20120919, 16:06  #34 
Aug 2002
Buenos Aires, Argentina
7^{2}×29 Posts 
Based on the previous post: let n = (2^{2p} 2^{p}+1)/3 where p is a prime number, is it true that 2^{n1} = 1 (mod n) ?

20120919, 16:55  #35  
"Robert Gerbicz"
Oct 2005
Hungary
2·3^{2}·5·17 Posts 
Quote:
Observe that: (2^(2*p)2^p+1)/3  2^(6*p)1 so it is enough to prove that for p>2: 2^(n1)==1 mod 2^(6*p)1 it is equivalent to n1==0 mod (6*p) so (2^(2*p)2^p2)/3==0 mod (6*p) (18*p)2^p*(2^p1)2 Let p>3 then it is true mod p (use Fermat's little theorem), 2 also divides, this is trivial since 2^p is even. For 9 use: p=6k+1 then 2^p=={2,5} mod 9, for the two cases: 2^p*(2^p1)2=={0,18}==0 mod 9. Since 2,9,p are relative prime numbers for p>3 it means that the divisibility is also true for 2*9*p=18*p. For p=3: 2^(n1)==1 mod n is also true. 

20120919, 17:18  #36 
Aug 2002
Buenos Aires, Argentina
1421_{10} Posts 
Thanks.

20210320, 08:03  #37 
Mar 2021
1_{10} Posts 
Hi everyone
I know I am a bit late.... BUT I have the answer to your problem: take N = 2^524287  1. Notice that 524287 = 2^19  1, this is a clue for the proof ;) Indeed, you just need to consider the sequence X(n+1) = 2^X(n)  1 with X(0) = 19 and proove that all terms of the sequence verify X(n)*X(n+1) divide 2^X(n)  2. Have a nice day! 
20210321, 16:14  #38 
Mar 2021
France
12_{16} Posts 
Hi
I would like to know if a^(2^n) mod (2^n1) = a² when a is between 0 and sqrt(2^n1) and a is an integer when 2^n1 is prime and only if it's prime ? I tried this with some Mersenne exposant and it apparently works. It can't be used to eliminate nonMersenne exponant quickly ? Thanks :D 
20210321, 22:43  #39 
"Matthew Anderson"
Dec 2010
Oregon, USA
1760_{8} Posts 
In My Humble Opinion (IMHO) the even numbers are easier to keep track of

20210322, 03:03  #40  
Romulan Interpreter
"name field"
Jun 2011
Thailand
10011010000100_{2} Posts 
Quote:
Yes, it can be used to eliminate composite Mp's, but not "fast". Well, not faster than we are doing already. 

20210523, 15:32  #41  
Mar 2021
France
2×3^{2} Posts 
Quote:
And by the way I noticed something interesting : if a+b+c = 2^n1 you can found number with a²+b²+c² = (4^n1)/3 and it seems it works with Mersenne number (it seems it doesn't work with composite Mersenne number like 2^111 or 2^231) For example 4+2+1 = 7 (2^31) and 4²+2²+1² = 21 (4^31)/3 another example : 14+9+8 = 31 (2^51) and 14²+9²+8² = 341 (4^51)/3 There is a way to explain this ? Thanks :) 

20210523, 16:55  #42 
Jan 2021
California
3·101 Posts 
I'm not sure what you are trying to say here...
4+3+0 = 7 (2^31), but 4^2 + 3^2 + 0^2 = 25 which is not (4^31)/3 13 + 10 + 8 = 31 (2^51) but 13^2 + 10^2 + 8^2 = 333 which is not (4^51)/3 So clearly there's something I'm missing. 
20210523, 17:09  #43  
Mar 2021
France
12_{16} Posts 
Quote:
a+b+c = 2^n1 and a²+b²+c²=(4^n1)/3 If a+b+c is a prime Mersenne number. you can't find any numbers for example for a+b+c = 2^111 and a²+b²+c² for (4^111)/3 it seems it doesn't exist. 

Thread Tools  
Similar Threads  
Thread  Thread Starter  Forum  Replies  Last Post 
Twin Prime Conjecture Proof  Steve One  Miscellaneous Math  53  20190318 00:34 
Semiprime factorization conjecture  Alberico Lepore  Alberico Lepore  7  20180216 08:27 
Conjecture prime numbers, demonstration possible?  Godzilla  Miscellaneous Math  5  20160516 12:44 
Prime abc conjecture b == (a1)/(2^c)  miket  Miscellaneous Math  6  20130522 05:26 
The Twin Prime Conjecture Song  Templus  Lounge  9  20060314 16:30 