Details about the mathematics class C #, Math, floating point number (medium)

The float and double Floating Point data types are defined to be consistent with ANSI/IEEE Std 754-1985 (an IEEE Standard for binary floating point Arithmetic.

A float value consists of a 24-Bit Signed ending number and an 8-Bit Signed index. The precision is greater than 7 digits after the decimal point. The value range is from

-The minimum float values from 3.402823*1038 to 3.402823*1038 greater than 0 are 1.401298*10-45.

Float. MinValue, float. MaxValue, float. Epsilon.

 

The double value consists of a 53-Bit Signed ending number and an 11-Bit Signed index. The precision is about 15 to 16 digits after the decimal point. The value range is from

-1.79769313486232*10308 to 1.79769313486232*10308. The minimum double value greater than 0 is 4.94065645841247*10-324, respectively:

Double. MinValue, double. MaxValue, double. Epsilon.

 

The following code divides a floating point number by 0;

Float f1 = 1;

Float f2 = 0;

Float f3 = f1/f2;

 

If you see the above Code, will you immediately think of DividedByZeroException. If these numbers are integers, an exception is thrown, but these numbers are floating-point numbers, and no exception is thrown here,

In fact, floating-point calculation never produces exceptions, because f2 is not really 0, it is just a number very close to 0 (For details, refer to my previous article ). In this case, f3 is equal to a special value. It is Infinity (positive Infinity ),

If you change f1 to-1, the f3 symbol is negative, and it is-Infinity (negative Infinity ).

You can even perform arithmetic operations on Infinity. For example, expression 1/f3 is equal to 0.

If f1 is changed to 0, f3 is equal to NaN (Not a Number, unknown ).

Determine whether a number is Infinity or NaN uses a Single static method:

Bool IsInfinity (float fValue); // determines whether it is infinite.

Bool IsPositiveInfinity (float fValue); // determines whether it is positive or not.

Bool IsNegativeInfinty (float fValue); // determines whether it is negative infinity.

Bool IsNaN (float fValue); // judge whether it is NaN

 

The Single structure also defines some constant fields.

Single. PositiveInfinity.

Single. NegativeInfinity.

Single. NaN.