holy tethered cow! new Mersenne prime? (M43-related)

[gbvalor]

Can we know what is the status of the two running verifications ?
What percent they have reached ?

T. Rex's verifcation already surpased the 50% line

Guillermo

[tha]

I was trying to find the encoded residue, but as I was adding bits to one side of my 14" screen they fell of at the other side. So I gave up.

[Mystwalker]

Originally Posted by ewmayer

But George has already re-run the last 5000 iterations from the user's last savefile (which is practically impossible to fake so as to indicate primality for an exponent not in fact yielding a prime)

I'd think that it is not that difficult to fake the savefile. After all, we already had the "start from the beginning forward and from the end backwards and see whether they meet in the middle" discussion.

It wouldn't be time-efficient and such, but I don't see a real problem in generating a residue that will be 0x0 in 5000 iterations for a particular exponent.

It's only highly unlikely that a simple computer error is responsible.

[ColdFury]

I'd think that it is not that difficult to fake the savefile. After all, we already had the "start from the beginning forward and from the end backwards and see whether they meet in the middle" discussion.

I don't see how anyone would be able to generate such a savefile in any reasonible amount of time. You'd have to examine both square roots in each iteration and unless you had some clever way to cut off branches of the computation (?) the tree would grow exponentially. Maybe you could do it for a couple iterations, but I would think 5000 would be out of the question. Of course, maybe there's some tricks I don't know about which would make it feasible.

[Wacky]

Originally Posted by ColdFury

You'd have to examine both square roots in each iteration and unless you had some clever way to cut off branches of the computation (?) the tree would grow exponentially.

Is this correct? In order to "meet in the middle" you would need to do something of that sort because you do not know which is the meeting point. However, there should be many possible (pseudo-saved) starting points that will lead to "0" after the correct number of iterations. If we are only making it appear that we have a correct save point, we don't care which one we choose.

Therefore, although the inverse iteration is not unique, you need take only one of the possible values at each step.

[ColdFury]

Originally Posted by Wacky

Is this correct? In order to "meet in the middle" you would need to do something of that sort because you do not know which is the meeting point. However, there should be many possible (pseudo-saved) starting points that will lead to "0" after the correct number of iterations. If we are only making it appear that we have a correct save point, we don't care which one we choose.

Therefore, although the inverse iteration is not unique, you need take only one of the possible values at each step.

You're right, one can arbitrarily choose one of the results of the square root since the starting point does not matter.

[philmoore]

But isn't just one square-root computation of the same overall complexity as a complete LL test?

[Lumo]

Originally Posted by georgekh

OMG SOMEONE JUST POST SOMETHING RELEVANT TO THE PROBABLE NUMBER!!!!!!

Have you tried internet search? What about a Google search for a Calabi-Yau fibered Mersenne number?

[F960897]

Hi guys,
new to the forums. Do not have much time to spend with prime numbers any more but still follow the news (used to be obsessed). Here is my encrypted guess:

0x36C9248137FFFE

Those who know how the false residues are calculated may figure out my method for encryption.

Regards,
Mike Eaton

[georgekh]

I meant about the new Merseene number like everyone is talking about encryption this and encryption that but none of it is a help for me and i'm completely lost about what pacionet and jinydu, t.rex and moo are talking about

[jinydu]

Originally Posted by georgekh

I meant about the new Merseene number like everyone is talking about encryption this and encryption that but none of it is a help for me and i'm completely lost about what pacionet and jinydu, t.rex and moo are talking about

Well, basically, some people think they have managed to guess what the exponent is and would like to be acknowledged for having guessed it before it was announced. However, they don't want to spoil it for everyone else by simply posting their guess (and because of several other reasons listed in this thread). So instead, they post an encrypted version of their guess; that is, they use some kind of code (that they keep secret) to scramble their guess, then post it on this thread. After the verification is complete and the exponent is officially reported, the guessers reveal their code. Then, everyone can apply the code to see that they really did guess correctly.