EX-13 SQS within USPEX - Seed Species Counts Overriding Main Input Composition - Composition Control Issue

[נדב מואב‎]

Hi everyone,

I'm working on the EX-13 exercise using the Special Quasirandom Structures (SQS) method within the USPEX framework. I'm encountering a critical issue with composition control that I'm hoping someone can help me understand.

Specifically, I'm finding that the species counts specified in the "seed" file are consistently overriding the species counts specified in the main USPEX input file (e.g., INPUT), directly impacting the occupation of cation sites.

I'm generating SQS structures for the alloy system in EX-13, where two elements compete for the same cation sites. My problem is that regardless of what I specify for the number of species in the main USPEX input file, the distribution of the two elements is determined solely by the numbers present in the "seed" file.

This is causing significant issues for me to move forward and understand how this method works via USPEX. Even when the main USPEX input file (e.g., INPUT) specifies different species counts that conflict with those in the "seed" file, the final generated structure reflects the "seed" counts.

Therefore, my primary questions are:

1. Why are the species counts in the "seed" file overriding the species counts in the main USPEX input file (e.g., INPUT) for EX-13?
2. Is this the intended behavior within USPEX, or is there a bug or misconfiguration in my workflow?
3. How can I regain control over the composition of the cation sites using the main USPEX input file (e.g., INPUT), as expected?
4. How do I correctly implement calculations with relaxation, and what are the steps involved?

I'm concerned that this issue is preventing me from accurately modeling the alloy system in EX-13. Any insights or suggestions would be greatly appreciated.

Thanks in advance.