Had a new look at the problem. Reworked verbs as I'd not kept the
old. First pass, pretty brutal, took ~100 sec, in j806beta with avx.
Later efforts to speed it up were worse! Evidently better to compare
sorted character reps of numbers than sorted base 10 decodes.
M
Please reply to mike_liz....@tiscali.co.uk.
Sent from my iPad
> On 15 Mar 2017, at 07:48, 'Mike Day' via Programming
> <programm...@jsoftware.com> wrote:
>
> Only just up - having breakfast! I evidently solved it, as the project's
> discussion thread
> is open to me, but I don't have a record of my method.
>
> Early-ish PE problems were amenable to brute force, and I expect that's how
> I would have
> done it.
>
> AFAIrecall, later problems involving phi require insight how phi is
> calculated:
>
> if n = */p^q, phi(n) = */(p^(q-1)) * p-1
> eg
> q:360
> 2 2 2 3 3 5
> p =: 2 3 5
> q =: 3 2 1
>
> */(p^q-1) * p-1
> 96
>
> 5 p: 360
> 96
>
> so n/phi(n) is
> */@:(%<:) p
> 3.75
>
> No time for more - away from wifi for 10 hours or so!
> Mike
>
>
>
>> On 15/03/2017 03:09, Don Guinn wrote:
>> My second approach was brute force like you did. It gave the same number as
>> my first approach. And interestingly it was faster than my first approach.
>> Happens sometimes. The only thing I can think of is that I found the answer
>> but I didn't supply what they wanted. I supplied the first number that had
>> as a permutation of the digits given sorted by (%totient)n . I cheated.
>> Stole totient from J phrases.
>>
>> The only thing I can think of is that they wanted something other than the
>> n that I found.
>>
>> On Tue, Mar 14, 2017 at 8:12 PM, 'Jon Hough' via Programming <
>> programm...@jsoftware.com> wrote:
>>
>>> I just tried it and got the right answer. But my approach is essentially
>>> brute force:
>>> I basically stringified (":) the totient result, sorted it, and compared
>>> to the sorted stringified original number.
>>>
>>> I can be more specific if you like.
>>>
>>> Regards,
>>> Jon
>>> --------------------------------------------
>>> On Wed, 3/15/17, Don Guinn <dongu...@gmail.com> wrote:
>>>
>>> Subject: [Jprogramming] Project Euler
>>> To: "Programming forum" <programm...@jsoftware.com>
>>> Date: Wednesday, March 15, 2017, 9:37 AM
>>>
>>> Has anyone out there solved problem
>>> 70? I have worked it two ways which
>>> give the same answer but it is given as incorrect. I don't
>>> want to divulge
>>> what I did as that is against their rules. I must be missing
>>> something and
>>> presenting the wrong number for the result. Or is it
>>> possible that their
>>> answer is wrong?
>>>
>>> Glad to discuss it in the forum, but if anyone wants to
>>> contact me
>>> privately so we don't break Project Euler rules, contact me
>>> at
>>> dongu...@gmail.com
>>>
>>> Thanks.
>>> ----------------------------------------------------------------------
>>> For information about J forums see http://www.jsoftware.com/forums.htm
>>> ----------------------------------------------------------------------
>>> For information about J forums see http://www.jsoftware.com/forums.htm
>> ----------------------------------------------------------------------
>> For information about J forums see http://www.jsoftware.com/forums.htm
>
>
>
> ---
> This email has been checked for viruses by Avast antivirus software.
> https://www.avast.com/antivirus
> ----------------------------------------------------------------------
> For information about J forums see http://www.jsoftware.com/forums.htm
----------------------------------------------------------------------
For information about J forums see http://www.jsoftware.com/forums.htm
