[QUOTE=T.Rex;142133]
About the 100M prime number, my guess is that it will be discovered in 2023 (if there is still petrole...).
Tony[/QUOTE]
I thought computers run on coal, nuclear power, natural gas, and renewables, not oil :confused:

Status of F33 known in 2025 !!!!
 

[QUOTE=uigrad;142145]At the rate of one prime a week, it should come much sooner than that. If hardware continues to follow Moore's law, in 2023, the average pc will be 1000 times faster than today.[/QUOTE]Verifying a 10 times bigger Mersenne number requires about 10*15 more times than a 10M number. After 14 years, I guess that PCs will be about 128 times faster than now. And there will be much more exponents to tests... 2023 !!
[QUOTE] My guess is that we have at least another 5-7 years before we know "the smallest 10 million+ digit prime".[/QUOTE]My guess is that we know all Mersenne primes below M46.
And I guess that we will know the status of the 33th Fermat number in [B]2025[/B], after 6 months of computation done on the biggest computer and with a program using my LLT-proof for Fermats (and not the Pépin's test) with a program derived from Prime95 or from Mlucas, using very fast assembler optimisations.
Tony

[QUOTE=Jeff Gilchrist;142189]Sorry, busy day, my verification finished shortly after midnight last night.[/QUOTE]Oh, nuts ! I'm late ! I was so used to be the first... Not easy to not be under the sunlights !
Nice run, Jeff ! The wind is blowing in Ottawa ! [QUOTE]Tony will be finished shortly I'm sure.[/QUOTE]Tomorrow late in France, I think. So Sunday for USA.[QUOTE]Tony: Yes, the system because a little less busy so I was able to speed up a little, but still not much, just enough to stay ahead.[/QUOTE]I don't understand how you did... Can you send me the glucas.out file ?

Have you read my guess about [B]F33 [/B]?

Tony

[QUOTE=T.Rex;142199]I guess that we will know the status of the 33th Fermat number in [B]2025[/B], after 6 months of computation done on the biggest computer and with a program using my LLT-proof for Fermats (and not the Pépin's test) with a program derived from Prime95 or from Mlucas, using very fast assembler optimisations.
Tony[/QUOTE]
Tony, is your LLT-based Fermat test in some way faster to compute than Pépin's test? If so, why not just use Pépin? [Assuming that a small factor has not yet been found by then].

LLT-based test for Fermat numbers
 

[QUOTE=ewmayer;142203]Tony, is your LLT-based Fermat test in some way faster to compute than Pépin's test? If so, why not just use Pépin? [/QUOTE]In fact, Lucas provided some information about such a test, but with no clear proof. My proof is based on Ribenboim and Williams techniques. Not rocket-science techniques... but that works. It is not faster than the Pépin's test. But, since the Prime95 and Mlucas have now assembler code, they should be faster than the very old code (Fortran ?) implementing the Pépin's tests, I think. Who knows where such code is ? Page 495 of version 2 of his book (Prime Numbers, a Computational Perspective), Crandall provides a DWT modulo Fermat numbers. I don't know the work required to adapt the Mersenne IBDWT of Prime95 or Mlucas for Fermat numbers. For the fastest PC available now (Core2 Duo 3GHz), it should take 3000 years now... (it was about 4000 one year before !!)
Maybe we should ask Crandall ?
Tony

Mlucas already has all the generic C code needed for a fast Fermat-mod, and most of the FFT for the LL and Pe'pin tests share the same code, i.e. the SSE2 code I'm writing for the LL-test portion will help both kinds of tests. The nice thing about Fermat mod is that for the most part, the differences between it and Mersenne-mod make the resulting Fermat-mod code *simpler*, e.g. no complex/real-FFT wrapper code and intricate pairwise index correlations needed surrounding the dyadic squaring step between the forward and inverse FFT.

By the time F33 becomes computationally attractive, the code will be ready.

[QUOTE=T.Rex;142023]23th of August: nearly 86% done.
6th of september: nearly 55% done.
[/QUOTE]

[QUOTE=T.Rex;142126]Aug23th : 93.06% done. Never so close !
Sept6th : 67.50% done. Soon.
[/QUOTE]

So the August verification is getting roughly 7% in 20 hours. That means it should finish about 20 hours after the second quote, less than 10 hours from now. Why wait all the way until Sunday US time?

[QUOTE=Jeff Gilchrist;142189]Sorry, busy day, my verification finished shortly after midnight last night. Verified as prime! [/QUOTE]

Congratulations Jeff! :beer:

the new banner is trying to tell us something :shock: :w00t: --

[QUOTE=Batalov;142231]the new banner is trying to tell us something :shock: :w00t: --[/QUOTE]If that banner is correct then it would suggest that Msep > Maug > M44.

[QUOTE=retina;142232]If that banner is correct then it would suggest that Msep > Maug > M44.[/QUOTE]

That would give a strong indication that the September exponent is 43041727.