C# Mathematics Calculation not working properly

Question

OK, so I'm performing an annoying math computation here, trying to solve for one of the cubic roots.

Now, here is my C# code:

public void CubeCalculate()
{
    //Calculate discriminant

    double insideSquareRoot = (18 * cubicAValue * cubicBValue * cubicCValue * cubicDValue) + (-4 * (Math.Pow(cubicBValue, 3) * cubicDValue) + (Math.Pow(cubicBValue, 2) * Math.Pow(cubicCValue, 2)) + (-4 * cubicAValue * Math.Pow(cubicCValue, 3)) + (-27 * Math.Pow(cubicAValue, 2) * Math.Pow(cubicDValue, 2)));

    if (insideSquareRoot < 0)
    {
        //One real solution, two imaginary
        double onecuberootradical1 = (1 / 2) * (((2 * Math.Pow(cubicBValue, 3)) + (-9 * cubicAValue * cubicBValue * cubicCValue) + (27 * Math.Pow(cubicAValue, 2) * cubicDValue)) + (Math.Sqrt(Math.Pow((2 * Math.Pow(cubicBValue, 3)) + (-9 * cubicAValue * cubicBValue * cubicCValue) + (27 * Math.Pow(cubicAValue, 2) * cubicDValue), 2) + (-4 * Math.Pow(Math.Pow(cubicBValue, 2) + (-3 * cubicAValue * cubicCValue), 3)))));
        double onecuberootradical2 = (1 / 2) * (((2 * Math.Pow(cubicBValue, 3)) + (-9 * cubicAValue * cubicBValue * cubicCValue) + (27 * Math.Pow(cubicAValue, 2) * cubicDValue)) - (Math.Sqrt(Math.Pow((2 * Math.Pow(cubicBValue, 3)) + (-9 * cubicAValue * cubicBValue * cubicCValue) + (27 * Math.Pow(cubicAValue, 2) * cubicDValue), 2) + (-4 * Math.Pow(Math.Pow(cubicBValue, 2) + (-3 * cubicAValue * cubicCValue), 3)))));
        x1 = (-cubicBValue / (3 * cubicAValue)) + ((-1 / (3 * cubicAValue)) * (Math.Pow(onecuberootradical1, 1 / 3))) + (-1 / (3 * cubicAValue) * Math.Pow(onecuberootradical2, 1 / 3));
        x2 = double.NaN;
        x3 = double.NaN;
    }

OK, I'm trying to figure out what's going wrong here.

First of all, since this is part of an MVC application, I have made sure my other roots are working properly, so this is purely the fault of the following computation, not a problem from anywhere else.

Now, I've checked many times here and I have not found anything wrong.

You can compare to the proper formula here:

It is the x1 root that I'm trying to replicate here.

Also, if you would like to know the official discriminant form the same Wikiepdia article here it is:

Do you guys see anything wrong???

Answer 1

Right here is something obvious:

(1 / 2)

You are performing integer division there when you should be using a floating point number:

(1 / 2.0)

You don't show us the declaration of your cubic*Value variables, so I will assume that those are doubles.

Answer 2

This is or should be one of frequently asked question, and the answer is: What Every Computer Scientist Should Know About Floating-Point Arithmetic ( link )