( ( root
  =   n a d x0 x1 d2 rnd 10-d
    .   ( rnd       { For 'rounding' rational numbers = keep number of digits within bounds. }
        =   N r
          .   !arg:(?N.?r)
            & div$(!N*!r+1/2.1)*!r^-1
        )
      & !arg:(?n,?a,?d)
      & !a*!n^-1:?x0
      & 10^(-1*!d):?10-d
      &   whl
        ' (   ( rnd$(((!n+-1)*!x0+!a*!x0^(1+-1*!n))*!n^-1.10^!d)
              . !x0
              )
            : (?x0.?x1)
          & (!x0+-1*!x1)^2:~<!10-d   { Exit loop when required precision is reached. }
          )
      & flt$(!x0,!d)      { Convert rational number to floating point representation. }
  )
& ( show
  =   N A precision
    .   !arg:(?N,?A,?precision)
      & out$(str$(!A "^(" !N^-1 ")=" root$(!N,!A,!precision)))
  )
& show$(10,1024,20)
& show$(3,27,20)
& show$(2,2,100)
& show$(125,5642,20)
)