(35s) y^x for y < 0 and x < 1 (complex roots)

[brianddk]

(04-29-2016 09:09 AM)Dieter Wrote:  According to this rule –8^(3/5) should work, but it doesn't => INVALID y^x.

I think it's much simpler: the error in your example occurs because you want the calculator to evaluate (–8)^0,333333333333.
But this does NOT equal (–8)^(1/3), so no (real) result exists.

That's why there is a XROOT function (x√y, on the K key). –8 [ENTER] 3 [x√y] returns –2, as expected. Here no errors occur for y<0 if x is an odd integer.

So instead of –8^(3/5) you can use –8^3 = –512 and then the 5th root of this yields –3,4822.

Your correct, I wasn't clear. What I meant was

Code:

Given: 
  x = n/d where d is odd
  y where y < 0

f(x,y) = abs((y+0i)^(n/d)) * -1    ; where n is odd
f(x,y) = abs((y+0i)^(n/d))         ; where n is even

The complex mantissa forces the matter, and ABS reflects it back to reals.