Cross-My-Primes
Hello Math-Fun,
this game (hope not old hat) is played on a kind of infinite Scrabble board where all squares are white, and no square has any text on it. The starting "star" square, in the center of the grid, has a 7 on it (the number, not the word, as this game is played with digits only that will form numbers – instead of letters forming words).
The single player must now form prime numbers at every turn on the board – at the first turn placing the integer 1, at the second turn placing 2, then 3, then 4, then 5, etc. – this is the natural order of the positive integers as they appear (and because 7 starts the game, the player will jump from 6 to 8. No other jump will be allowed during the game).
Those integers, as in the traditional game of Scrabble, must be attached to the existing structure at least by a digit (one digit per square).
All visible numbers, before and after any turn, must be prime: they are read horizontally from left to right or vertically from top to bottom (again, like words in a traditional Scrabble game).
Example here:
We start the game with 7 on the star:
....................7....................
then comes 1 and we form the prime 71;
....................71....................
then comes 2 and we form the prime 271;
...................271....................
then comes 3 and we form the prime 3271;
..................3271....................
then comes 4 and we form the prime 43271;
.................43271....................
then comes 5 and we form the prime 53 (5 on top of 3 on the grid);
..................5.......................
.................43271....................
..................5..6.....................
.................43271.....................
..................8........................
..................5..6.....................
.................43271.....................
..................8........................
..................5..6.....................
.................43271.....................
...................9.......................
.....................1.....................
..................8..0.....................
..................5..6.....................
.................43271.....................
...................9.......................
now comes 11 – we have many choices here; we can place vertically 11 sticking tot he upper digit 1 of 1061 to form twice the prime 11 (horizontally and vertically) :
......................1....................
.....................11....................
..................8..0.....................
..................5..6.....................
.................43271.....................
...................9.......................
Etc. After a few discussions with friends, we've decided to forbid to split – as this is too much a help in building the structure. The other problem being that once a number is split, it is difficult to find it looking at a completed grid. This goes against the rules of Scrabble, we know. So, the placement of 11 hereunder is now forbidden, though 101 is a genuine prime:
...........................................
.....................1.....................
..................8.101....................
..................5..6.....................
.................43271.....................
...................9.......................
Is it possible to place, one after the other, starting with 7 in the "center" of the grid, the first 100 natural integers on it? If yes, is there a compact solution that beats the others?
It might be impossible to place the first 100 integers, I do not know – but in this case, is there another start than 7 on the "center star" that would allow that?
This suggests to start a new game every now and then with other primes than 7; one could start a new game with primes like 11, 13, 17 or 17623 (it seems to me that the game halts quickly with the starts 2, 3 and 5).
And now, perhaps, a sequence for the OEIS (any takers?):
We start we the empty board and the first prime on the "star":
...........................................
...........................................
...........................................
....................2......................
...........................................
...........................................
S = 2, ...We always want to extend S with the smallest integer not yet present in S that can be placed on the grid, according to the above rules; a(2) will be equal to 3 as no possible prime can be built with 1 and 2:
...........................................
...........................................
...........................................
....................23.....................
...........................................
...........................................
S = 2, 3, ...Now 1 can be placed (forming the prime 31, for instance):
...........................................
...........................................
...........................................
....................23.....................
.....................1.....................
...........................................
S = 2, 3, 1, ...Now 4 can be placed (forming the prime 431):
...........................................
...........................................
.....................4.....................
....................23.....................
.....................1.....................
...........................................
S = 2, 3, 1, 4 ...Now 5 can be placed (forming the prime 5431):
...........................................
.....................5.....................
.....................4.....................
....................23.....................
.....................1.....................
...........................................
S = 2, 3, 1, 4, 5, ...Now 6 cannot be placed (on this grid) – but 7 yes (forming the prime 75431 for instance):
...........................................
.....................7.....................
.....................5.....................
.....................4.....................
....................23.....................
.....................1.....................
...........................................
S = 2, 3, 1, 4, 5, 7, ...Now there is a place for 6 – which will form the prime 67:
...........................................
....................67.....................
.....................5.....................
.....................4.....................
....................23.....................
.....................1.....................
...........................................
S = 2, 3, 1, 4, 5, 7, 6, ... etc.Question:
What is the lexicographically earliest sequence of distinct positive terms fitting the "Cross-My-Primes" rules?
Best,
É.
__________________
Update of May 5th, 2020
Maximilian H. computed quickly the hereunder grid, starting on 2 and placing (without splitting them) the first 100 integers. Only 5 integers (in yellow) appear a bit later than they should – but not very late!
The sequence I was looking for is thus quite boring. Here are the first 37 terms of S (the next ones are 38, 39, 40, ... 98, 99, 100 with no "yellow inversion"):
S = 2,3,1,4,5,7,6,8,9,10,11,12,13,15,14,16,17,18,19,20,21,22,23,24,25,26,27,28,29,30,31,33,32,35,36,34,37,...
How was computed Maximilian's wonderful structure?
1) start with the prime 2;
2) always try to place on the grid the smallest integer not yet present;
3) any new integer must be linked to the previous structure by at least a single digit (like in the game of Scrabble);
4) always try to keep the structure compact;
5) splitting integers is forbidden.
Voilà!
[In French (merci Maximilian!) :
1° - placer toujours le plus petit nombre possible (on ne peut pas placer un plus petit, à n'importe quel instant) ; de plus :
2° - parmi ceux-là, ne considérer que ceux qui permettent le plus petit placement au coup suivant (par exemple, avec le 1 on peut faire 13 et 31, mais on retient le 31 permettant de placer le 4 au coup suivant, ce qui n'est pas possible en faisant un 13) ;
3° - parmi ceux là (minimal au coup n et aussi n+1) on choisit celui dont le "barycentre" (milieu) est le plus proche de l'origine (0,0), là où il y a le 2 bleu initial – pas du barycentre de la structure déjà produite) ;
4° - parmi ceux qui sont toujours "équivalents", on prend celui qui a les plus petits "square spiral numbers" (le max de ces nombres pour chaque case sur laquelle est placé un de ses chiffres doit être minimal).]
Under the structure is a table showing the primes produced by the placement of the successive natural numbers – see what 41 does!
Maximilian has used a very convenient notation to locate any integer k on the grid:
2 has coordinates [0,0], 3 = [1,0], 4 = [1,-1], 5 = [-1,0] and 6 = [-1,-1].
2 forms
2
|
2=[0,0]
|
3 forms 23
|
3=[1,0]
|
1 forms 31
|
1=[1,1]
|
4 forms 431
|
4=[1,-1]
|
5 forms 523
|
5=[-1,0]
|
7 forms 7523
|
7=[-2,0]
|
6 forms 67
|
6=[-2,-1]
|
8 forms 8431
|
8=[1,-2]
|
9 forms 84319
|
9=[1,2]
|
10 forms 109
|
10=[-1,2]
|
11 forms 11 and 167
|
11=[-2,-2]
|
12 forms 12109
|
12=[-3,2]
|
13 forms 113
|
13=[2,1]
|
15 forms 157523
|
15=[-4,0]
|
14 forms 14157523
|
14=[-6,0]
|
16 forms 163
|
16=[3,-1]
|
17 forms 17 and 11
|
17=[-1,-3]
|
18 forms 1811
|
18=[-4,-2]
|
19 forms 19 and 89
|
19=[2,-3]
|
20 forms 2011
|
20=[-1,-5]
|
21 forms 11321
|
21=[4,1]
|
22 forms 222011
|
22=[-1,-7]
|
23 forms 223
|
23=[-2,3]
|
24 forms 2484319
|
24=[1,-4]
|
25 forms 2512109
|
25=[-5,2]
|
26 forms 263
|
26=[-4,4]
|
27 forms 248431927
|
27=[1,3]
|
28 forms 263287
|
28=[-1,4]
|
29 forms 229
|
29=[2,3]
|
30 forms 30263287
|
30=[-6,4]
|
31 forms 11
and 31
|
31=[4,-2]
|
33 forms 233
|
33=[0,-6]
|
32 forms 32233
|
32=[-3,-6]
|
35 forms 351811
|
35=[-6,-2]
|
36 forms 3631
|
36=[4,-4]
|
34 forms 343631
|
34=[4,-6]
|
37 forms 23,
17 and 37
|
37=[4,2]
|
38 forms 383
|
38=[-6,-4]
|
39 forms 3919 and 29
|
39=[2,-5]
|
40 forms 401
|
40=[-4,-4]
|
41 forms 41, 541 and 1671223
|
41=[-3,1]
|
42 forms 4232233
|
42=[-5,-6]
|
43 forms 167122343
|
43=[-2,5]
|
44 forms 443
|
44=[-4,6]
|
45 forms 45351811
|
45=[-8,-2]
|
46 forms 463
|
46=[1,-8]
|
47 forms 47 and 47
|
47=[-3,-5]
|
48 forms 48463
|
48=[1,-10]
|
49 forms 24843192749
|
49=[1,5]
|
50 forms 503
|
50=[-8,-4]
|
51 forms 1151
|
51=[5,-1]
|
52 forms 521
|
52=[6,-3]
|
53 forms 953
|
53=[3,4]
|
54 forms 54521
|
54=[6,-5]
|
55 forms 5545351811
|
55=[-10,-2]
|
56 forms 56443
|
56=[-6,6]
|
57 forms 557
|
57=[4,4]
|
58 forms 5814157523
|
58=[-8,0]
|
59 forms 59 and 59
|
59=[0,6]
|
60 forms 60343631
|
60=[4,-8]
|
61 forms 461 and 3147
|
61=[-5,-5]
|
62 forms 622512109
|
62=[-7,2]
|
63 forms 115163
|
63=[7,-1]
|
64 forms 643
|
64=[8,-3]
|
65 forms 65222011
|
65=[-1,-9]
|
66 forms 66347
|
66=[-3,-8]
|
67 forms 3767
|
67=[6,2]
|
68 forms 685545351811
|
68=[-12,-2]
|
69 forms 95369
|
69=[3,6]
|
70 forms 7030263287
|
70=[-8,4]
|
71 forms 71,
1777 and 61
|
71=[5,3]
|
72 forms 727 and 7622512109
|
72=[-8,2]
|
73 forms 73, 571777 and 545213
|
73=[5,0]
|
74 forms 74545213
|
74=[6,-7]
|
75 forms 755814157523
|
75=[-10,0]
|
76 forms 764232233
|
76=[-7,-6]
|
77 forms 577, 37 and 67
|
77=[4,5]
|
78 forms 78643
|
78=[8,-5]
|
79 forms 79 and 79
|
79=[-1,7]
|
80 forms 809
|
80=[-3,8]
|
81 forms 6781
|
81=[5,6]
|
82 forms 823
|
82=[-7,-8]
|
83 forms 83 and 89
|
83=[2,7]
|
84 forms 8448463
|
84=[1,-12]
|
85 forms 858448463
|
85=[1,-14]
|
86 forms 86461
|
86=[-5,-8]
|
87 forms 376787
|
87=[8,2]
|
88 forms 887
|
88=[9,0]
|
89 forms 89 and 83
|
89=[1,8]
|
90 forms 6907 and 739
|
90=[7,0]
|
91 forms 691
|
91=[-5,7]
|
92 forms 92764232233
|
92=[-9,-6]
|
93 forms 55793 and 619
|
93=[6,4]
|
94 forms 941
|
94=[-7,8]
|
95 forms 9560343631
|
95=[4,-10]
|
96 forms 9666347
|
96=[-3,-10]
|
97 forms 97, 61991 and 37
|
97=[6,5]
|
98 forms 987030263287
|
98=[-10,4]
|
99 forms 8999, 5779 and 89
|
99=[4,7]
|
100 forms 10056443
|
100=[-9,6]
|
Bravo et merci, Maximilian, quelle merveille !
If a number might be inserted forming a prime with already existing digits, is it forbidden to do it if it would create a non-prime in some other way (by touching some other existing digits somewhere else)?
RépondreSupprimerI find that after the initial disorder (1 after 3 and 6 after 7), one can place each n at the n-th move, at least up to n=999, and that even in the greedy way (using the first of "usually" 2 possibilities, where my program checks first upward and leftward extensions).
SupprimerJust to clarify, the above comment concerns the variant with splitting, not the one without splitting mentioned in the part (update May 5) of this story, where many more "inversions" (numbers n placed only after move n) occur, depending on the criteria used to choose among different possible placements of the least possible number.
SupprimerIndeed, Maximilian, thanks!
RépondreSupprimer