Actually,

Type n 2^(n-1)-1 mod n

--------- ---- -

composite 1 => 0

prime 2 => 1

composite 341 => 1

composite 561 => 1

composite 645 => 1

composite 1105 => 1

composite 1387 => 1

composite 1729 => 1

composite 1905 => 1

prime 1093 => 1, fails

Thanks for your help.

--- In

primenumbers@yahoogroups.com, Alan McFarlane

<alan.mcfarlane@...> wrote:

>

> Given a few test values, your function returns:

>

> Type n 2^(n-1)-1 mod n

> --------- ---- -

> composite 1 => 0

> prime 2 => 1

> composite 341 => 0

> composite 561 => 0

> composite 645 => 0

> composite 1105 => 0

> composite 1387 => 0

> composite 1729 => 0

> composite 1905 => 0

>

> Look at

>

> http://www.research.att.com/~njas/sequences/A001567

>

> Extract:

>

> "It is known that all primes p divide 2^(p-1) - 1"

>

>

>

> Paul E. Schippnick wrote:

> >

> >

> > I recently discovered a prime number solution for all prime

numbers. It

> > works. And one of its proofs is where it fails. Yes, where it

fails is a

> > proof that it will work for all other prime numbers.

> >

> > Where it fails: Testing Mersenne Primes. Because of how it proves

a

> > prime, all Mersenne numbers test as if prime. And only the

Mersenne

> > composite numbers. That means non-Mersenne composite numbers will

never

> > test as prime.

> >

> > Here is the test formula: 2n-1 - 1 mod n = 0, then n is prime.

> >

> > Now it has other issues. But it is a binary test for prime.

> > ----------

> >

> > No virus found in this outgoing message.

> > Checked by AVG Free Edition.

> > Version: 7.5.467 / Virus Database: 269.7.6/815 - Release Date:

5/22/07

> > 3:49 PM

> >

> > [Non-text portions of this message have been removed]

> >

> >

>