ECP calculation of La using Gaussian 98

From: Denny <dilys98 at.at mails.tsinghua.edu.cn>
Organization: Tsinghua University
Subject: ECP calculation of La using Gaussian 98
Date: Thu, 30 Aug 2001 16:15:29 +0800

Hi all,
 When I calculate the energy of La atom, if I change the initial
 orbit guess with the keyword guess=alter, I even find lower
 enery. Then it means Gaussian 98 can't give the ground state of
 La automatically? If so then how can I be sure the orbital order
 what I get correspond to the ground state?
 In addition I even found using the same input file, Gaussian 98
 for windows A3 and Gaussian 98 for linux A7 can give different
 results. Even the number of primitive basis functions are
 different. What goes wrong?
             Denny
             dilys98 at.at mails.tsinghua.edu.cn