Oops - New Prime! (M49 related)

[airsquirrels]

I'm surprised no one has used ETA, card specs timing and the slipped FFT size to try and narrow in. Precision of units might be a problem :)

[aurashift]

"The oops prime"

[Madpoo]

Originally Posted by aurashift

"The oops prime"

I'm curious to see how George treats this discovery's unusual history in the press release.

[NBtarheel_33]

Originally Posted by aurashift

"The oops prime"

Well, this prime and the prime from April 2009, both being "oops" primes, could perhaps form the basis of a new class of primes known as Britney Spears primes.

[Madpoo]

Originally Posted by airsquirrels

I'm surprised no one has used ETA, card specs timing and the slipped FFT size to try and narrow in. Precision of units might be a problem :)

I crunched some data myself because I wondered if it were easily guessable from the known info.

In the 4M FFT size (AVX2 table) there are 60,100 currently unverified exponents or with no known factors. Not only that but over 22K are unassigned, so it's kind of a big pool. Had it not been for the 4M FFT slip, it'd be even more obscure. LOL

And besides, the method(s) used to hide it on the website seem to be just fine. I did uncover one odd quirk that may have tipped someone off if they were *really* diligent, which I'll bring up with George to make sure that's not the case next time, but otherwise, yeah, I think it's doing a good job of keeping it under wraps.

I don't think anyone would stumble on it, but you're a bunch of smart folks and some of you are very persistent. Well, not too much longer before it's public anyway.

[aurashift]

Originally Posted by NBtarheel_33

Well, this prime and the prime from April 2009, both being "oops" primes, could perhaps form the basis of a new class of primes known as Britney Spears primes.

oops.

The second oops prime.

[ewmayer]

Originally Posted by airsquirrels

I'm also verifying with clLucas, which if all goes well should make this the first prime verified on prime95, mlucas, cudalucas and clLucas

I don't spend much time in the GPU fora - are the sources for cudalucas and clLucas independent?

Originally Posted by Prime95

Yes. Ernst let slip that a 4M FFT was required.

Hey, had to give hoi polloi a little hint of something to keep them from rioting.

Anyhow, given some of the roundoff weirdness several of the DC/TC/QC have encountered, this could plausibly be spun as a sub-4M-range exponent, just one with a bad attitude - maybe sour at having sat there all those months, waving its little byte-sized arms frantically at an uncaring world and shouting, "yoohoo!"

[frmky]

Originally Posted by ewmayer

I don't spend much time in the GPU fora - are the sources for cudalucas and clLucas independent?

No. CUDALucas is derived from MacLucasFFTW, and clLucas from CUDALucas. CUDALucas uses the closed-source nVidia cuFFT library, and clLucas uses the open-source clFFT library.

Edit: gpuLucas is an independent implementation that also uses the cuFFT library.

[cuBerBruce]

Originally Posted by Prime95

As to official announcement, this may take some time. I need to write a press release, Scott needs to approve and gather his list of press contacts. Chris Caldwell needs to have his web site updated. GIMPS main page needs updating. Patience please.

And "WR" work type request handling must be changed so as not to give out exponents smaller than the newly discovered prime's exponent. Or does that happen automatically?

[Batalov]

Actually, the same can be said about another WR search effort... Hmmm...
In it, b needs to be > SIKRIT_b_value ( = 211807) to be a contender again.

P.S. Will retroactively edit this post when it will no longer be secret. Done. It is 211807

[Dubslow]

Originally Posted by airsquirrels

I'm surprised no one has used ETA, card specs timing and the slipped FFT size to try and narrow in. Precision of units might be a problem :)

We've tried that before... but the precision is pretty bad, 200K-500K in terms of exponent.