Newer milestone thread

[Uncwilly]

Originally Posted by Aramis Wyler

Honestly I'm surprised primenet still gives out TF work at all, while gputo72 is keeping up.

What do you do when someone asks only for TF, or if someone has a machine that is only good for LMH-TF?
I think that you give them something to do.

[axn]

Originally Posted by Uncwilly

What do you do when someone asks only for TF, or if someone has a machine that is only good for LMH-TF?
I think that you give them something to do.

Not sure about the first part, but small ECM work type is suitable for the second type.

[Aramis Wyler]

Sure, LMH-TF, ok. I wasn't explicit enough. They were talking about the 81-90M range, not the 332M range. That's a whole unopened space up there.

[petrw1]

Originally Posted by ckdo

March 6, 2013: The trial factoring wavefront hits 89M.
March 25, 2013: The trial factoring wavefront hits 91M.

That's 23 days for 4M.

It hasn't happened yet in the 80 range but it still could. The proceeding 60 and 70 ranges were "re-primed" a few times for TF to take them deeper. I think it required some manual intervention.

[bcp19]

Originally Posted by davieddy

Currently, new LL tests are being dished out at 62M.
We currently expect the next prime discovery when this "wavefront" hits
90M in 4 years time.
ATM we are trying to ensure new LLs are TFed to 74 bits. By 90M, this will be 75 or 76 bits.

HOW MUCH IS TFing to 70 bits contributing to this effort, aside from stealing all the low-lying fruit factor finds?

David

Hmm, let's take a look...

All exponents below 26,083,693 have been tested and double-checked. 
All exponents below 44,250,841 have been tested at least once. 
Countdown to testing all exponents below M(57885161) once: 44,071 
Countdown to proving M(30402457) is the 43rd Mersenne Prime: 6,198 
Countdown to proving M(32582657) is the 44th Mersenne Prime: 43,686 
Countdown to proving M(37156667) is the 45th Mersenne Prime: 134,620 
Countdown to proving M(42643801) is the 46th Mersenne Prime: 248,550 
Countdown to proving M(43112609) is the 47th Mersenne Prime: 258,509

Looks to me like there are some 258 and a fraction thousand exponents that have not been double-checked. While the possibility is remote, there is a chance there is a hidden prime in there somewhere. Look at the 48th known MP... it was assigned to anonymous twice before CC worked his magic and brought it to our attention. You love working the numbers, what's the probability of there being a missed prime in those 258k exponents?

[petrw1]

Originally Posted by petrw1

It hasn't happened yet in the 80 range but it still could. The proceeding 60 and 70 ranges were "re-primed" a few times for TF to take them deeper. I think it required some manual intervention.

Ask and ye shall receive....

http://www.mersenne.org/primenet/

[Uncwilly]

Originally Posted by davieddy

The raw totals at the bottom (which you are plotting) need careful interpretaion to get the DC and LL rates of completion:

Here is a nice visual that illustrates how swell things are going. The little yellow sliver is not even visible.

[davieddy]

Originally Posted by bcp19

Hmm, let's take a look...

All exponents below 26,083,693 have been tested and double-checked. 
All exponents below 44,250,841 have been tested at least once. 
Countdown to testing all exponents below M(57885161) once: 44,071 
Countdown to proving M(30402457) is the 43rd Mersenne Prime: 6,198 
Countdown to proving M(32582657) is the 44th Mersenne Prime: 43,686 
Countdown to proving M(37156667) is the 45th Mersenne Prime: 134,620 
Countdown to proving M(42643801) is the 46th Mersenne Prime: 248,550 
Countdown to proving M(43112609) is the 47th Mersenne Prime: 258,509

Looks to me like there are some 258 and a fraction thousand exponents that have not been double-checked. While the possibility is remote, there is a chance there is a hidden prime in there somewhere. Look at the 48th known MP... it was assigned to anonymous twice before CC worked his magic and brought it to our attention. You love working the numbers, what's the probability of there being a missed prime in those 258k exponents?

About one or two percent, depending on the probability of the first LL test being wrong.

Forum wisdom says that the early 10M digit region is likely to have a higher error rate, the prize having lured testing here before the hardware was up to it.

[davieddy]

Originally Posted by Uncwilly

Here is a nice visual that illustrates how swell things are going.

In what sense are we doing swell?

[davieddy]

Originally Posted by davieddy

In what sense are we doing swell?

OK I'll start the ball rolling.
At ~8.00 am UTC "expected primes" will click down to 0.898.
Every 3 days, we have a 0.2% probability of a new prime.
Until recently, this was every 4 days.

To maintain this expectation of a prime every 4 years on average, the wavefronts need to advance by 10% each year, requiring an increase of 1/3 in computing. Not unreasonable.

What I will monitor more carefully is whether or not the new LL assignments are keeping up with the 1000 per 3 days required ATM to sustain this.

D

[davieddy]

Originally Posted by davieddy

At ~8.00 am UTC "expected primes" will click down to 0.898.

Oops.
Must have put my clock back instead of forwards.