[go: up one dir, main page]

login
Revision History for A061341 (Bold, blue-underlined text is an addition; faded, red-underlined text is a deletion.)

Showing entries 1-10 | older changes
Numbers not ending in 0 whose cubes are concatenations of other cubes.
(history; published version)
#26 by Peter Luschny at Sun Aug 08 01:49:11 EDT 2021
STATUS

reviewed

approved

#25 by Sean A. Irvine at Sat Aug 07 00:26:59 EDT 2021
STATUS

proposed

reviewed

#24 by François Marques at Thu Jul 15 04:51:40 EDT 2021
STATUS

editing

proposed

#23 by François Marques at Thu Jul 15 04:51:31 EDT 2021
LINKS

François Marques, <a href="/A061341/b061341.txt">Table of n, a(n) for n = 1..100</a>

STATUS

proposed

editing

#22 by François Marques at Mon Jul 12 10:09:36 EDT 2021
STATUS

editing

proposed

#21 by François Marques at Mon Jul 12 09:56:01 EDT 2021
COMMENTS

Conjecture: the last digit of a(n) is in {1, 2, 3, 4, 5, 6, 8} and if it's 3 then there is k>=2 such that a(n) = 3*10^k+3.

- 1 then there is k>=3 such that a(n) = 6*10^k+1,

- 2 then a(n) = 22 or a(n) = 2272 or there is k>=4 such that a(n) = 15*10^k+12,

- 3 then there is k>=2 such that a(n) = 3*10^k+3.

PROG

(PARI) can_split(v, b=10, p=3)=if(#v==0, 1, for(k=1, #v, if(ispower(fromdigits(v[1..k], b), 3p) && can_split(v[k+1..#v], b, p), return(1))); return(0));

isok(n, b=10, p=3)=if(n%10b==0, return(0)); my(v=digits(n^3p, b)); for(k=1, #v-1, if(ispower(fromdigits(v[1..k], b), 3p) && can_split(v[k+1..#v], b, p), return(1))); return(0); \\ François Marques, Jul 11 12 2021

STATUS

proposed

editing

Discussion
Mon Jul 12
10:04
François Marques: I have changed the program for those who want to explore this kind of sequences for other bases or powers. It's interesting: in base 8 the cube of a value is a concatenation of at least 2 cubes if and only if it is even, for bases 5, 11, 13, 25 and 29 there seems to be no solution and for some other bases the solutions seem to be finite (only 3 for base 19; only 10 for base 31; only 1449462 for base 33).
10:06
François Marques: I will publish a b-file soon for at least the first hundred values.
#20 by Michel Marcus at Mon Jul 12 04:18:27 EDT 2021
STATUS

editing

proposed

Discussion
Mon Jul 12
09:52
François Marques: Thanks Marcus for the corrections but I am computing more terms and the conjecture must be changed : It's wrong : for values ending in 1 there are more patterns (like 375000001) ; ending in 2 : 6750000002 and also for 8 there exists the pattern 375000000008 (I don't find a smaller one) !
#19 by Michel Marcus at Mon Jul 12 04:18:03 EDT 2021
COMMENTS

- 1 then exists there is k>=3 such that a(n) = 6*10^k+1,

- 2 then a(n) = 22 or a(n) = 2272 or exists there is k>=4 such that a(n) = 15*10^k+12,

- 3 then exists there is k>=2 such that a(n) = 3*10^k+3.

#18 by François Marques at Sun Jul 11 13:50:38 EDT 2021
COMMENTS

From François Marques, Jul 11 2021: (Start)

If a and b are integers such that 3*a^2*b and 3*a*b^2 can be splited obtained as concatenations of cubes then a*10^k+b is a term in the list for k greater than the maximum of the number of digits in b^3, 3*a^2*b and 3*a*b^2.

Conjecture: the last digit of a(n) is in {1, 2, 3, 4, 5, 6} and if it's

- 1 then exists k>=3 such that a(n) = 6*10^k+1,

- 2 then a(n) = 22 or a(n) = 2272 or exists k>=4 such that a(n) = 15*10^k+12,

- 3 then exists k>=2 such that a(n) = 3*10^k+3.

(End)

Discussion
Sun Jul 11
13:57
François Marques: The conjecture is just an observation from the values. For values ending in 4, 5 or 6, there are several possible patterns. So far I have not found any other particular values such as 22 or 2272.
14:14
François Marques: For values ending with 1 there is a pattern starting from 375000001 ! So I will remove the part  of cases from the conjecture.
#17 by Jon E. Schoenfield at Sun Jul 11 12:14:40 EDT 2021
EXAMPLE

2272^3 = 11728027648 = 1_1728_0_27_64_8 (where the underscores indicate concatenation).

STATUS

reviewed

editing

Discussion
Sun Jul 11
12:15
Jon E. Schoenfield: (And note that "splited" is not a word in English.) :-)