Let's call the first root a. So the second root is a2. Let's see if we can come up with a quadratic equation that has these two roots.
When you know the roots of a quadratic equation (call them a and b), you know that x-a and x-b are both factors of the quadratic polynomial. So you can say (x - a) (x - b) = 0.
But the polynomial can also be multiplied by some constant without changing the roots.
In this case, we have roots a and a2
That means that the equation can be written as k (x - a) (x - a2) = 0, where k is some constant.
That's the same as:
k (x2 - ax - a2x + a3) = 0
k (x2 - (a + a2)x + a3) = 0
kx2 - k(a + a2)x + ka3 = 0
Let's match this up with the original equation:
2x2 + bx + 16 = 0
Since the coefficient of x2 in the original equation is 2, and the coefficient of x2 in the equation we derived is k, we can say that it will probably work if we say that k = 2.
So now our derived equation is:
2x2 - 2(a + a2)x + 2a3 = 0
Now the constant term in the original equation was 16 and the constant term in our derived equation is 2a3. So lets set 2a3 equal to 16 and solve for a:
2a3 =16
a3 = 8
a = 2
So now we know that a is 2 and the other root, a2, is 4.
Finally, in our derived equation, the coefficient of x is -2(a + a2), and the coefficient of x in the original equation was b. So substitute 2 for a in 2(a + a2), and we will get the value of b:
b = -2(a + a2) = -2 (2 + 4) = -12
So b is -12, and the original equation was 2x2 - 12x + 16 = 0