OFFSET
0,2
COMMENTS
a(33) > 2*10^9; then sequence continues 62, 2275, 272, 22577, 118, 17, 40, 43, 7, 1339, 136, 25, 154, 143, 128, 125599, 34, 5619, 352, 1483, 18, 145, 8, 15457, 173, 14963, 60, 1727, 517, 1197, 1787456, 235, 642, 53, 116, ... - Robert G. Wilson v, Dec 14 2005
a(33) > 2*10^16. - Floris M. Velleman, Dec 17 2014
a(33) = 0. Modification of David W. Wilson's proof for A038369 shows that if a(33) > 0, then a(33) has at most 84 digits. This allows an exhaustive search of numbers of the form 2^a*3^b*5^c*7^d which shows that no such number exists. Other values of n for which a(n) is currently unknown and may be equal to 0 (based on analysis of numbers with at most 20 digits) are: 69, 111, 127, 146, 168, 172, 233, 243, 249, 273, 279, 281, 316, 327, 372, 533, 557, 579, 587, 621, 623, 647, 649, 676, 683, 713, 721, 816, 819, 821, 827, 861, 872, 917, 926, 927, 928, 939, 983, 996, 999, ... - Chai Wah Wu, Nov 22 2015
a(69) = a(111) = 0. To compute a(111), numbers of at most 85 digits were checked. - Chai Wah Wu, Dec 04 2015
LINKS
Chai Wah Wu, Table of n, a(n) for n = 0..126
Eric Weisstein's World of Mathematics, Sum-Product Number
MATHEMATICA
a[n_] := Block[{k = 1}, While[id = IntegerDigits@k; Abs[(Plus @@ id)(Times @@ id) - k] != n, k++ ]; k];
Table[ a[n], {n, 0, 54}] (* Robert G. Wilson v, Dec 14 2005 *)
PROG
(PARI) f(k) = my(d=digits(k)); abs(sum(j=1, #d, d[j])*prod(j=1, #d, d[j]) - k);
a(n) = {k = 1; while(f(k) != n, k++); k; } \\ Michel Marcus, Jan 02 2015
CROSSREFS
KEYWORD
nonn,base
AUTHOR
Eric W. Weisstein, Nov 28 2005
EXTENSIONS
Added a(33), edited definition and verified a(34)-a(68) by Chai Wah Wu, Nov 22 2015
STATUS
approved