Re: getting nth prime
- From: "Joseph Ashwood" <ashwood@xxxxxxx>
- Date: Tue, 25 Apr 2006 06:09:48 GMT
"Francois Grieu" <fgrieu@xxxxxxxxxxxx> wrote in message
news:fgrieu-CFE5AA.07244525042006@xxxxxxxxxxxxxxxxxxxxxx
To the question
Is there an algorithm other than Eratosthene Sieve to find the Nth
prime number
In article <8Of3g.69272$H71.28527@xxxxxxxxxxxxxxxxxxxxxxxxxx>,
"Joseph Ashwood" <ashwood@xxxxxxx> wrote:
There are variations of counting and seiving, probably hundreds of
them throughout the years, but it currently appears that there is
no way to find the n-th prime number without knowing the n-1th prime
number (except for directly sievable values). More specifically it
appears to be impossible to resolve n to an integer without knowing
either the n-1th or n+1th prime (e.g if I know 5 is the third prime
I can figure out that 7 is the fourth, if I know 11 is the 5th prime
I can figure out 7 is the fourth).
The above I think can't be reconciled with results on Pi(x), the
prime-counting function. To compute the Nth prime, we can find
x such that Pi(x) is close to N, compute Pi(x) exactly, then move
from here sequentially. Computing Pi(x) exactly is fast, a day
or so is enough to determine that the 2220819602560918841 th prime
is 100000000000000000039.
I see where my error is, if we know the number of primes <= k +/- 1 we can
resolve n (where k is the nth prime), otherwise we have to determine the
count. My main point though was that a count or sieving of some kind is
necessary to perform, even though we have hundred of variations (no
citation, just a guess) throughout the centuries.
Joe
.
- References:
- getting nth prime
- From: Dhairyas
- Re: getting nth prime
- From: Joseph Ashwood
- Re: getting nth prime
- From: Francois Grieu
- getting nth prime
- Prev by Date: Re: JSH: Hyperbolic factoring method
- Next by Date: Re: getting nth prime
- Previous by thread: Re: getting nth prime
- Next by thread: tools for lecture
- Index(es):
Relevant Pages
|