Prime finding rate, Sierp vs. Riesel? - Page 3

CGKIII · 2012-09-12, 19:08

What sort of project-level data are available? The stats pages are fantastic, but is there any way to get at the underlying data? (Receiving a snapshot of the current database would be sufficient, real-time updates aren't necessary).

My main interest is basically...
Does the following exist?
For each base b, for each k < (conjectured k), the min(n) for which k*b^n (+/-) 1 is prime?

Heck, even graphing b vs. conjectured k would be interesting, but I'd rather not have to manually type all the numbers into a spreadsheet.

I think that would be really interesting to look at / toy around with. Additionally, it could lend itself to a (thinking very long term) double-check effort, since it seems like on multiple occasions, manual efforts (even pretty well-run ones like this) can have some errors, due to all sorts of things.

If this doesn't exist, is there any way to start tracking / maintaining it, and consolidating what we can piece together?

rogue · 2012-09-12, 19:44

The attached is freely available, but I don't recall where. Only a couple of people were responsible for collecting the list and they used a program written by one of them (source available in the forum somewhere). It is possible that some of the conjectures with conjectured k > 2^32 have lower conjectured k due to the potential of overflows in the program that computed the conjectured k.

There are two possible double-checks that I can think of. The first is to modify the program mentioned above to avoid overflows (or write something different). The second is to verify that all reported primes for a conjecture are truly prime.

henryzz · 2012-09-12, 20:33

Quote:

Originally Posted by rogue

The attached is freely available, but I don't recall where. Only a couple of people were responsible for collecting the list and they used a program written by one of them (source available in the forum somewhere). It is possible that some of the conjectures with conjectured k > 2^32 have lower conjectured k due to the potential of overflows in the program that computed the conjectured k.

There are two possible double-checks that I can think of. The first is to modify the program mentioned above to avoid overflows (or write something different). The second is to verify that all reported primes for a conjecture are truly prime.

At some point it would be nice to create a database that can hold all the data required to prove each base.

rogue · 2012-09-12, 21:18

Quote:

Originally Posted by henryzz

At some point it would be nice to create a database that can hold all the data required to prove each base.

That is many terabytes of data, possibly as much as a petabyte.

henryzz · 2012-09-12, 21:39

Quote:

Originally Posted by rogue

That is many terabytes of data, possibly as much as a petabyte.

Eventually. We have yet to generate a lot of it yet. By the time we do storing it will be possible.

rogue · 2012-09-12, 23:16

Quote:

Originally Posted by henryzz

Eventually. We have yet to generate a lot of it yet. By the time we do storing it will be possible.

I see little benefit in storing most of it as much of it can be easily regenerated.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
Bases 2 & 4 reservations/statuses/primes	Jean Penné	Conjectures 'R Us	466	2021-07-25 04:05
Mersenne prime discovery rate	davieddy	PrimeNet	0	2012-11-26 03:26
Riesel and Sierp numbers bases <= 1024	R. Gerbicz	Conjectures 'R Us	22	2009-12-29 20:21
Sieving Riesel & Sierp base 16	gd_barnes	Conjectures 'R Us	13	2009-12-14 09:23
Riesel/Sierp #'s for bases 3, 7, and 15	Siemelink	Conjectures 'R Us	105	2009-09-04 06:40

2012-09-12, 19:08	#23
CGKIII Aug 2012 31₈ Posts	What sort of project-level data are available? The stats pages are fantastic, but is there any way to get at the underlying data? (Receiving a snapshot of the current database would be sufficient, real-time updates aren't necessary). My main interest is basically... Does the following exist? For each base b, for each k < (conjectured k), the min(n) for which k*b^n (+/-) 1 is prime? Heck, even graphing b vs. conjectured k would be interesting, but I'd rather not have to manually type all the numbers into a spreadsheet. I think that would be really interesting to look at / toy around with. Additionally, it could lend itself to a (thinking very long term) double-check effort, since it seems like on multiple occasions, manual efforts (even pretty well-run ones like this) can have some errors, due to all sorts of things. If this doesn't exist, is there any way to start tracking / maintaining it, and consolidating what we can piece together?