Current location - Training Enrollment Network - Mathematics courses - What is machine zero? When floating-point numbers take the form of machine numbers, can all "0s" represent machine zeros?
What is machine zero? When floating-point numbers take the form of machine numbers, can all "0s" represent machine zeros?
The definition of floating point number is actually 0. Five x syndrome

*

2^XXXX

What is the power of zero times two?

How many zeros are mantissas and how many powers of 2 are rank codes. No matter how many digits a computer uses to represent rank codes, there must be a representation range. A * * * has four restrictions: negative maximum, negative minimum, positive maximum and positive minimum.

The minimum negative value (negative sign, maximum absolute value) is the question you ask, when the sequence code in a representation is negative, and this negative number is smaller than the sequence code that the computer can represent.

For example, in the XXX power of 2 above, XXX is negative (the absolute value is very large) and very small. Mathematically, this value is very close to zero. On the computer, because it cannot be expressed as less than this value, this floating point number is treated as zero.

Supplement: Never mind the digits, look at the ranking code first.

Suppose a number 10, 0.9999 *10-99999999999999999,

The power of 10 is-99999999999999, so is this number very close to 0?

Is the previous 0.9999 positive or negative?

The same is true of binary, 0. 1 1 1 1 * 2^- 1 1 1 1 1 1 1 1 1 1 1 / kloc-0/ 1 1 1 1 1 1 1 1 16 5438+0 1 1 1 1 1 1 1 1

The number of bits stored in a computer is limited, and there is also a range of the minimum number that can be represented. When a representation is less than the number it can represent, the computer can't represent it, so it is regarded as 0. In fact, this number is also very close to 0, mainly because of mathematical understanding.