home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!cis.ohio-state.edu!magnus.acs.ohio-state.edu!slc3.ins.cwru.edu!agate!linus!linus.mitre.org!faron!bs
- From: bs@faron.mitre.org (Robert D. Silverman)
- Subject: Re: Primes
- Message-ID: <1992Aug17.110234.13501@linus.mitre.org>
- Sender: news@linus.mitre.org (News Service)
- Nntp-Posting-Host: faron.mitre.org
- Organization: Research Computer Facility, MITRE Corporation, Bedford, MA
- References: <1992Aug17.160252.10145@waikato.ac.nz>
- Date: Mon, 17 Aug 1992 11:02:34 GMT
- Lines: 44
-
- In article <1992Aug17.160252.10145@waikato.ac.nz> bill@waikato.ac.nz writes:
- :How many primes are there less than 2 to the power of p?
- :
- 2^p/(p log 2) [approximately]
-
- :Is there a way of calculating this, or does anyone know of where I can get my
-
- How big is p? If p isn't too large it can be calculated exactly using
- methods developed by Lagarias, Odlyzko, and Coppersmith. These are
- improvements on the Meisel-Lehmer method.
-
- :hands on a list of primes so I can find out how many primes there are
- :less than 2^16 or 2^32 for example?
-
- For p THIS small you can do it directly by sieving. 2^16 will take
- milliseconds. How long 2^32 will take is memory dependent, but should
- only be a few hours at most.
-
- :Is there some way of determining (or estimating) the expected magnitude of the
- :nth prime? Or of the growth of the difference between the nth and (n+1)th
-
- the n'th prime is approximately n log n.
-
- the growth of the difference is not well understood. studying it involves
- some deep methods in analytic number theory. For example, if one has
- a 'random' sequence of integers going to infinity, whose normal order
- difference is log n, then one has lim sup (n-->infinity) p_n/log n = 2,
- whereas for primes, the lim sup is infinite. Further, Rankin has shown
- that the difference is
-
- C log n loglog n loglogloglog n
- ---------------------------------
-
- logloglog n ^2
-
- infinitely often.
-
- I suggest you start by reading a book on number theory. Start with
- Hardy & Wright.
- --
- Bob Silverman
- These are my opinions and not MITRE's.
- Mitre Corporation, Bedford, MA 01730
- "You can lead a horse's ass to knowledge, but you can't make him think"
-