mersenneforum.org  

Go Back   mersenneforum.org > Extra Stuff > Blogorrhea > gophne

Reply
 
Thread Tools
Old 2017-12-27, 03:47   #1
gophne
 
Feb 2017

3×5×11 Posts
Exclamation "New" primality test/check

I think that I might have discovered a new "primality" test, as well as a new prime sieve. I think that the primality test might be computer-time friendly as well, as it consists of a simple straight calculation/logic test only. I think it is big. What would be the most appropriate route to submit it to the math community for scrutiny?
gophne is offline   Reply With Quote
Old 2017-12-27, 03:57   #2
CRGreathouse
 
CRGreathouse's Avatar
 
Aug 2006

134628 Posts
Default

The usual way to announce a mathematical discovery is to write a paper, push a preprint to the arXiv, then publish the paper. In the fortunate case that the algorithm can be coded easily I would recommend programming a reference implementation and including the timing results in the preprint and paper. If they improve on the current state of the art, it might be worth posting a note to NMBRTHRY while you're writing the paper. (Since you're here, you could also post timings here; depending on what they are we could advise further.)
CRGreathouse is offline   Reply With Quote
Old 2017-12-27, 04:35   #3
gophne
 
Feb 2017

16510 Posts
Default

Thanks CRGreathouse.

I had previous attempts at prime innovation on this site, with a bit of embarrassing results!

I am a hobbyist only, and have made some interesting discoveries (using hobbyist approach of "go where my fancy takes me"). I think my present "test" is indeed interesting and had not found anything like it in the "primality" literature, especially on the internet as well -not to say that it is unique/new, but submitting it would of course quickly verify the authenticity of the system/test in a split second.

I was thinking of submitting the "test" on this forum, but I could write a "paper" as I can ask my son/daughter (who is more competent w.r.t writing academic papers) to assist if required.

I am really exited, but it could of course turn out to be a damp squib as well.

Thanks for the initial advise.

Last fiddled with by gophne on 2017-12-27 at 04:37 Reason: spelling error
gophne is offline   Reply With Quote
Old 2017-12-27, 05:11   #4
CRGreathouse
 
CRGreathouse's Avatar
 
Aug 2006

2×2,969 Posts
Default

Quote:
Originally Posted by gophne View Post
I had previous attempts at prime innovation on this site, with a bit of embarrassing results!
In that case I'd suggest doing the same again: show what you have and let the forum give feedback. If it's good they can give you advice on how to prepare the work for publication, if not you're saved the effort and possible embarrassment.

Here are some things we (and others) would likely be interested in checking:
  • Is the description clear enough that we can understand the intended algorithm?
  • Does it classify all primes as prime?
  • Does it classify all composites as composite?
  • Does it run in reasonable time on average primes?
  • Does it run in reasonable time on worst-case primes?
  • Can you prove the above claims?
  • Is the algorithm easy to implement?
CRGreathouse is offline   Reply With Quote
Old 2017-12-27, 14:14   #5
rogue
 
rogue's Avatar
 
"Mark"
Apr 2003
Between here and the

2·3,001 Posts
Default

Can you write a computer algorithm to execute the test? If so, have you tested it against known primes and composites? Do you know how fast/slow the test is compared to other primality tests? Can it be used as a primality test for numbers of any form?
rogue is offline   Reply With Quote
Old 2017-12-27, 18:11   #6
gophne
 
Feb 2017

3×5×11 Posts
Default New Frontier

Hi CRGreathouse

The algorithm is very clear.
ALL primes are covered- tested to M34 -2^1,257,787-1, Seriously!
ALL composites are identified!
Time O(log n) -Very very fast. Serious.
ALL primes are defined -NO false primes expected as per algorithm logic.
The algorith is a logical formulation, so can be verified quickly.
If authenticated, will compare to simplicity of Euclid's Proof for Infinite number of Primes.
If true, will be a new frontier in PNT

Call me crazy

Last fiddled with by gophne on 2017-12-27 at 18:36 Reason: Correction for M34- 2^1,257,787-1
gophne is offline   Reply With Quote
Old 2017-12-27, 18:31   #7
chalsall
If I May
 
chalsall's Avatar
 
"Chris Halsall"
Sep 2002
Barbados

221528 Posts
Default

Quote:
Originally Posted by gophne View Post
The algorithm is very clear.
Under US of A law, an algorithm cannot be patented. So, you are unlikely to make any money from this discovery directly.

On the other hand, this would be a ground breaking discovery if it was true.

Why don't you give us the algorithm so the experts here can vet it?
chalsall is offline   Reply With Quote
Old 2017-12-27, 18:34   #8
gophne
 
Feb 2017

101001012 Posts
Default Grease lightning

Hi Rogue

Yes, I run the algorithm in SAGE. Time, is the time needed in SAGE to caculate straight math formula/logic test, of course involving very large numbers. I tested M34 -2^1,257,787-1. I won't mention the time taken as I will be debarred from this site. (I use a Intel(R) Core(TM)2 CPU T5500 @ 1.66 GHz 1.67 GHz RAM 2.00 GB, 64 bit O/S x64 based processor).

I have tested it to M34. If vindicated, the algorithm will replace GIMP. (Your are allowed to scream out in vexation at this point w.r.t the claims I am making...I don't expect anybody to consider my claims seriously at this stage -I am looking for the best way to submit my algorith/system

Tests ALL known Primes...it is that surreal.

The algorithm maps ALL primes in sequence.

Now you can release your breath and burst out laughing!!!
gophne is offline   Reply With Quote
Old 2017-12-27, 18:47   #9
gophne
 
Feb 2017

3×5×11 Posts
Default

Quote:
Originally Posted by chalsall View Post
Under US of A law, an algorithm cannot be patented. So, you are unlikely to make any money from this discovery directly.

On the other hand, this would be a ground breaking discovery if it was true.

Why don't you give us the algorithm so the experts here can vet it?
Hi Chalsall

That was my intention, to submit the algorithm to the world on this site, with the advantage that the algorithm will be ripped to pieces in a matter of micro seconds, if it is a hoax, with myself probable banned for life from this site, if not any other serious math site.

However, I would like to believe that a Site such as this would be able to offer me some sort of protection of my work/intellectual property, even if is open source.

If the algorithm is confirmed, this would have a serious implication on Prime Number Theory.

This is my dilemma.
gophne is offline   Reply With Quote
Old 2017-12-27, 18:51   #10
gophne
 
Feb 2017

3·5·11 Posts
Default

Post erased as it was a duplicate of #9. Apologies.

Last fiddled with by gophne on 2017-12-27 at 19:10 Reason: Reply duplicated for some reason
gophne is offline   Reply With Quote
Old 2017-12-27, 19:01   #11
guptadeva
 
Dec 2017

1100102 Posts
Default

Quote:
Originally Posted by gophne View Post
What would be the most appropriate route to submit it to the math community for scrutiny?
the mathematical community was the first scientific group to understand, that classical publications in books, journals or other printed media has limits.

you could always publish your results in arxiv (https://arxiv.org/help/submit) and/or here in mersenneforum where the density of prime experts is very high and where you would get deep and quick feedback needed to determine the future of your new primality check ...
guptadeva is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
gpuOwL: an OpenCL program for Mersenne primality testing preda GpuOwl 2607 2020-11-25 02:05
GQQ: a "deterministic" "primality" test in O(ln n)^2 Chair Zhuang Miscellaneous Math 21 2018-03-26 22:33
Aouessare-El Haddouchi-Essaaidi "test": "if Mp has no factor, it is prime!" wildrabbitt Miscellaneous Math 11 2015-03-06 08:17
"New primality proving test from Alex Petrov" ewmayer Math 11 2007-04-23 19:07
P-1 B1/B2 selection with "Test=" vs "Pfactor=" James Heinrich Software 2 2005-03-19 21:58

All times are UTC. The time now is 02:39.

Wed Nov 25 02:39:25 UTC 2020 up 75 days, 23:50, 4 users, load averages: 1.48, 1.33, 1.36

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2020, Jelsoft Enterprises Ltd.

This forum has received and complied with 0 (zero) government requests for information.

Permission is granted to copy, distribute and/or modify this document under the terms of the GNU Free Documentation License, Version 1.2 or any later version published by the Free Software Foundation.
A copy of the license is included in the FAQ.