[Scilab-users] Loop query

[Stefan Du Rietz]

On 2021-08-15 15:54, Samuel Gougeon wrote:
> Le 15/08/2021 à 11:28, Lester Anderson a écrit :
>> Hello Samuel,
>>
>> The size of ns (number of steps) and seq (sequence of values) are 
>> variable depending on the integer input, and this seems to be one issue.
> 
> For this reason, seq must be a list, leading to
> 
> function  [ns, seq] = collatz(p)
>      seq  =  p
>      while  %T
>          if  pmodulo(p, 2)
>              p  =  p*3+1
>          else
>              p  =  p/2
>          end
>          seq  =  [seq  p]
>          if  p==1
>              ns  =  length(seq)
>              break
>          end
>      end
> endfunction
> 
> prime = primes(20); [ns, seq] = ([],list()); for  i  =  1:length(prime)
>      [ns(i),  seq(i)]  =  collatz(prime(i));
> end --> ns' ans = 2. 8. 6. 17. 15. 10. 13. 21. --> seq seq = (1) = [2,1] 
> (2) = [3,10,5,16,8,4,2,1] (3) = [5,16,8,4,2,1] (4) = 
> [7,22,11,34,17,52,26,13,40,20,10,5,16,8,4,2,1] (5) = 
> [11,34,17,52,26,13,40,20,10,5,16,8,4,2,1] (6) = 
> [13,40,20,10,5,16,8,4,2,1] (7) = [17,52,26,13,40,20,10,5,16,8,4,2,1] (8) 
> = [19,58,29,88,44,22,11,34,17,52,26,13,40,20,10,5,16,8,4,2,1]
> 
> Samuel
> 
Of course, Samuel, I didn't realize that the length of seq was not 
monotonically increasing (which I had if I had looked at the displayed 
matrix ...)!

Stefan