COMPLETE!!!! Thinking out loud about getting under 20M unfactored exponents

[petrw1]

Originally Posted by chalsall

It might be amusing to create a new metric for this project. Toy Stories...

"Oh, yeah... 14.0M took 3.7 Toy Story Movie (TSM) equivalents in compute..." (Pulling a number out of my butt... I have no idea how much compute went into rendering that movie... But it might be a cute thing to estimate; a bit of a nod towards Wayne's naming convention for his machines... )

That said it would be interesting to determine how much total compute was done on 14.0.
I'd need access to all P-1, PP1, ECM and TF work done by any of us in the last couple years.
And even if I had it the cyphering itself would be quite onerous.

[petrw1]

Originally Posted by masser

Thanks for the offer, but VBCurtis and I have it covered, I think. It will give our modest GPUs something to work on while we hammer on the range with P-1. Sorting by B1 or B2 will be a good feature as we approach the more difficult ranges.

I trust you have my post from a few weeks ago suggesting how to get a sorted bound report and how to calculate the appropriate bounds?

Enjoy
Wayne

[petrw1]

Originally Posted by petrw1

That said it would be interesting to determine how much total compute was done on 14.0.
I'd need access to all P-1, PP1, ECM and TF work done by any of us in the last couple years.
And even if I had it the cyphering itself would be quite onerous.

What I do have is my TF estimate from 2017/07/24 (Day 0):
Doing TF only would take:
10 more bit levels to 77 bits.
Almost 4.4M GhzDays
That would be over 2.5 years with a 2080Ti GPU at 4,600 GhzDays/Day.

This was probably optimistic because it used the formula from the GIMPS Math page that said

the chance of finding a factor between 2X and 2X+1 (bits) is about 1/X

I'm guessing that is based on NO ECM or P-1 done prior.
I am seeing factor rates much lower than that; 1/85 with moderate P-1 done and 1/100 with aggressive P-1 done.

[lycorn]

I have started some factoring work in the 12M range.
To avoid toe stepping, here is what I´m doing:

TF: 12.9M to 73 bits. Will take at least 3 months to complete. If/when I stop/finish I´ll let you know.
P-1: 12.0M. I am still experiencing with the bounds to use, so for the moment I don´t have any estimate for the time to finish. Will keep you posted.

[lycorn]

Originally Posted by lycorn

I am still experiencing with (...)

LOL, I meant "experimenting on"!

[petrw1]

Originally Posted by lycorn

I have started some factoring work in the 12M range.
To avoid toe stepping, here is what I´m doing:

TF: 12.9M to 73 bits. Will take at least 3 months to complete. If/when I stop/finish I´ll let you know.
P-1: 12.0M. I am still experiencing with the bounds to use, so for the moment I don´t have any estimate for the time to finish. Will keep you posted.

Experimenting and experiencing are both fun

[petrw1]

Originally Posted by lycorn

TF: 12.9M to 73 bits. Will take at least 3 months to complete.

Looking here someone has 12.7, 12.8 and 12.9 all owned to 73 bits.

Unless you reserved all of 12.9 for yourself already it is probably Anonymous.
And if so he will have all 3 ranges at 73 bits in about a month.

[firejuggler]

Have we considered PP1 as a viable alternative to pm1 for low expo?

[lycorn]

Originally Posted by petrw1

Looking here someone has 12.7, 12.8 and 12.9 all owned to 73 bits.

Unless you reserved all of 12.9 for yourself already it is probably Anonymous.
And if so he will have all 3 ranges at 73 bits in about a month.

It´s probably chalsall that assigned 12.9 to me, I don´t know. There´s no indication in Primenet that the range is assigned.

[SethTro]

I'm working in the 6.9M range doing P-1 (B1=4M, B2=110M). I have roughly a 2.5% chance of finding a factor per exp.

I got super lucky today and found 4 factors from my last 22 tests!

https://www.mersenne.org/report_expo...6975583&full=1
https://www.mersenne.org/report_expo...6908203&full=1
https://www.mersenne.org/report_expo...6907297&full=1
https://www.mersenne.org/report_expo...6906943&full=1

I should find the remaining 11 factors over the next ~9 days.

[chalsall]

Originally Posted by lycorn

It´s probably chalsall that assigned 12.9 to me, I don´t know. There´s no indication in Primenet that the range is assigned.

Super busy... But...

12.9M et al was taken by our mysterious Anon, reserved through GPU72.

Please note GPU72 does not reserve from Primenet for the <2K sub-project. That was the reason for the nomenclature change from "Owned" to "Curated" on this report.

12.0M or 12.5M might be of interest to you to work... Please let us know, and either reserve it through the GPU72 interface, or I can "release" it. Your choice.

P.S. Sorry for the high bandwidth. I'm juggling a few things...