-
Notifications
You must be signed in to change notification settings - Fork 0
/
upper-bounds_with_residue.pl
282 lines (203 loc) · 7.52 KB
/
upper-bounds_with_residue.pl
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
#!/usr/bin/perl
# Author: Daniel "Trizen" Șuteu
# Date: 28 January 2019
# https://github.com/trizen
# A new algorithm for generating Fermat pseudoprimes to base 2.
# See also:
# https://oeis.org/A050217 -- Super-Poulet numbers: Poulet numbers whose divisors d all satisfy d|2^d-2.
# https://oeis.org/A214305 -- Fermat pseudoprimes to base 2 with two prime factors.
# See also:
# https://en.wikipedia.org/wiki/Fermat_pseudoprime
# https://trizenx.blogspot.com/2018/08/investigating-fibonacci-numbers-modulo-m.html
use 5.020;
use warnings;
use strict;
use ntheory qw(:all);
use experimental qw(signatures);
use List::Util qw(uniqnum);
# a(n) is the smallest odd composite k such that prime(n)^((k-1)/2) == -1 (mod k) and b^((k-1)/2) == 1 (mod k) for every natural b < prime(n).
# a(n) : 3277, 5173601, 2329584217, 188985961, ... Is this sequence infinite?
# I found upper-bounds only for the next 5 terms:
# a(5) <= 32203213602841
# a(6) <= 323346556958041
# a(7) <= 2528509579568281
# a(8) <= 5189206896360728641
# a(9) <= 12155831039329417441
# Notice that the upper-bounds a(5)-a(9) are all semiprimes and are of the form p * (2*(p-1) + 1), for some prime number p.
# I couldn't find an upper-bound for a(10), which may be larger than 2^64 (assuming that it exists).
# More generally, we can look for numbers of the form k = p*q, where p and q are odd primes, such that d|((p-1)/2) and d|((q-1)/2), satisfying:
# prime(n)^d == -1 (mod p)
# prime(n)^d == -1 (mod q)
# This can be implemented efficiently by iterating over the odd primes bellow a certain limit and grouping primes together by divisors `d` that satisfy the wanted property.
# When a group has two or more primes, we iterate over all the combinations of two of those primes, say p and q, and generate k = p*q.
# Once we have k = p*q, it's very likely that:
# prime(n)^((k-1)/2) == -1 (mod k)
# For each such number k, we check if:
# b^((k-1)/2) == 1 (mod k) for every natural b < prime(n)
# Several more upper-bounds (example), with possible gaps, for each a(2)-a(7):
# a(2) <= { 5173601, 6787327, 15978007, 314184487, 581618143, 682528687, 717653129, 1536112001, 1854940231, 1860373241, 2693739751, 3344191241, 4916322919, ... }
# a(3) <= { 2329584217, 4394140633, 18491988217, 71430713857, 146463308377, 306190965697, 542381064457, 662181331897, 1749760817617, 2600044182457, 3420965597017, ... }
# a(4) <= { 188985961, 922845241, 41610787921, 55165631041, 85103495641, 130173194161, 178420380841, 451403438041, 577255246921, 669094855201, 842611640041, 1038185022241, ... }
# a(5) <= { 32203213602841, 92223430598881, 181122243848281, 244362410507881, 323731175307121, 359643366744481, 383326325785201, 503561507122801, ... }
# a(6) <= { 323346556958041, 449287527233161, 871447814622001, 2419909071073201, ... }
# a(7) <= { 2528509579568281, 51558565269914641, 251985537187183801, ... }
# Let b(n) be the smallest odd composite k such that q^((k-1)/2) == -1 (mod k) for every prime q <= prime(n).
#~ b(1) = 3277
#~ b(2) = 1530787
#~ b(3) <= 3697278427
#~ b(4) <= 118670087467
#~ b(5) <= 2152302898747
#~ b(6) <= 614796634515444067
#~ b(7) <= 614796634515444067
my $N = 5;
my $multiplier = 4;
my $P = nth_prime($N);
my $MAX = ~0;
my $LIMIT = 1e5;
use Math::GMPz;
sub prod {
my $prod = 1;
foreach my $n(@_) {
$prod *= $n;
}
$prod;
}
# a(5) = 11530801
# a(7) = 15656266201
my @primes = @{primes(100)};
my @primes_bellow = @{primes($P)};
sub non_residue {
my ($n) = @_;
foreach my $p (@primes) {
#if ($p > $Q) {
# return -1;
# }
sqrtmod($p, $n) // return $p;
}
return -1;
}
#~ #139309114031, a(9) <= 7947339136801, a(10) <= 72054898434289.
#~ say non_residue(139309114031, nth_prime(8));
#~ say non_residue(7947339136801, nth_prime(9));
#~ say non_residue(72054898434289, nth_prime(10));
#~ __END__
#~ say non_residue(72054898434289, nth_prime(10));
#~ __END__
sub isok {
my ($k) = @_;
my $t = $k-1;
my $u = $t>>1;
vecall { powmod($_, $u, $k) == $t } @primes_bellow;
}
#say isok(614796634515444067);
#say isok(8614572538322761627);
#392044843, 1568179369
#~ foreach my $p(factor(2152302898747)) {
#~ foreach my $d(divisors(($p-1)>>1)) {
#~ if (vecall {my $t = powmod($_, $d, $p); ($t == $p-1) } @primes_bellow) {
#~ say "$p : $d -> ", join(', ', divisors($d));
#~ }
#~ }
#~ say '';
#~ }
# Let a'(n) be the smallest odd prime p such that b^((p-1)/2) == 1 (mod p) for every natural b < prime(n).
# a(n) is the least number such that the n-th prime is the least quadratic nonresidue (not necessarily coprime) modulo a(n).
#~ __END__
#~ forprimes {
#~ if (is_prime(($_-1)*$multiplier + 1)) {
#~ my $k = Math::GMPz->new($_)*(($_-1)*$multiplier + 1);
#~ if (isok($k)) {
#~ die "a($N) <= $k\n";
#~ }
#~ }
#~ } 1e11, 1e12;
#~ __END__
sub fermat_pseudoprimes ($limit, $callback) {
say "Collecting primes...";
my %common_divisors;
forprimes {
my $p = $_;
my $common = 0;
foreach my $d (divisors(($p-1)>>1)) {
#if (powmod($P, $d, $p) == $p-1 ) {
if (vecall { powmod($_, $d, $p) == $p-1 } @primes_bellow) {
#push @{$common_divisors{$d}}, $p;
foreach my $g (divisors($d)) {
if ($g * 1000 >= $d) {
push @{$common_divisors{$g}}, $p;
}
}
}
}
#if (@div) {
# push @{$common_divisors{"@div"}}, $p;
#}
#~ if ($common) {
#~ foreach my $n(2..100) {
#~ if (is_prime(($p-1)*$n + 1)) {
#~ my $q = ($p-1)*$n + 1;
#~ foreach my $d (divisors($q - 1)) {
#~ if (powmod($P, $d, $q) == $q-1) {
#~ push @{$common_divisors{$d}}, $q;
#~ }
#~ }
#~ }
#~ }
#~ }
# }
} 3, $limit;
say "Generating combinations...";
#~ forprimes {
#~ my $p = $_;
#~ foreach my $d (divisors($p - 1)) {
#~ if (powmod($P, $d, $p) == $p-1) {
#~ push @{$common_divisors{$d}}, $p;
#~ }
#~ }
#~ } $limit>>1, $limit;
my %seen;
foreach my $value (values %common_divisors) {
my $arr = [uniqnum(@$value)];
my $l = $#{$arr} + 1;
if ($l > 100) {
say "Combinations: $l";
}
foreach my $k (2..$l) {
forcomb {
my $n = prod(@{$arr}[@_]);
if ($n <= $MAX) {
$callback->($n) #if !$seen{$n}++;
}
} $l, $k;
}
}
}
#~ sub is_fermat_pseudoprime ($n, $base) {
#~ powmod($base, $n - 1, $n) == 1;
#~ }
#~ sub is_fibonacci_pseudoprime($n) {
#~ (lucas_sequence($n, 1, -1, $n - kronecker($n, 5)))[0] == 0;
#~ }
my @values;
my %seen;
fermat_pseudoprimes(
$LIMIT,
sub ($k) {
# is_fermat_pseudoprime($n, 2) || die "error for n=$n";
#~ if (kronecker($n, 5) == -1) {
#~ if (is_fibonacci_pseudoprime($n)) {
#~ die "Found a special pseudoprime: $n = prod(@f)";
#~ }
#~ }
#~ if (powmod($P, ($k-1)>>1, $k) != $k-1) {
#~ say "false: $k";
#~ }
if (isok($k) and !$seen{$k}++) {
say "Found: $k";
push @values, $k;
}
# push @pseudoprimes, $n;
}
);
say "a($N) <= { ", join(', ', sort { $a <=> $b} @values), " }";
say "Min: a($N) <= ", vecmin(@values);