Why is C # the keyword return true for double but false for float, although casting for float works?

Question

Motivation: I have a method that returns a dynamic data type. The value comes from the database, and I know that this value will be either float, double, or string. I do not want to use the value if it is a string, so I wrote the following code:

if (value is float)
{
    myVariable = (float)value;
}

My expectation was that this code would execute whether the actual value type was double or floating due to the following snippet from the 'is' keyword documentation:

expression evaluates to true if the provided expression is not null and the provided object can be added to the provided type without causing an exception.

Found here: (C # link)

However, when the type is double, (value is float), returns false, and the assignment is not performed. But I changed my code to this:

if (value is double)
{
    myVariable = (float)value;
}

and it works fine when the value type is double - even if according to the documentation I can't use for float because (value is float) returns false.

My question is: Why (value is float) returns false when the value is double (which can be passed to float without exception)?

EDIT is a short program demonstrating

class Program
{
    static void Main(string[] args)
    {
        dynamic d = 1.0;
        Console.WriteLine(d is double);
        Console.WriteLine(d is float);
        float f = (float)d;
        Console.WriteLine(f);
        Console.ReadKey();
    }
}

Answer 1

That's right, that's because type dynamic. This basically means that the cast value floatdepends on the type of execution of the value.

is , float double - , false.

, double float, . MSDN # , - .

Answer 2

, , , . . ( #)

Answer 3

float. . , . :

if (value is double || value is float)
{
    myVariable = (float)value;
}