Constant n Search

[kar_bon]

Constant n Searches

This thread is for collecting all Constant n Searches.

The most popular kind of such search was/is n=333333 at PrimeGrid (also n=195000).

Please everyone who has some ranges to report for constant n, post here.
Post the primes found (not only Top5000) and the k-range searched.

The page can be found under the 'Data Section' at the bottom.

Notes:

- The k-ranges are mostly not for sure (like for P.Kenny's n's): i got those information only from the Top5000 pages.

- The n's and k-ranges from L.Zhou are first-hand data.

- All primes are listed in the 'normal' data pages when there exist a Top5000 link (or note).

So post your k-ranges for any n and i'll include and reserve them.

[henryzz]

does anyone if there is such a thing as a low or high weight n

[mdettweiler]

Originally Posted by henryzz

does anyone if there is such a thing as a low or high weight n

Well, though I don't know whether it would officially be termed "high weight" or "low weight", I *have* noticed that when scrolling through the large "master sieve files" that we use over at NPLB, that some n definitely seem to have a lot more k's listed for them than others--i.e., they look "heavier". For example:

405 300001 
427 300001 
531 300001 
541 300001 
619 300001 
769 300001 
789 300001 
895 300001 
951 300001 
407 300002 
435 300002 
437 300002 
537 300002 
551 300002 
617 300002 
633 300002 
635 300002 
665 300002 
677 300002 
857 300002 
887 300002 
987 300002 
451 300003 
459 300003 
565 300003 
585 300003 
609 300003 
613 300003 
679 300003 
709 300003 
735 300003 
781 300003 
849 300003 
913 300003 
539 300004 
549 300004 
657 300004 
825 300004 
857 300004 
885 300004 
927 300004 
939 300004 
469 300005 
511 300005 
525 300005 
547 300005 
657 300005 
697 300005 
715 300005 
799 300005 
925 300005 
937 300005 
425 300006 
437 300006 
527 300006 
537 300006 
635 300006 
725 300006 
783 300006 
821 300006 
921 300006 
945 300006 
421 300007 
465 300007 
489 300007 
559 300007 
603 300007 
619 300007 
625 300007 
645 300007 
715 300007 
741 300007 
813 300007 
919 300007 
939 300007 
951 300007 
961 300007 
975 300007

In this particular selection, n=300007 is the heaviest, whereas n=300004 is the lightest weight. (Feel free to correct me if I'm wrong on any of those--I came up with those based on quick mental/visual calculations.)

Edit: I've added colors to the table. This can be useful in helping the brain to "pick out" which ones look heavier and lighter weight more easily.

[Mini-Geek]

I'd be surprised if the apparent weight of n's is due to anything other than random chance giving some more than other (but I have noticed this apparent "weight" in my own NPLB work, especially when it unbalances my cores due to one file having more "high weight" n's than the other). Weighted k's, on the other hand, actually make some sense.

[Svenie25]

Karsten, as you read in another thread, I´m doing a constant n search at n=666777, with k=2-1000M(1G), trying to find twinprime. Atm the sieving is at 110T.

kar_bon: reservation is online

[Svenie25]

Karsten, sorry for the work you have with me. But please unreserve my reservation for n=666777. There is so much to do on other places.