Aouessare-El Haddouchi-Essaaidi "test": "if Mp has no factor, it is prime!"

wildrabbitt · 2015-03-05, 17:30

Hi,

I'm not sure whether this has been discussed. I searched for it here on the forums but didn't find anything.

I'm expecting that it's naïve to think it's not known about here especially since there was a post about this
on the mailing list about a year ago.

The post was about a new way of testing whether or not Mersenne numbers are prime or not. The post included
a link to this article :

http://ijcaonline.org/archives/volum...er3/17505-8053

I was half expecting a new program to be in use by now by the GIMPS and I don't believe one is so I'm starting to wonder
why (since the document clearly outlines the details and is mathematically sound).

So can someone update me on the effect this news had on the mersenne.org community - i.e what the response was,
and if there are any plans to use this algorithm.

Dubslow · 2015-03-05, 18:04

http://www.mersenneforum.org/showthread.php?p=381194

That article, the test, isn't worth the figurative paper it's printed on.

petrw1 · 2015-03-05, 19:10

... to determine it is prime vs 11 iterations for Prime95... WOW?!?!?!?

However, it took 1,525 iterations for MP59 to find the first factor
748 iterations to determine MP31 is prime

and it quickly gets much worse.

wildrabbitt · 2015-03-05, 20:13

Thanks Guys.

So is it just the claim that it's more efficient CPU time - wise that's wrong ?

Does the algorithm actually work?

There was no proof on the paper so I'm unable to even attempt to work the answer out for myself.

Dubslow · 2015-03-05, 20:33

Quote:

Originally Posted by wildrabbitt

Thanks Guys.

So is it just the claim that it's more efficient CPU time - wise that's wrong ?

Does the algorithm actually work?

There was no proof on the paper so I'm unable to even attempt to work the answer out for myself.

Read the other thread I linked. The algorithm is basically trial factoring, like mfakt*, except done really really really really poorly.

petrw1 · 2015-03-05, 20:34

Quote:

Originally Posted by wildrabbitt

Thanks Guys.

So is it just the claim that it's more efficient CPU time - wise that's wrong ?

Does the algorithm actually work?

There was no proof on the paper so I'm unable to even attempt to work the answer out for myself.

Using Excel I was able to verify that it works at least up to MP61.....a LONG LONG way from where Prime95 is currently

wildrabbitt · 2015-03-05, 20:41

Without proof it really is useless except to them if they've got one, but why would they keep the proof secret?

Batalov · 2015-03-05, 22:17

The proof is obvious, however the test is impractical. (I am trying to use only nice words.)

It is important to know that "ijcaonline" is a predatory "journal" that publishes anything as long as the submitters pay the required fee. As such, it should be avoided for reading (and as for writing/publishing - only if one likes to publish on the walls of public restrooms; it's the same sort of thing).

ewmayer · 2015-03-05, 22:47

Quote:

Originally Posted by wildrabbitt

Without proof it really is useless except to them if they've got one, but why would they keep the proof secret?

The point is, even with proof of correctness the test is useless in practice.

In theory I can do a rigorous '1-iteration' test of any M(p) by simply feeding it to e.g. the Pari 'factor' command. In practice, once p gets larger than a few hundred bits, the needed runtime becomes impractically large.

Uncwilly · 2015-03-06, 00:30

Quote:

Originally Posted by Batalov

It is important to know that "ijcaonline" is a predatory "journal" that publishes anything as long as the submitters pay the required fee.

Same idea

Quote:

Originally Posted by rogue

WHY A FAKE ARTICLE TITLED "CUCKOO FOR COCOA PUFFS?" WAS ACCEPTED BY 17 MEDICAL JOURNALS

science_man_88 · 2015-03-06, 02:11

I know I shouldn't play with this, However I manipulated the equation in Theorem 1 to:

$k>(2np^2-2np+\frac{3}{2}p)$ where k is such that 2*k*p+1 is Mp I was thinking of manipulating it more, but if I've messed up already it's pointless to try to extrapolate further. EDIT: I think I messed up the +3/2 part I think it should be -3/2 now that I looked over my manipulation on paper. EDIT2: I found a fatal error I failed to correct further up if only I hadn't gone through 27 steps.

2015-03-05, 17:30	#1
wildrabbitt Jul 2014 2×3²×5² Posts	Aouessare-El Haddouchi-Essaaidi "test": "if Mp has no factor, it is prime!" Hi, I'm not sure whether this has been discussed. I searched for it here on the forums but didn't find anything. I'm expecting that it's naïve to think it's not known about here especially since there was a post about this on the mailing list about a year ago. The post was about a new way of testing whether or not Mersenne numbers are prime or not. The post included a link to this article : http://ijcaonline.org/archives/volum...er3/17505-8053 I was half expecting a new program to be in use by now by the GIMPS and I don't believe one is so I'm starting to wonder why (since the document clearly outlines the details and is mathematically sound). So can someone update me on the effect this news had on the mersenne.org community - i.e what the response was, and if there are any plans to use this algorithm.

2015-03-05, 19:10	#3
petrw1 1976 Toyota Corona years forever! "Wayne" Nov 2006 Saskatchewan, Canada 2³×5×7×19 Posts	Yes MP13 does only take 4 iterations ... ... to determine it is prime vs 11 iterations for Prime95... WOW?!?!?!? However, it took 1,525 iterations for MP59 to find the first factor 748 iterations to determine MP31 is prime and it quickly gets much worse.

2015-03-06, 02:11	#11
science_man_88 "Forget I exist" Jul 2009 Dartmouth NS 2×5²×13² Posts	did I do the math correct ? I know I shouldn't play with this, However I manipulated the equation in Theorem 1 to: $k>(2np^2-2np+\frac{3}{2}p)$ where k is such that 2kp+1 is Mp I was thinking of manipulating it more, but if I've messed up already it's pointless to try to extrapolate further. EDIT: I think I messed up the +3/2 part I think it should be -3/2 now that I looked over my manipulation on paper. EDIT2: I found a fatal error I failed to correct further up if only I hadn't gone through 27 steps. Last fiddled with by science_man_88 on 2015-03-06 at 02:53 Reason: had to edit what I said needed editing.

Similar Threads
Thread	Thread Starter	Forum	Replies	Last Post
GQQ: a "deterministic" "primality" test in O(ln n)^2	Chair Zhuang	Miscellaneous Math	21	2018-03-26 22:33
Speeding up double checking when first test returns "prime"	Unregistered	PrimeNet	16	2006-02-28 02:00
P-1 B1/B2 selection with "Test=" vs "Pfactor="	James Heinrich	Software	2	2005-03-19 21:58
Would Minimizing "iterations between results file" may reveal "is not prime" earlier?	nitai1999	Software	7	2004-08-26 18:12

2015-03-05, 18:04	#2
Dubslow Basketry That Evening! "Bunslow the Bold" Jun 2011 40<A<43 -89<O<-88 3×29×83 Posts	http://www.mersenneforum.org/showthread.php?p=381194 That article, the test, isn't worth the figurative paper it's printed on.

2015-03-05, 20:13	#4
wildrabbitt Jul 2014 1C2₁₆ Posts	Thanks Guys. So is it just the claim that it's more efficient CPU time - wise that's wrong ? Does the algorithm actually work? There was no proof on the paper so I'm unable to even attempt to work the answer out for myself.

2015-03-05, 20:41	#7
wildrabbitt Jul 2014 111000010₂ Posts	Without proof it really is useless except to them if they've got one, but why would they keep the proof secret?

2015-03-05, 22:17	#8
Batalov "Serge" Mar 2008 Phi(4,2^7658614+1)/2 3³·13·29 Posts	The proof is obvious, however the test is impractical. (I am trying to use only nice words.) It is important to know that "ijcaonline" is a predatory "journal" that publishes anything as long as the submitters pay the required fee. As such, it should be avoided for reading (and as for writing/publishing - only if one likes to publish on the walls of public restrooms; it's the same sort of thing).