home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: sci.math
- Path: sparky!uunet!newsgate.watson.ibm.com!yktnews!admin!yktnews!victor
- From: victor@watson.ibm.com (Victor Miller)
- Subject: Re: Primes
- Sender: news@watson.ibm.com (NNTP News Poster)
- Message-ID: <VICTOR.92Aug17102144@terse4.watson.ibm.com>
- In-Reply-To: bs@faron.mitre.org's message of Mon, 17 Aug 1992 11:02:34 GMT
- Date: Mon, 17 Aug 1992 14:21:44 GMT
- Reply-To: victor@watson.ibm.com
- Disclaimer: This posting represents the poster's views, not necessarily those of IBM
- References: <1992Aug17.160252.10145@waikato.ac.nz> <1992Aug17.110234.13501@linus.mitre.org>
- Nntp-Posting-Host: terse4.watson.ibm.com
- Organization: IBM, T.J. Watson Research Center
- Lines: 55
-
- >>>>> On Mon, 17 Aug 1992 11:02:34 GMT, bs@faron.mitre.org (Robert D. Silverman) said:
- Bob> Nntp-Posting-Host: faron.mitre.org
-
- Bob> In article <1992Aug17.160252.10145@waikato.ac.nz> bill@waikato.ac.nz writes:
- Bob> :How many primes are there less than 2 to the power of p?
- Bob> :
- Bob> 2^p/(p log 2) [approximately]
-
- Bob> :Is there a way of calculating this, or does anyone know of where I can get my
- Bob>
- Bob> How big is p? If p isn't too large it can be calculated exactly using
- Bob> methods developed by Lagarias, Odlyzko, and Coppersmith. These are
-
- Make that Lagarias, Odlyzko and Miller (not Coppersmith!).
-
- Bob> improvements on the Meisel-Lehmer method.
-
- Bob> :hands on a list of primes so I can find out how many primes there are
- Bob> :less than 2^16 or 2^32 for example?
- Bob>
- Bob> For p THIS small you can do it directly by sieving. 2^16 will take
- Bob> milliseconds. How long 2^32 will take is memory dependent, but should
- Bob> only be a few hours at most.
-
- Bob> :Is there some way of determining (or estimating) the expected magnitude of the
- Bob> :nth prime? Or of the growth of the difference between the nth and (n+1)th
- Bob>
- Bob> the n'th prime is approximately n log n.
-
- Bob> the growth of the difference is not well understood. studying it involves
- Bob> some deep methods in analytic number theory. For example, if one has
- Bob> a 'random' sequence of integers going to infinity, whose normal order
- Bob> difference is log n, then one has lim sup (n-->infinity) p_n/log n = 2,
- Bob> whereas for primes, the lim sup is infinite. Further, Rankin has shown
- Bob> that the difference is
-
- Bob> C log n loglog n loglogloglog n
- Bob> ---------------------------------
-
- Bob> logloglog n ^2
-
- Bob> infinitely often.
-
- Bob> I suggest you start by reading a book on number theory. Start with
- Bob> Hardy & Wright.
- Bob> --
- Bob> Bob Silverman
- Bob> These are my opinions and not MITRE's.
- Bob> Mitre Corporation, Bedford, MA 01730
- Bob> "You can lead a horse's ass to knowledge, but you can't make him think"
- --
- Victor S. Miller
- Vnet and Bitnet: VICTOR at WATSON
- Internet: victor@watson.ibm.com
- IBM, TJ Watson Research Center
-