[Wien] Number of memories in MSR1a for big calculations

[Laurence Marks]

Dear All,

A small piece of information posted for general use. I am now fairly
convinced that for large calculations (e.g. 80 unique, 350 total) in
MSR1a it is better to increase the number of memories that MSR1a uses,
e.g. use in case.inm

MSR1a  0.0   YES  (BROYD/PRATT, extra charge (+1 for additional e), norm)
0.20            mixing FACTOR for BROYD/PRATT scheme
1.00  1.00      PW and CLM-scaling factors
9999  12        idum, HISTORY

For small calculations 8 is good, with 40-50 unique perhaps 10
memories (not 12) is better; I doubt that increasing this much more is
useful unless you are doing vast calculations (200 unique) which is
probably not viable anyway.

N.B., I cannot prove this is optimal as this would require running the
same calculation many times and I don't have the computer resources to
do this.

-- 
Professor Laurence Marks
Department of Materials Science and Engineering
Northwestern University
www.numis.northwestern.edu 1-847-491-3996
"Research is to see what everybody else has seen, and to think what
nobody else has thought"
Albert Szent-Gyorgi