Talk:Untouchable numbers: Difference between revisions
→Nice recursive solution
m (→Number of untouchable numbers up to 1 million: added a comment concerning the REXX output.) |
|||
Line 72:
N=None so I am done. 2 and 5 have been identified as untouchable. To find larger untouchables, probably time to turn to your poor long suffering computer. Where [[Talk:Numbers_with_prime_digits_whose_sum_is_13#Nice_recursive_solution]] may be considered 'Trees 101' This is more difficult to implement. It is described above as a breadth first search, but for for large values it may be better to use depth first. The main saving is in how well any implementation prunes the tree.--[[User:Nigel Galloway|Nigel Galloway]] ([[User talk:Nigel Galloway|talk]]) 14:55, 11 February 2021 (UTC)
:I was thinking along (vaguely) similar lines, seeing it spelt out like that clarified a couple of points:
:It is guaranteed to iterate through all composite numbers (as long as we carry on until Σ too big), and we have no interest in primes.
:Obviously it would be very silly to use trial division when we already have all the prime factors.
:However, I am sure there is an easy/clever way to calculate Σ, from those primes [and/or prev Σ?], but I just can't see it.
:Is that Σ=14 for Π=12 a typo, should it be 16 ie sum({1,2,3,4,6})? --[[User:Petelomax|Pete Lomax]] ([[User talk:Petelomax|talk]]) 08:03, 12 February 2021 (UTC)
| |||