mersenneforum.org  

Go Back   mersenneforum.org > Great Internet Mersenne Prime Search > Math > Number Theory Discussion Group

Reply
 
Thread Tools
Old 2020-10-23, 11:16   #12
mart_r
 
mart_r's Avatar
 
Dec 2008
you know...around...

25×19 Posts
Default

Quote:
Originally Posted by Dr Sardonicus View Post
Once upon a time, long long ago, I posted a link to a paper discussing this, Prime Number Races.
It's like you can read my mind, I have a printout of that paper (actually one of the first math papers I have made a printout of) and wanted to look into it today to recap on logarithmic measures.

Also, https://www.semanticscholar.org/paper/The-hunt-for-Skewes'-number-Smith/a05c0eadc3eda9301f1b48e1622d3d62d581a04f
mart_r is offline   Reply With Quote
Old 2020-10-23, 12:37   #13
Dr Sardonicus
 
Dr Sardonicus's Avatar
 
Feb 2017
Nowhere

2×1,901 Posts
Question Sequence of 3159 numbers containing 447 primes

Quote:
Originally Posted by R2357 View Post
Anyway, if there indeed is such a sequence, then the first or the occurrence will have been reached by 32 589 158 477 190 044 730, thus way below the lower band of 10^174.
By 53#? I don't know where in the world you got that idea.
Dr Sardonicus is offline   Reply With Quote
Old 2020-10-23, 14:11   #14
R2357
 
"Ruben"
Oct 2020
Nederland

468 Posts
Smile Answer

Quote:
Originally Posted by Dr Sardonicus View Post
By 53#? I don't know where in the world you got that idea.
Simply because in any sequence 53#*n to 53#*n+1, all the numbers that are congruent to x*{2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53} modulo 53# will be divisible by a prime up to 53.

59²=3481, 3481>3159 and 59 is the next prime after 53.

That's why I think the conjecture likely!
R2357 is offline   Reply With Quote
Old 2020-10-23, 16:12   #15
mart_r
 
mart_r's Avatar
 
Dec 2008
you know...around...

25×19 Posts
Default

Just because I can...

Take, for instance,
n = 1566280308578217520031412816790827048467516641360946961779273951
and set p = 163#, or in full,
p = 5766152219975951659023630035336134306565384015606066319856068810

then the numbers n+x are all coprime to p for the following 447 values of x:
Code:
0
2
6
8
12
20
30
32
42
48
50
56
68
72
78
90
92
96
98
102
110
116
128
132
138
146
152
156
158
162
176
186
200
210
212
216
228
230
240
242
246
252
260
266
272
278
282
288
300
306
308
312
320
326
336
338
342
350
366
372
378
380
386
396
398
420
422
428
432
438
440
450
462
470
476
482
488
492
498
506
510
512
516
518
536
546
548
558
572
576
590
600
602
606
618
620
638
648
650
656
662
672
690
692
702
708
716
722
726
740
746
756
762
768
776
782
792
798
806
810
812
818
828
840
846
858
860
866
870
872
876
882
888
890
900
912
926
930
938
942
950
956
960
966
968
980
986
992
1008
1020
1022
1026
1040
1052
1056
1058
1062
1068
1080
1082
1086
1098
1110
1112
1118
1122
1142
1146
1148
1166
1170
1176
1178
1188
1190
1196
1202
1212
1220
1236
1238
1250
1260
1262
1266
1272
1278
1290
1296
1302
1308
1328
1332
1346
1350
1352
1356
1370
1376
1392
1400
1406
1416
1418
1428
1430
1436
1442
1448
1458
1470
1476
1478
1488
1496
1500
1502
1520
1530
1532
1538
1542
1560
1566
1568
1580
1590
1602
1608
1610
1626
1632
1638
1640
1646
1652
1656
1670
1680
1682
1692
1706
1712
1716
1722
1728
1730
1748
1758
1766
1770
1772
1790
1796
1800
1806
1812
1818
1826
1832
1836
1838
1848
1850
1862
1866
1868
1878
1880
1892
1896
1898
1902
1910
1916
1926
1932
1938
1940
1946
1962
1980
1982
1988
1992
2000
2006
2016
2022
2028
2036
2046
2048
2052
2060
2066
2070
2072
2076
2078
2088
2102
2108
2112
2120
2130
2132
2148
2162
2178
2186
2190
2192
2198
2202
2210
2226
2240
2252
2256
2258
2268
2280
2286
2310
2312
2316
2318
2322
2336
2342
2352
2358
2360
2366
2372
2388
2396
2400
2406
2408
2412
2430
2438
2442
2450
2456
2462
2468
2478
2486
2490
2492
2496
2508
2510
2522
2528
2538
2540
2552
2562
2570
2576
2580
2582
2592
2598
2606
2616
2618
2622
2630
2640
2648
2652
2658
2660
2666
2676
2682
2690
2696
2706
2708
2718
2732
2736
2742
2750
2756
2760
2762
2778
2786
2790
2798
2808
2820
2822
2826
2828
2850
2856
2858
2862
2868
2882
2888
2892
2900
2910
2912
2918
2928
2930
2940
2952
2958
2966
2970
2972
2982
2990
2996
3002
3018
3036
3042
3050
3056
3060
3066
3072
3080
3086
3092
3098
3102
3110
3122
3126
3128
3138
3150
3152
3158
And, viola, you have 447 potential prime number candidates in an interval of 3159.
The only thing you have to do now is to add p to n as many times until you find an example where all 447 numbers are prime. Simple as that
mart_r is offline   Reply With Quote
Old 2020-10-23, 18:12   #16
R2357
 
"Ruben"
Oct 2020
Nederland

2·19 Posts
Question

Co prime doesn't mean prime, and I still don't understand why the odds are to think that the conjecture is wrong, okay, it's incompatible with the first, but there is no particular reason to think it's false, why would we rather think that.
R2357 is offline   Reply With Quote
Old 2020-10-23, 18:38   #17
mart_r
 
mart_r's Avatar
 
Dec 2008
you know...around...

60810 Posts
Default

Then, I'm afraid, the odds that anyone can help you understand are about as good as the odds of finding an actual prime-447-tuplet.

Edit, FWIW:
If the first Hardy-Littlewood conjecture was wrong, this would have far-reaching consequences for prime number theory, probably of a magnitude as disproving the Riemann hypothesis.
The second conjecture is based merely on human intuition; conjectures of this kind have been proven wrong a myriad of times in the past.

Last fiddled with by mart_r on 2020-10-23 at 19:13
mart_r is offline   Reply With Quote
Old 2020-10-23, 18:46   #18
Dr Sardonicus
 
Dr Sardonicus's Avatar
 
Feb 2017
Nowhere

EDA16 Posts
Default

Quote:
Originally Posted by mart_r View Post
Just because I can...

Take, for instance,
n = 1566280308578217520031412816790827048467516641360946961779273951
and set p = 163#, or in full,
p = 5766152219975951659023630035336134306565384015606066319856068810

then the numbers n+x are all coprime to p for the following 447 values of x:
Code:
0
2
6
8
12
<snip>
3158
And, viola, you have 447 potential prime number candidates in an interval of 3159.
The only thing you have to do now is to add p to n as many times until you find an example where all 447 numbers are prime. Simple as that
There is an "admissibility condition" to the effect that there can be no inevitable prime factor.

To check this, I told Pari-GP to compute the degree-447 polynomial

f = x*(x+2)*(x+6)*(x+8)*(x+12)*...*(x+3158)

and then to compute Mod(Mod(1,p)*f, x^p - x) for all primes p up to 447, to see if any were zero. None were. (Of course, for p > 447, the degree of f is less that p, so the polmod will be Mod(1,p)*f, which isn't 0.)

Looks like you're good to go.

Of course, as indicated by 447 tuples calculations, finding an example where the numbers are all prime might take a while.

Last fiddled with by Dr Sardonicus on 2020-10-24 at 00:31 Reason: awkward phrasing; nixfig ostpy
Dr Sardonicus is offline   Reply With Quote
Old 2020-10-23, 19:20   #19
R2357
 
"Ruben"
Oct 2020
Nederland

1001102 Posts
Cool

Quote:
Originally Posted by R2357 View Post
Simply because in any sequence 53#*n to 53#*n+1, all the numbers that are congruent to x*{2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53} modulo 53# will be divisible by a prime up to 53.

59²=3481, 3481>3159 and 59 is the next prime after 53.

That's why I think the conjecture likely!
Okay, in my calculation, I overlooked the fact that numbers congruent to a composite up to 53# divisible by primes from 59 onwards, modulo 53# may be prime, mea culpa :|

But still, there remains two conditions :
- first the positions of the 3159 numbers must contain 447 numbers that are congruent to numbers not divisible by {2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53} modulo 53#;
- secondly, the sequence chosen must not contain more numbers divisible by primes from 59 onwards than the difference (positive of course) between the numbers congruent to numbers not divisible by {2, 3, 5, 7, 11, 13, 17, 19, 23, 29, 31, 37, 41, 43, 47, 53} modulo 53# and 447.

Good luck successfully reaching the two above conditions :)
R2357 is offline   Reply With Quote
Old 2020-10-24, 14:27   #20
sweety439
 
sweety439's Avatar
 
Nov 2016

26·3·13 Posts
Default

Quote:
Originally Posted by R2357 View Post
Hello,

I've looked at the way primes behave in sequences which have a primorial length and, as I mentionned a few weeks ago, I came across the 2nd Hardy-Littlewood conjecture, which was expained to me as being incompatible with the first Hardy-Littlewood conjecture.

What I don't understand is why do we believe that it's the first one which is right, personnaly, I believe it's the second, here's why I think so :

The second Hardy-Littlewood conjecture states that with x>1, pi(x)>=pi(x+y)-pi(y).
That seems to me as very probable! Already, the first sequence of 30 contains 10 primes, none of the others will contain more than 7.

Why would this conjecture seem to be false?
If Schinzel's hypothesis H is true, then the second Hardy-Littlewood conjecture is false, since Schinzel's hypothesis H covers both Bunyakovsky conjecture and Dickson's conjecture, Dickson's conjecture covers first Hardy–Littlewood conjecture, but the first & second Hardy–Littlewood conjecture cannot be both true since ....


primepi(3159) = 446, but there may be 3159 consecutive positive integers which contain 447 primes (unlike the case of 6 consecutive positive integers, any 6 consecutive positive integers >= {4,5,6,7,8,9} contain at most 2 primes, where prmiepi(6) = 3), this is a case of law of small numbers.

However, Schinzel's hypothesis H is widely believed to be true, like generalized Riemann hypothesis and abc conjecture.

Last fiddled with by sweety439 on 2020-10-24 at 14:28
sweety439 is offline   Reply With Quote
Old 2020-10-26, 17:07   #21
R2357
 
"Ruben"
Oct 2020
Nederland

2·19 Posts
Default

Quote:
Originally Posted by mart_r View Post
The second conjecture is based merely on human intuition;
I don't understand what you really mean by "human intuition", when we look at the primes from 1 to 3159, this makes more than 14% of the natural numbers, the non-multiples of one and two digit numbers make up barely 12% of them.

furthermore, it's not at all similar to the twin-prime conjecture, because here, we're looking for a bigger prime density than in the beginning.

And it's not just an intuition : each p# takes away the number of possible primes in pn-1# for every p#, not to mention all of the non-primes divisible by numbers bigger than p : in 17#, in the first sequence of 17# is already made out of more composites only by numbers from19, than primes, I let you imagine in for example, let's say 10³³*17# to 10³³+1*17# :)
R2357 is offline   Reply With Quote
Old 2020-10-26, 19:30   #22
CRGreathouse
 
CRGreathouse's Avatar
 
Aug 2006

2·2,969 Posts
Default

Quote:
Originally Posted by R2357 View Post
I don't understand what you really mean by "human intuition", when we look at the primes from 1 to 3159, this makes more than 14% of the natural numbers, the non-multiples of one and two digit numbers make up barely 12% of them.
There are 446 primes up to 3159, so 14.1%. There are 422 numbers coprime to 1, 2, ..., 99, or 13.4%. What does this show? 3159 isn't a counterexample to the second Hardy-Littlewood conjecture. But there are infinitely many other cases to check, and we should be cautious in generalizing from 'one case holds' to 'all cases hold': as is often the case in number theory, the first counterexample may be large.
CRGreathouse is offline   Reply With Quote
Reply

Thread Tools


Similar Threads
Thread Thread Starter Forum Replies Last Post
Can we prove Beal conjecture assuming ABC conjecture? didgogns Miscellaneous Math 1 2020-08-05 06:51
I think I have a conjecture MathDoggy Miscellaneous Math 11 2019-04-17 07:05
This conjecture may be useful. reddwarf2956 Prime Gap Searches 2 2016-03-01 22:41
The constant in Hardy-Littlewood's Conjecture F CRGreathouse Math 4 2010-07-06 17:47
Incompatible Hardy-Littlewood conjectures: a DC project? GP2 Math 2 2003-09-14 04:36

All times are UTC. The time now is 09:27.

Fri Nov 27 09:27:52 UTC 2020 up 78 days, 6:38, 4 users, load averages: 1.33, 1.19, 1.20

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.