Estimate the new primes

davieddy · 2008-09-12, 09:51

Based on all the evidence, I guess exponents 42M for Aug, 38M for Sept.

Flatlander says 42M and 28M.

We have about 3 days to go before all is revealed.

retina · 2008-09-12, 10:02

I think that both are >10M digits (>33M exponent). This begin based upon my assumption that it would seem odd to hold back the announcement of the first discovery for a rather uninteresting (relativity speaking) 28M exponent.

mcduck · 2008-09-12, 10:27

I have to say 42M and 28M

davieddy · 2008-09-12, 10:34

On the contrary, it would be surprising if any more sub 10M digit
Mersenne primes were discovered.

rgiltrap · 2008-09-12, 11:13

Just to confuse things for you. We started the first verification on one server using old Mlucas code and finished on another faster one using the new Mlucas code. The second verification was started on the slower server as well but this time with the new code but also got moved to the faster server once it became available after completing the first verification.

The performance delta between the old Mlucas code and the new Mlucas code was quite significant.

Based on this, speculation based on the times we provided is going to be ambiguous at best

Andi47 · 2008-09-12, 11:35

My guess is 41M for the first exponent and just below 10M digits for the second one.

davieddy · 2008-09-12, 11:38

THX Rob.
I stand by 38M for September.
You could use 2048K instead of 4096K FFT, and it is
in the zone where most exponents are currently being tested.

davieddy · 2008-09-12, 12:12

From the notorious "Misc Math" forum thread comes:

Quote:

Originally Posted by cochet

It would be logic, for me, that M discovered on 6 th september is : 37156667

This exponent is into the interval I have always privileged. All tests are conclusive.

But I am a little, little "amateur".

Alain

What more proof is needed?

R. Gerbicz · 2008-09-12, 15:54

Quote:

Originally Posted by davieddy

Based on all the evidence, I guess exponents 42M for Aug, 38M for Sept.
Flatlander says 42M and 28M.
We have about 3 days to go before all is revealed.

You are a little impatient!
I would wait for independent verifications on a Commodore64.

Andi47 · 2008-09-12, 16:27

Quote:

Originally Posted by R. Gerbicz

You are a little impatient!
I would wait for independent verifications on a Commodore64.

ROTFL!!!!

(Then we would need a great amount of liquid helium to cool the overclocked C64...)

Kevin · 2008-09-12, 16:28

Quote:

Originally Posted by davieddy

What more proof is needed?

After the Olympics, I don't think many people take predictions from people named Alain very seriously...

2008-09-12, 09:51	#1
davieddy "Lucan" Dec 2006 England 2·3·13·83 Posts	Estimate the new primes Based on all the evidence, I guess exponents 42M for Aug, 38M for Sept. Flatlander says 42M and 28M. We have about 3 days to go before all is revealed. Last fiddled with by davieddy on 2008-09-12 at 10:26

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
I need a cost estimate for some data.	Kathegetes	Data	40	2014-07-28 18:49
Benchmark Estimate	Primeinator	Information & Answers	8	2009-06-11 23:39
V25.7 TF estimate way out ... or am I?	petrw1	PrimeNet	5	2008-11-08 02:23
A tricky way to estimate pi(x)	XYYXF	Math	13	2006-09-01 15:44
Estimate of sieve duration for 2^811 - 1	dsouza123	NFSNET Discussion	16	2004-03-12 15:28

2008-09-12, 10:02	#2
retina Undefined "The unspeakable one" Jun 2006 My evil lair 14120₈ Posts	I think that both are >10M digits (>33M exponent). This begin based upon my assumption that it would seem odd to hold back the announcement of the first discovery for a rather uninteresting (relativity speaking) 28M exponent.

2008-09-12, 10:27	#3
mcduck Sep 2008 Kristiansand, Norway 3·37 Posts	I have to say 42M and 28M

2008-09-12, 10:34	#4
davieddy "Lucan" Dec 2006 England 2×3×13×83 Posts	On the contrary, it would be surprising if any more sub 10M digit Mersenne primes were discovered.

2008-09-12, 11:13	#5
rgiltrap Apr 2006 Down Under 59₁₆ Posts	Just to confuse things for you. We started the first verification on one server using old Mlucas code and finished on another faster one using the new Mlucas code. The second verification was started on the slower server as well but this time with the new code but also got moved to the faster server once it became available after completing the first verification. The performance delta between the old Mlucas code and the new Mlucas code was quite significant. Based on this, speculation based on the times we provided is going to be ambiguous at best

2008-09-12, 11:35	#6
Andi47 Oct 2004 Austria 2·17·73 Posts	My guess is 41M for the first exponent and just below 10M digits for the second one.

2008-09-12, 11:38	#7
davieddy "Lucan" Dec 2006 England 2·3·13·83 Posts	THX Rob. I stand by 38M for September. You could use 2048K instead of 4096K FFT, and it is in the zone where most exponents are currently being tested.