... Off list, Edward replied to me, saying that ... I think that he and I now agree that pi(x) ~ x/(gap(x) - 1) is his empirical form of the PNT result pi(x) ~...
... Footnote: I have now learned that Minnesotans are the most civilized. David Underbakke does a quadruple-sieve over {k*2^(n-1)-1, k*2^n-1, k*2^n+1,...
... Here's a couplet for an older, weaker limit on prime gaps: Euclid has shown us: a prime must accrue 'twixt any prime p and p-hash plus two. David...
... First of all we had to carry out the parallelisation of our sieving program using MPI. Then we could run our software on a supercomputer (Aster) SGI Altix...
... By the way having to defend my nostrum has been extremely useful to me and I owe you much thanks for sticking with this << I am happy to conclude on that...
... The third column of the table at http://www.primegrid.com/orig/top_prime.php indicates healthy initial progress: currently about 50 primes found at 333...
[Reposted to Yahoo, after a suitable pause, since my first attempt seemed to be refused. Apologies if something like this appears twice.] ... I believe so....
... Congratulations. http://hjem.get2net.dk/jka/math/aprecords.htm is updated. This combined AP21/AP22 record ended a 4-day period (the first) where no AP was...
(statement also posted in the twinprimeforum - sieving discussion) The last few weeks the project TPS was critisized because we are not searching for bitwins,...
... Thanks, Reto, for the information. It might be interesting to hear how David Underbakke's TwinGen http://primes.utm.edu/bios/page.php?id=440 managed to...
... I understand the reluctance for Reto Keiser to use a combination sieve. It can be time consuming. NewPGen is a fine sieve, but I do not know its ...
... 486T and ... million ... David U: if a hypothetic pfgw twin-sophie prime search were to use your quad sieve at n=333,335: 1.) how wide would the sieve need...
... Thank you David for those, and other, details. Going up to 333 kbits would probably require a range of k that is (333/172)^2 = 3.7 times wider, say k up to...
... John Blazek sent me a helpful message, off list, letting me know -- inter alia -- that one should divide the 333 kbit tally in this table by a ... David...
... Grand news! That makes the 53-bit limit on k for gwnums in LLR look as irrelevant as I expected it to be, for a quadruple sieve, even at 500 kbits, and...
I just got 2036800114407689 + 66*41#*n, n=0..19 This improves AP20 record: 260.307P -> 383.566P For n=-2,-1 the formula gives negative primes. Jarek...
I just got the following sequence 1925228725347080393 + 47#*n, n=0..20 with 20-digit last term: 14223024377116908593. I was aiming at AP20 and was surprised to...
... I do find your messages exciting, Jarek. Please continiue updating us like this. [... fumbles for nonsensical, ungrammatical, Polish phrase ...] Gratulacj...
Having problems with message search?
Fill out this form to ensure your group is one of the first to be migrated to the new message search system.
