[LongGone]

Originally Posted by Reincarnate

Also some numbers will have more than 1 unique prime factors (e.g. 30 -> unique primes 2, 3, and 5)

Stargroup originally came up with the idea that having a composite being a product of 3 primes isn't efficient and will never be part of the maximum set. I had a thought and will back this up by saying that, in order for a composite to be in the set it has to have at least two terms from group 1 (otherwise it will exceed n). And it is intuitively known to be not efficient to group together two sets of numbers from set 1 (as you remove the chance of it being paired up with a set 2 number, which is more efficient to produce larger composites. So I'll assume that the solution set involves composites with at most two distinct primes (raised to whatever power). So in this case, 30 which has 3 prime factors will not be part of the solution set of Co