At 6 Aug 2002 13:16:44 +1000, "Sisyphus" <[EMAIL PROTECTED]> wrote: >What's an n-tuplet prime ? >My idea of triplet prime, for example, would be 3 consecutive odd integers >that are prime ...... but it wouldn't take long to find out how many of >*them* there are :-)
n-tuplet primes are a set of n primes "as close together as possible". For triplets, for example, there are two types: {p,p+2,p+6}, and {p,p+4,p+6}. At 6 Aug 2002 08:38:17 -0400, Jason Papadopoulos <[EMAIL PROTECTED]> wrote: >Enumerating the primes to 10^16 means storing something like 10^14 numbers, >doesn't it? If you only store the gaps, wouldn't that imply 10^14 bytes >of storage? If we were actually going to store them, yes. The idea was to work with a small block at a time and only store the "census values" and any exceptional numbers. At 06 Aug 2002 22:39:21 -0400, David Willmore <[EMAIL PROTECTED]> wrote: >I have a little bit of code lying around, uhh, somewhere, that can generate >bitmaps (via a sieve) of arbitrary 8K chunks of primes up to 2^32. It would >be pretty easy to extend it higher. Could these 'statistics' be easily >generated if fed a 'stream' of primes? If so, this project doesn't look all >that hard. Sieving, by itself, isn't going to identify pseudoprimes and strong pseudoprimes. Since I'm going to be performing strong pseudoprimality tests anyway, I might as well use those to construct my list of primes. We already know how many primes there are less than 10^15, so if we count the integers less than 10^15 which pass 15 strong pseudoprime tests and get the same answer, we'll know that all those values were in fact prime. Colin Percival _________________________________________________________________________ Unsubscribe & list info -- http://www.ndatech.com/mersenne/signup.htm Mersenne Prime FAQ -- http://www.tasam.com/~lrwiman/FAQ-mers