[CPMD-list] test calculation problems...

[Michael Brian Sullivan]

I wrote that little input file. I too am new to CPMD. I wanted something simple that a couple coworkers could use just to get a rough idea about what it's suppose to look like. I don't even know that it makes any sense. I've had trouble getting my other jobs to run and haven't had a chance to sort out the problems. I'd use the test cases from http://www.cpmd.org/CPMD/contrib/cpmd-test.tar.Z which requires the username and password you used to download CPMD to test the installation. There they have a whole set of cases that actually work. :) Perhaps I should remove that page...

Hope that helps clear things up.

Mike Sullivan
smambs at nus.edu.sg
http://homepage.mac.com/mbsullivan/home.html


> -----Original Message-----
> From: cpmd-list-admin at cpmd.org 
> [mailto:cpmd-list-admin at cpmd.org] On Behalf Of Walter Penits
> Sent: Wednesday, February 18, 2004 8:17 PM
> To: cpmd-list at cpmd.org
> Cc: Holger Hesske
> Subject: Re: [CPMD-list] test calculation problems...
> 
> 
> try a single run with one node (see below)
> 
> > Sorry, I´m a very beginner.
> me too
> > In addition I´ve added the test-input file (nacl.in) and my output
> > (nacl.out), maybe there is a point....
> 
> looks like they are from the sullivan website ?
> http://homepage.mac.com/mbsullivan/prog/cpmdtut.html
> 
> but there is a curious(?) remark on this site:
> 
> "Note: Things seem a bit flakey on the regatta. It should 
> work with 1 processor otherwise it will give this error message: 
> 
> PROGRAM STOPS IN SUBROUTINE BUILDSHELL| BAD NUMBER OF 
> ROTATIONS [PROC=   0] "
> 
> For our simple linux cluster it´s the same (single okay, 
> parallel no). 
> Maybe the input file is the point.
> 
> 
> walter
> 
> ------------------------------------------------------------ 
> Walter Penits 
> IT-Management Computational Physics Group 
> Informatik Lehre 
> Institute for Experimental Physics 
> University of Vienna 
> Walter.Penits at univie.ac.at 
> "to many chieftains, not enough indians" 
> _______________________________________________
> CPMD-list mailing list
> CPMD-list at cpmd.org
> http://www.cpmd.org/mailman/listinfo/cpmd-list
>