[molpro-user] a pspace error propblem with multi

H. -J. Werner werner at theochem.uni-stuttgart.de
Mon Dec 4 07:48:25 GMT 2006


You can try thresh,pspace=1.d2 to avoid the problem. The program
should actually print a message telling you about this possibility.
Best regards
H.-J. Werner
On Mo, 04 Dez 2006, Wang Bingwu wrote:

>Dear Molpro users,
>
>I encounted a confused problem when doing the multi calcultion.
>Using the same input including basis, initial orbital and multi input
>specification, it can get converged results sometime, but often occured a
>non-zero pspace error depending the time and platform. For example, I got
>well converged results on a AMD opteraon + redhat9 platform, but occured
>non-zero pspace error when run at a Itanium2 + redhat platform using
>complete same input.  It's real confused and disappointed problem when need
>transfer a big job from small platform to a big platform. Does anyone else
>have this problem?
>
>Best regards,
>
>Wang Bingwu
>Chemistry Department, Peking University
>Tel: 86-10-62765703
>email: bwwang at pku.edu.cn

-- 
Prof. Hans-Joachim Werner
Institute for Theoretical Chemistry
University of Stuttgart
Pfaffenwaldring 55
D-70569 Stuttgart, Germany
Tel.: (0049) 711 / 685 64400
Fax.: (0049) 711 / 685 64442
e-mail: werner at theochem.uni-stuttgart.de



More information about the Molpro-user mailing list