD computation from sligthly different molecular geometry could be possible, I suppose? (sure, one should fix AO-MO coefficients by .PHCOEFF in both runs).

Best regards,

Jakub