At 08:55 PM 6/22/99 -0700, Lang Pal wrote:
>I recommend to replace the expression " (3/2)^n " by "
>(PI*SQRT(2))/3". Then
>we receive
>1,480960969 instead of 3/2 and the correlation coefficient would be
>almost
>just the same
>(0,996117397). Why then to replace? I think, there must be some
>relation
>between the Mersenne problem and the partition function (considering
>the
>Hardy-Ramanujan-Rademacher formula). Other factors must be found
>somehow and
>once in the (I hope "next" ) future.
That sounds very interesting. 3/2 happens to be close to the value based on
the current data, but otherwise I don't know of any justification for 3/2.
+----------------------------------------------+
| Jud "program first and think later" McCranie |
+----------------------------------------------+
________________________________________________________________
Unsubscribe & list info -- http://www.scruz.net/~luke/signup.htm
