2013 MCSQS paper, example

Alloy Theoretic Automated Toolkit (ATAT)
1

As a starter to MCSQS I have attempted to try the example the in the 2013 mcsqs paper titled "Efficient stochastic generation of special quasirandom structures, the one used for reference purposes.

I followed all the steps mentioned and when the time came to run mcsqs -n 60, I ran into a few questions:

  • How long did it take for the results in the paper to be determined? My objective function becomes -3.2 and then code stays in the MC loop, continuously swapping atoms (with presumably the same value)
    -I noticed at any given point (I ran for upwards of 1 day), the bestcorr file was not complete, specifically something that concerned me was the multiplicity column, which I was getting 0 for all entries. What would this imply?
2

The code does not need to stop in order for a good answer to be found. It keeps searching and you can stop it at any time. The bestsqs.out and bestcorr.out will contain the best answer so far.

bestcorr.out does not contain multiplicities! From mcsqs -h
bestcorr.out
For each of the clusters selected,
its number of pointd (1st column)
its diameter (2nd column)
the correlations of the best SQS found so far (3rd column)
along with the target disordered state correlation (4th column)
and the difference between the two (5th column).