[seqfan] Re: nonpowers in A001481

[Joerg Arndt]

Me stupid.  Mentally took zero-exponents as ones:

How about rewording the current

 These are the numbers n = 2^i Product_{p == 3 mod 4} p^2j Product_{p
 == 1 mod 4 } p^k where gcd of the nonzero exponents {i, 2j's, k's} is 1.

to

  Numbers of the form 2^i * S^j * T^(2*k) where S is a product of
  primes 4k+1, T a product of primes 4k+3, and i,j,k>=0, and
  gcd(i, j, 2*k)==1.

?

...hoping the latter is correct as well.


* Charles Greathouse <charles.greathouse at case.edu> [Jan 16. 2011 17:42]:
> But for 9,  gcd{i, 2j's, k's} = gcd {2} = 2 ≠ 1, so it shouldn't be a
> member.  For 18, gcd{i, 2j's, k's} = gcd {1, 2} = 1 so it should be a
> member.
> 
> Charles Greathouse
> Analyst/Programmer
> Case Western Reserve University
> 
> [...]