mersenneforum.org

mersenneforum.org (https://www.mersenneforum.org/index.php)
-   PARI/GP (https://www.mersenneforum.org/forumdisplay.php?f=155)
-   -   PARI's commands (https://www.mersenneforum.org/showthread.php?t=13636)

science_man_88 2010-08-23 13:56

I never said [TEX]k*b!^n+1[/TEX] is of form 4x+1 you stated [TEX]k*b!^n/x^2[/TEX] for x>1= integer for some x.

this says that 4x+1 can be a [TEX]k*b!^n+1[/TEX]

and same for the rest so really I was just taking your last statement that the statements above and that statement were equal.

science_man_88 2010-08-23 13:59

[url]http://www.research.att.com/~njas/sequences/?q=3,5,7,9,11,13,17,19&sort=0&fmt=0&language=english&go=Search[/url]

was the best i could find with all of some group in one sequence.

science_man_88 2010-08-23 14:03

note also 65 = 2^6+1=2!^6+1 = 2!^3!+1

science_man_88 2010-08-23 14:15

one thing I notice is the first sequence of difference of your sequence pi is that they all seem to be powers of 2. can you confirm this ?

3.14159 2010-08-23 14:35

[QUOTE=science_man_88]one thing I notice is the first sequence of difference of your sequence pi is that they all seem to be powers of 2. can you confirm this ?
[/QUOTE]

3, 5: Diff = 2
5, 7: Diff = 2
7, 9: Diff = 2
9, 13: Diff = 2^2
13, 17: Diff = 2^2
17, 19: Diff = 2^2.
[B]19, 25: Diff = 6 =/= 2^n.[/B]

Strong law of small numbers at work.

science_man_88 2010-08-23 14:36

[QUOTE=3.14159;226658]3, 5: Diff = 2
5, 7: Diff = 2
7, 9: Diff = 2
9, 13: Diff = 2^2
13, 17: Diff = 2^2
17, 19: Diff = 2^2.
[B]19, 25: Diff = 6 =/= 2^n.[/B]

Strong law of small numbers at work.[/QUOTE]

okay sorry I forgot I took out 19 lol but the rest all have first difference of power of 2.

3.14159 2010-08-23 14:37

The sieve project is coming along.. Just that we have made zero progress.

Anyone have any spare ideas on it?

My best idea is to begin with the k-range loop, then a forprime loop which will be linked to a forstep loop, which, hopefully, will eliminate all the unwanted values. Then, it will save the remaining numbers in a file, and then I make the appropriate modifications to make the range testable, and then proceed to test.

science_man_88 2010-08-23 14:43

[CODE]for(k=,,forprime(_=,,forstep(-=,,_,write(,))))[/CODE]

something on the lines of this ?

3.14159 2010-08-23 14:44

Or: I trial-divide each candidate to 1048576, and test those which have no small factors.

[QUOTE=science_man_88]something on the lines of this ?
[/QUOTE]

Yes. Although I doubt it will work as I want it to.

3.14159 2010-08-23 14:46

The forstep loop will have to contain the liftmod for the primes, to eliminate bad values of k (Those which are divisible by a certain prime p.)

3.14159 2010-08-23 14:50

I can print the bad values that are divisible by a certain prime p, so far. But I can't get a good enough text editor to remove those specific values.

Ex: 757.

[code]623
1380
2137
2894
3651
4408
5165
5922
6679
7436
8193
8950
9707
10464
11221
11978
12735
13492
14249
15006
15763
16520
17277
18034
18791
19548
20305
21062
21819
22576
23333
24090
24847
25604
26361
27118
27875
28632
29389
30146
30903
31660
32417
33174
33931
34688
35445
36202
36959
37716
38473
39230
39987
40744
41501
42258
43015
43772
44529
45286
46043
46800
47557
48314
49071
49828
50585
51342
52099
52856
53613
54370
55127
55884
56641
57398
58155
58912
59669
60426
61183
61940
62697
63454
64211
64968
65725
66482
67239
67996
68753
69510
70267
71024
71781
72538
73295
74052
74809
75566
76323
77080
77837
78594
79351
80108
80865
81622
82379
83136
83893
84650
85407
86164
86921
87678
88435
89192
89949
90706
91463
92220
92977
93734
94491
95248
96005
96762
97519
98276
99033
99790
100547
101304
102061
102818
103575
104332
105089
105846
106603
107360
108117
108874
109631
110388
111145
111902
112659
113416
114173
114930
115687
116444
117201
117958
118715
119472
120229
120986
121743
122500
123257
124014
124771
125528
126285
127042
127799
128556
129313
130070
130827
131584
132341
133098
133855
134612
135369
136126
136883
137640
138397
139154
139911
140668
141425
142182
142939
143696
144453
145210
145967
146724
147481
148238
148995
149752
150509
151266
152023
152780
153537
154294
155051
155808
156565
157322
158079
158836
159593
160350
161107
161864
162621
163378
164135
164892
165649
166406
167163
167920
168677
169434
170191
170948
171705
172462
173219
173976
174733
175490
176247
177004
177761
178518
179275
180032
180789
181546
182303
183060
183817
184574
185331
186088
186845
187602
188359
189116
189873
190630
191387
192144
192901
193658
194415
195172
195929
196686
197443
198200
198957
199714[/code]

Those would be the bad k-values for k * 11![sup]4[/sup] + 1, because they form candidates divisible by 757.

Proof: 15006 * 11![sup]4[/sup] + 1 = 757 * 50325944549556534016411764332893


All times are UTC. The time now is 23:08.

Powered by vBulletin® Version 3.8.11
Copyright ©2000 - 2021, Jelsoft Enterprises Ltd.