'where'in with duplicates或'select distinct' - 表现

I would like to return some results basing on subquery results and I was wondering what would be faster - having subquery results be distinct or where in clause having duplicates? I would like to return some results basing on