alexm
Another? You would be better off finding a first one:
>>> digitsets(5024)
n**3 126808653824
(n+1)**3 126884390625
s0 digitset ['0', '1', '2', '2', '3', '4', '5', '6', '6', '8', '8',
'8']
s1 digitset ['0', '1', '2', '2', '3', '4', '5', '6', '6', '8', '8',
'9']
Oops! Try using your toes next time.
Dittoopps! Well, does such a pair exist (using toes or fingers)?
Alexm
None less than n=10000000. You might have better luck if not
restricted
to adjacent integers such as you first example. But that's another
program.
>
> Alexm
It looks unlikely that n and n+1 will work - by your result. Thanks.
Alexm
If n^3 and (n+1)^3 have the same digits
(up to rearrangement), then these would
be congruent mod 9, hence congruent mod
3 in particular.
But (n+1)^3 - n^3 = 3n^2 + 3n + 1.
regards, chip
E.g. 5^3 and 8^3.
Some others: 355^3 and 427^3; 656^3 and 674^3; 3772^3 and 4036^3;
5852^3 and 6470^3; 8744^3 and 8864^3. There are many more.
There are some cheats, like 1001^3, 1010^3 and 1100^3. OR 1002^3,
1020^3, 2001^3 and 2010^3.
I think more interesting than adjacent integers (which Chip has showed
impossible) would be finding pairs that are far apart (farthest I find
for 4 digits is 9938^3 and 4661^3), or that have a simple ratio. For
example, are there any others that have the ration 8/5?
Or how many are 3 apart, like these: 1095^3 and 1098^3; 1991^3 and
1994^3; 3714^3 and 3717^3; 4173^3 and 4176^3; 4823^3 and 4826^3;
6473^3 and 6476^3.
Bob H
I noticed that one or more of Bob's examples
of 3 apart pairs has both numbers divisible
by 3, e.g. 1095^3 and 1098^3.
Thus the cubes 365^3 = 48627125 and 366^3 =
49027896 have different digit sets, but on
multiplication by 27 the digit sets become
equal.
This suggests a more general question: If
m,n are positive integers, when does k exist
s.t. mk and nk have equal digit sets?
regards, chip
Interesting.
Especially m=855 and n=236.
838 571 414
['2', '3', '3', '4', '6', '9']
['2', '3', '3', '4', '6', '9']
718 536 9261
['3', '4', '6', '6', '8', '9', '9']
['3', '4', '6', '6', '8', '9', '9']
291 205 2772
['0', '2', '5', '6', '6', '8']
['0', '2', '5', '6', '6', '8']
811 605 10179
['1', '2', '5', '5', '6', '8', '9']
['1', '2', '5', '5', '6', '8', '9']
370 414 675
['0', '2', '4', '5', '7', '9']
['0', '2', '4', '5', '7', '9']
342 570 321
['0', '1', '2', '7', '8', '9']
['0', '1', '2', '7', '8', '9']
855 236 44658
['0', '1', '2', '3', '5', '8', '8', '9']
['0', '1', '2', '3', '5', '8', '8', '9']
374 171 6192
['0', '1', '2', '3', '5', '8', '8']
['0', '1', '2', '3', '5', '8', '8']
275 446 46
['0', '1', '2', '5', '6']
['0', '1', '2', '5', '6']
588 447 2607
['1', '1', '2', '3', '5', '6', '9']
['1', '1', '2', '3', '5', '6', '9']
995 320 484
['0', '1', '4', '5', '8', '8']
['0', '1', '4', '5', '8', '8']
267 672 1
['2', '6', '7']
['2', '6', '7']
150 469 11493
['0', '1', '2', '3', '5', '7', '9']
['0', '1', '2', '3', '5', '7', '9']
744 923 7893
['2', '2', '3', '5', '7', '8', '9']
['2', '2', '3', '5', '7', '8', '9']
427 121 12
['1', '2', '4', '5']
['1', '2', '4', '5']
780 524 585
['0', '0', '3', '4', '5', '6']
['0', '0', '3', '4', '5', '6']
332 837 99
['2', '3', '6', '8', '8']
['2', '3', '6', '8', '8']
986 710 426
['0', '0', '2', '3', '4', '6']
['0', '0', '2', '3', '4', '6']
584 710 84
['0', '4', '5', '6', '9']
['0', '4', '5', '6', '9']
526 575 999
['2', '4', '4', '5', '5', '7']
['2', '4', '4', '5', '5', '7']
>
> regards, chip
[snip further examples]
Consider m = 1 and n = 2. One solution
for k is 241280512605, but this is not
minimal. What is the smallest k such
that k and 2k have equal digit sets?
It would seem that for m = 1 and n = 10
no k can possibly give equal digit sets.
However if we include leading zeros,
solutions are possible. Is there a pair
m,n > 0 for which no solution k exists?
regards, chip
> Consider m = 1 and n = 2. One solution
> for k is 241280512605, but this is not
> minimal. What is the smallest k such
> that k and 2k have equal digit sets?
Hint: The periodic decimal expansion of a
familiar rational provides integer k s.t.
k, 2k, 3k, 4k, 5k, and 6k all have equal
digit sets.
--c
If we're going to allow 3k and 4k I will ignore the answer you're
thinking of and say 153846. It is not the smallest but it's not far
behind.
Or I could go with 117647058823529 which works for 2k, 3k, ... 8k. Or
105263157894736842 (2k .. 9k). Or
1016949152542372881355932203389830508474576271186440677966.
Bob H
But k = 153846 has a different digit set
than 2k has.
> Or I could go with 117647058823529 which
> works for 2k, 3k, ... 8k. Or
If think you meant 1176470588235294.
> 105263157894736842 (2k .. 9k). Or
> 1016949152542372881355932203389830508474576271186440677966.
It's like you found a mysterious crank to
turn!!! ;-)
If a leading zero is allowed on k, we can
extend the pattern to 10k, 11k, etc. E.g.
k = 0588235294117647, then:
2k = 1176470588235294,
...
10k = 5882352941176470.
11k = 6470588235294117,
...
16k = 9411764705882352.
If it were known that 10 is primitive for
Z/pZ with arbitrarily large primes p, and
allowing leading zeroes on k, then for all
integer m,n > 0 there exists k s.t. mk and
nk have equal digit sets.
However this remains an open question:
[Artin's conjecture on primitive roots -- Wikipedia]
http://en.wikipedia.org/wiki/Artin%27s_conjecture_on_primitive_roots
regards, chip
Yep. I wasn't clear, but I meant "only" 3k and 4k.
> It's like you found a mysterious crank to turn!!! ;-)
Some would argue that *I* am a crank.
Of course this kind of number is familiar to many of us. Because of
that, I think it would be more interesting to *exclude* numbers whose
multiple is simply a cyclic permutation of itself. You demonstrated k
and 2k for k=241280512605, which is not cyclic. That is more
interesting, in my opinion.
Bob H
Trying my hand at that, I found a non-cyclic number using each digit
0-9 exactly once in k and 2k: 1934708526. Is this hard to do? I
can't see any obvious process to generate these.
Bob H
> Trying my hand at that, I found a non-cyclic number using each digit
> 0-9 exactly once in k and 2k: 1934708526. Is this hard to do? I
> can't see any obvious process to generate these.
Among the 3,265,920 pandigital numbers p (here pandigital in the
strict sense, that is numbers composed by 10 distinct digits),
exactly 184,320 have the property that 2*p is pandigital as well,
the smallest being
1023456789 x 2 = 2046913578 and the largest
4938271605 x 2 = 9876543210.
None of these 184,320 numbers have 2*p as a cyclic permutation of p.
I have summarized the search results for all the multiples
(k=2,...,9)
of pandigitals here: http://www.iread.it/panmult.php
There are also several (12,289) pandigital numbers with more than one
pandigital
multiple. The 2 champions (which are pandigital times {2,4,5,7,8})
are 1,098,765,432 and 1,234,567,890.
giovanni resta
--
http://anagram.it anagrams, alphametics, arithmogriphs, ask Dr.
Matrix, ...