Lucky 13 (M51 related)

[Gordon]

Originally Posted by chalsall

Many of us do. What are you going to do about it?

(There's a song in there, somewhere....)

Not spoil the surprise for everyone else...

[chalsall]

Originally Posted by Gordon

Not spoil the surprise for everyone else...

Sweet. Cool. Thanks.

A lot of work goes into this kinda thing (which almost no one sees)....

[science_man_88]

Originally Posted by preda

The secondary question is whether we should prioritize the testing of 1mod8 exponents.

well we know if p and 2p+1 are both primes of form 4x+3 then 2p+1 divides Mp so that's a slight skew in the mod 12 version (potentially has an affect on primes of form 11 mod 12 as that's the intersection of primes of form 4x+3 and 6y+5 that are both required, so that 2p+1 could also be prime). not sure of any other rules and their effects.

[chalsall]

Originally Posted by science_man_88

not sure of any other rules and their effects.

So your counsel is?

[axn]

Posts about twin primes moved to miscellaneous math: https://www.mersenneforum.org/showthread.php?t=23910

[aurashift]

Originally Posted by ATH

I do not think we have seen benchmarks from the latest generation(s) of crazy Xeon Platinum processors with 6-channel RAM.

I'll let you know here soon.... I think the biggest I had was a 22 core.



But I'd rather be testing some EPYC (2?) setups.

[Batalov]

There was a suggestion circled around to split all the Conjecture talk in a separate thread.
(And perhaps "the size of the kit" comparisons in the other. )

[chalsall]

Originally Posted by Batalov

[Funny image wasn't quotable...]

For dyslexics, right (or correct) and wrong, left and write et al are a very complex and difficult subject.

It all depends on the frame of reference and the context... We have a lot of trouble with mirrored frames of reference, for example.

"Dyslexics of the world, untie!

[GP2]

Originally Posted by Batalov

There was a suggestion circled around to split all the Conjecture talk in a separate thread.
(And perhaps "the size of the kit" comparisons in the other. )

Yes, that's a good idea.

It's only here because masser gave binary-based histogram bins for Mersenne primes, which led me to post about hexadecimal bins and hexadecimal final digits, and one thing led to another.

[Damian]

Congratulations everyone involved for the new prime!

[axn]

Originally Posted by kriesel

If they all pass, it does not necessarily mean the existing codes are without flaw.
If one fails, it does not necessarily mean the existing codes are flawed, it could be a transient hardware issue.There's a sliver of a chance of an error occurring outside the loop of iterations that are guarded for accuracy by the Gerbicz check.
No disrespect to any of the authors (or users, or hardware designers for that matter; some tasks are just very hard). I've fixed code that failed nonreproducibly at low rates. Ppb or lower frequency bugs can be hard to determine exist, much less identify and resolve. Current assignments can be of order 10¹⁸ core clocks to complete.

The random rotation means that FFT operates on completely different bit patterns for first time vs dc runs. Hence, if a DC run matches with same final residue, this is as good as 100% as we can humanly make.