[CPMD-list] PIMD AND OPENMP

A.Alaka aoa101 at york.ac.uk
Sun Aug 8 23:29:16 CEST 2004


The modifications (from the path-int tests) I meant are:

I'm performing the calculation on a bulk system rather than  a surface.
I am using uspps.
I am using PCG MINIMIZE rather than ODIIS.
Im using a TROTTER DIMENSION value of 4 (instead of 8).
I set the DEBROGLIE CENTROID value to 500K.

I am leaning towards the idea that the problem with the path-int 
calculation is to do with setting the states. In the example test-input, 
the STATES and OCCUPATION keywords are defined. I'm not sure how this 
would affect my calculation.

Your help is much appreciated. Thanks all.




More information about the CPMD-list mailing list