Getting others to do the work on exponents I like (was: Trial Factoring Progress)

[tuckerkao]

Originally Posted by LaurV

That is actually true. Which proves the fact that even a stopped clock shows the time with a very high accuracy, two times per day.

Maybe my day was super long because the second time had barely arrived - M168830323


M168,*3*,*23 survivors at the moment:

005, 074
128, 131, 168, 191, 195
233, 236, 275
332, 363
476, 483
501, 527, 593
609, 612, 644, 666, 681, 693, 699
702, 719, 758, 765, 768, 785
813, 825, 830, 846
906, 908, 911, 921, 941, 948, 971, 989

The quadrupler that I'll specifically working on: M168931123, M168932123, M168934123, M168937123

[tuckerkao]

Originally Posted by drkirkby

IF the aim of SRBase was to find large Mersenne primes, then they are clearly going about it the wrong way. It would be far more sensible to trial factor a bit ahead of the wavefront, not 100s' of millions ahead, which will probably not need a primality test for many tens or perhaps even >100 years.

I don't think SRBase will run P-1 factoring across the large range of exponents, the several Nvidia Geforce 3090 will run TF ultra fast, but P-1 requires AMD Threadripper 5970X or at least 5960X to advance that rapidly.

The trial factoring bits are mostly up to 2^76 in the M116M range, but P-1 are missing - https://www.mersenne.org/report_expo...6844000&full=1

UncWilly previously mentioned about a model that hinted a Mersenne Prime may be hidden within M116M or M117M ranges.

[drkirkby]

Originally Posted by tuckerkao

I don't think SRBase will run P-1 f actoring across the large range of exponents, the several Nvidia Geforce 3090 will run TF ultra fast, but P-1 requires AMD Threadripper 5970X or at least 5960X to advance that rapidly.

The trial factoring bits are mostly up to 2^76 in the M116M range, but P-1 are missing - https://www.mersenne.org/report_expo...6844000&full=1

UncWilly previously mentioned about a model that hinted a Mersenne Prime may be hidden within M116M or M117M ranges.

The statistics at this page
https://www.mersenne.org/report_top_500/
would indicate that SRBase, (along with a 14 others in the top 500), do 100% trial factoring.

The majority (perhaps 80%???) of exponents below 105.2 million have already had the P-1 done. As exponents get larger, less and less have had any P-1 factoring.

[tuckerkao]

Originally Posted by chalsall

Unless other arrangements have been made...

Do please try to keep up.

SRBase 2^73 to 2^74 (My Prediction of the Schedule)

M135M to M139M: 07/24/2021 - 08/04/2021
M140M to M144M: 08/03/2021 - 08/14/2021
M145M to M149M: 08/13/2021 - 08/24/2021
M150M to M154M: 08/23/2021 - 09/03/2021
M155M to M159M: 09/02/2021 - 09/13/2021
M160M to M164M: 09/12/2021 - 09/23/2021
M165M to M167M and M169M: 09/22/2021 - 10/02/2021
M170M to M174M: 10/01/2021 - 10/11/2021
M175M to M179M: 10/10/2021 - 10/20/2021
M180M to M184M: 10/19/2021 - 10/29/2021
M185M to M189M: 10/28/2021 - 11/07/2021
M190M to M194M: 11/06/2021 - 11/16/2021
M195M to M199M: 11/15/2021 - 11/25/2021
M200M to M204M: 11/24/2021 - 12/03/2021
M205M to M209M: 12/02/2021 - 12/11/2021

[tuckerkao]

Originally Posted by drkirkby

If you are concerned, why not go back and check a couple of exponents with known factors, and ensure you find the factors you know exist. Would that not give you confidence in your hardware, without redoing a lot of work?

I must have had a brain fart. You have done exactly what I suggested!!

It has been a while I was wondering why no factors were found between 2^75 to 2^78 for M168,***,*23, finally dug out 1 after 200+ attempts.

M168753223 -> Factor found between 2^75 to 2^76

My confidence towards my hardware had been restored now. Then after 19 more minutes, another factor was found through P-1 of the same type of exponent: M168926123

[mathwiz]

Dormammu tuckerkao, I've come to bargain!

Here's my deal: post here (or in the other thread, as Uncwilly allows), FIVE guesses from your sacred 168M "dozenal" range. I'll run them on an A100 GPU instance, should only take a few days.

If ANY of them is a hit, I'll split the prize money with you.

If NONE is a hit, you agree to not post to any threads on this forum for 1 month.

[Dr Sardonicus]

Originally Posted by mathwiz

<snip>
If NONE is a hit, you agree to not post to any threads on this forum for 1 month.

A month is insufficient. Make it at least a year.

[kriesel]

A month per missed guess?

[Viliam Furik]

Originally Posted by mathwiz

Dormammu tuckerkao, I've come to bargain!

I love a good reference, nice! Especially the connection of the rest of the post with the movie scene. Even though, Doctor Strange told Dormammu to never return. So maybe you could double down your bets? I personally am a kind of softie, when it comes to banishing, I tend to forgive much more quickly than I'd like, but it's not my say here.

[tuckerkao]

Originally Posted by mathwiz

Dormammu tuckerkao, I've come to bargain!

Here's my deal: post here (or in the other thread, as Uncwilly allows), FIVE guesses from your sacred 168M "dozenal" range. I'll run them on an A100 GPU instance, should only take a few days.

If ANY of them is a hit, I'll split the prize money with you.

If NONE is a hit, you agree to not post to any threads on this forum for 1 month.

Maybe a different time when I prepare the different sub-groups, I've gotten new factors 3 days in a row.

I'm currently on the highly composite side on the Mersenne Prime testing ground after nearly no factors for a month.

M168860123 -> F-PM1
M168767023 -> F 2^75 to 2^76

[Uncwilly]

Originally Posted by tuckerkao

I'm currently on the highly composite side on the Mersenne Prime testing ground after nearly no factors for a month.

Remember this post?

Originally Posted by Uncwilly

So, doing 40 and not finding a factor is well within that 1 in 20 average. Do lots of tests 300 or more, then worry about your averages.

Real random things can be quite lumpy.
Patience grasshopper.