home *** CD-ROM | disk | FTP | other *** search
- Path: sparky!uunet!mcsun!uknet!comlab.ox.ac.uk!pcl
- From: pcl@ox.ac.uk (Paul C Leyland)
- Newsgroups: sci.math
- Subject: Re: Distribution of primes mod 4
- Message-ID: <PCL.93Jan21100948@rhodium.ox.ac.uk>
- Date: 21 Jan 93 10:09:48 GMT
- References: <winer.727385758@husc.harvard.edu>
- Organization: Oxford University Computing Services, 13 Banbury Rd Oxford OX2
- 6NN
- Lines: 50
- In-reply-to: winer@husc10.harvard.edu's message of 18 Jan 93 19:35:58 GMT
-
- In article <winer.727385758@husc.harvard.edu> winer@husc10.harvard.edu (Adam Winer) writes:
-
- If you keep a running total of the number of prime congruent to 1 mod 4
- minus the number of primes congruent to 3 mod 4 (less than a given number,
- of course), this is _almost_ always non-positive. The first (and last)
- example I know of is 26861. Does anyone out there know of a second
- example? (and no, I don't mean 26862 :-)
-
- There were lots of other questions. For instance:
-
- Adam Winer, WINER@HUSC.HARVARD.EDU :
-
- 1) what the second number is.
- 2) if not, what can I use to get an upper bound for a search?
-
-
- Hans Riesel's book _Prime Numbers and Computer Methods for
- Factorization_, Pub. Birkhauser (1985) has a section on this topic
- (pp.79-84). The following information is lifted from this book.
-
- Please note: Riesel has the subtraction the other way round from
- Winer. I follow Riesel's sign convention, purely to help reduce
- transcription errors.
-
-
- Firstly, the second block of negative values starts at 616841. There
- are several short dips into the negative region, the deepest of which
- lies between 623437 and 623803, where the difference falls as low as
- -8 for 623681. The last value in this region is 641639. The third
- region is around 12366589 where the lowest value occurs (-24). The
- difference remains negative over the entire range 12361933 - 12377279.
- There are other long bands of negative values beginning around
- 9.5185e8; 6.34e9 and 1.854e10
-
- These results by Bays and Hudson, Math. Comp. _32_ (1978), pp. 281-286
-
- Allan Adler (ara@altdorf.ai.mit.edu) asks:
-
- What is the proof that it changes sign infinitely often?
-
- See: JE Littlewood, Comptes Rendu _158_ (1914) pp1869-1872
-
-
-
- If anyone is interested in the corresponding question about primes of
- the form $6n \pm 1$, see Bays and Hudson, Math. Comp. _32_ (1978),
- pp. 571-576
-
-
- Paul
- --
- Paul Leyland <pcl@oxford.ac.uk> | Hanging on in quiet desperation is
- Oxford University Computing Service | the English way.
- 13 Banbury Road, Oxford, OX2 6NN, UK | The time is come, the song is over.
- Tel: +44-865-273200 Fax: +44-865-273275 | Thought I'd something more to say.
- Finger pcl@black.ox.ac.uk for PGP key |
-