Integer operation is easy to understand, that is, the operation between integers. As for floating-point numbers, you must first understand the definition of floating-point numbers: floating-point numbers logically mean that the decimal point is not fixed, while fixed-point numbers can only represent numerical values with fixed decimal points. Whether floating point number or fixed point number is used to represent a number depends on what meaning the user gives to the number. The operation between floating-point numbers is called floating-point operation. You can look here:/view/339796.htm-so, the understanding upstairs is actually wrong. Floating-point numbers can represent not only very large numbers, but also very small numbers, such as decimals or even negative numbers. In addition, floating-point numbers can be used to indicate the accuracy of numerical values. The reason why computers separate integer operations from floating-point operations should not be such a distinction. To be precise, they should separate addition and subtraction from multiplication and division. Floating-point operations are actually multiplication operations. In fact, CPU has only one addition operator, subtraction is to add a negative number, and division is to multiply its reciprocal, so the computer has only addition and multiplication, and multiplication is also realized by complex addition (I'm not sure), so the amount of calculation is particularly complicated, which can be used to test the operation performance of CPU.