20190706, 13:01  #23 
Feb 2017
Nowhere
2·5^{2}·71 Posts 
Perhaps samuel is mipsplepping the calculus method I mentioned here.
I have every right to make sport of samuel's avatar, and to at his ordering me not to. Freedom of speech, you know. And if samuel wants to insist his avatar is a photo, I guess he must be, not a demon, but the serpent himself. 
20190706, 13:28  #24 
May 2011
Orange Park, FL
3×17^{2} Posts 
Perhaps you could tell us M53 and M54 so we could get a head start on verifying them....

20190706, 13:54  #25  
Undefined
"The unspeakable one"
Jun 2006
My evil lair
5788_{10} Posts 
Quote:
https://mersenneforum.org/showpost.p...&postcount=213 Generates ALL 51 exponents in order, perfectly. So do you agree that the output for M52 must be correct? 

20190709, 20:31  #26  
"silent magician!!"
Apr 2019
nowheresville califo
101 Posts 
Quote:


20190709, 20:51  #27 
Bamboozled!
"πΊππ·π·π"
May 2003
Down not across
2·3·5·337 Posts 

20190709, 20:57  #28 
Sep 2016
149_{16} Posts 

20190710, 00:05  #30 
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest
11DC_{16} Posts 
circuit
Hi, not sure if this is the right place to post this, but I have a conceptual design for a fast and relatively simple circuit that rather accurately predicts whether a prime mersenne exponent produces a mersenne prime. It's a 30input NOR. For p>0, its output is always low (indicating composite). It's only wrong for 0 and the mersenne primes, known or unknown. That's only about a 1.1 ppm error rate, which is way more accurate than most lab measurements or design calculations. :)
Edit: a more efficient design gets rid of the error on zero input and reduces power consumption and cost, and easily scales to higher p. Inputs are opens or may be omitted; output is grounded. Okay, how about some fancy math, say a polynomial, to improve the accuracy to the desired level? Nope; no polynomial can generate only primes. https://en.wikipedia.org/wiki/Formula_for_primes 
20190710, 00:26  #31  
"TF79LL86GIMPS96gpu17"
Mar 2017
US midwest
2^{2}·3^{2}·127 Posts 
Quote:
Use a randomly selected subset of almost all the known mersenne primes to predict the rest of the set of known mersenne primes. How does your method fare for predictions yielding the excluded few? Or notsorandomly; if you had only up to M50 as input, how does it fare in predicting M51 correctly? If anyone could provide a working predictor, I'd happily amend https://www.mersenneforum.org/showpo...04&postcount=5 The frequency difference for p mod 8 versus other values is intriguing. https://www.mersenneforum.org/showpo...00&postcount=4 Last fiddled with by kriesel on 20190710 at 00:28 

20190710, 01:11  #32  
"silent magician!!"
Apr 2019
nowheresville califo
101 Posts 
Quote:
Quote:
Quote:


20190710, 01:17  #33  
Undefined
"The unspeakable one"
Jun 2006
My evil lair
2^{2}×1,447 Posts 
Quote:
But maybe the python code is too advanced for you to understand. 

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Super Cullen & Woodall primes  Citrix  And now for something completely different  1  20171026 09:12 
Super Newbie Here  Hatterz  Information & Answers  15  20120911 22:28 
GIMPS people are Bgrade superagents!  kuratkull  Soap Box  2  20071203 16:50 
Intel Mac Minis Factoring super farm?  delta_t  Hardware  5  20060308 10:25 
personal super computer  crash893  Hardware  16  20051008 13:14 