[math-fun] Puzzle from Emissary about number shrinkage via digit summation
actually a more careful look at my nonrigorous argument last post suggests that in radix R, the number of steps needed is only O(logR).
Great! I've heard that Richard Stong says he can prove an absolute bound of 3 for all bases, but I don't know his idea. Victor On Wed, Nov 30, 2011 at 10:58 AM, Warren Smith <warren.wds@gmail.com> wrote:
actually a more careful look at my nonrigorous argument last post suggests that in radix R, the number of steps needed is only O(logR).
_______________________________________________ math-fun mailing list math-fun@mailman.xmission.com http://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun
Joe Buhler just pointed out to me that in bases bigger than 3, that the number 22'1' requires 3 steps to get to a single digit, where the digit x' means (b-x), where b is the base. So, for example in base 10, the number 289 requires 3 steps. Victor On Wed, Nov 30, 2011 at 11:23 AM, Victor Miller <victorsmiller@gmail.com>wrote:
Great! I've heard that Richard Stong says he can prove an absolute bound of 3 for all bases, but I don't know his idea.
Victor
On Wed, Nov 30, 2011 at 10:58 AM, Warren Smith <warren.wds@gmail.com>wrote:
actually a more careful look at my nonrigorous argument last post suggests that in radix R, the number of steps needed is only O(logR).
_______________________________________________ math-fun mailing list math-fun@mailman.xmission.com http://mailman.xmission.com/cgi-bin/mailman/listinfo/math-fun
participants (2)
-
Victor Miller -
Warren Smith