Sometimes the blog content will change. The first blog is the latest, and other blog addresses may not be synchronized. Check it carefully.
https://blog.zysicyj.top
First blog address [1]
Interview question manual [2]
Series article address [3]
1. What is Java floating point arithmetic?
In Java, floating-point operations refer to basic operations such as addition, subtraction, multiplication, and division of floating-point numbers. Java provides two floating point types: float and double.
2. Why are Java floating point operations inexact?
Inaccuracies in Java floating-point operations are mainly caused by the internal representation of floating-point numbers and limitations of computer hardware.
2.1 Internal representation of floating point numbers
Floating point numbers are represented in computers using binary scientific notation, which means that a real number is decomposed into two parts, the mantissa and the exponent, and is approximated by a binary number with a limited number of digits. For example, 0.1 cannot be represented exactly in binary, so there will be some error in the computer.
2.2 Computer hardware limitations
Computer hardware has certain limitations on the storage and calculation of floating point numbers. Typically, computers use fixed-length bytes to represent floating-point numbers, such as 32-bit or 64-bit. This means that floating point numbers have a limited number of significant digits, and any number beyond that number is truncated or rounded, introducing errors.
In addition, computers also need to perform rounding operations when processing floating point numbers to accommodate limited storage space. Rounding operations result in some loss of precision.
3. Implementation principles of Java floating point operations
Java's internal representation of floating point numbers uses the IEEE 754 standard, which defines the binary format and basic operation rules of floating point numbers. When performing floating point operations, Java processes floating point numbers in accordance with the IEEE 754 standard.
Specifically, Java uses signed, exponent, and mantissa bits to represent floating point numbers. Among them, the exponent bit is used to represent the magnitude of the floating point number, and the mantissa bit is used to represent the precision of the floating point number. By adjusting the values of the exponent and mantissa bits, floating point numbers of different ranges and precisions can be represented.
When performing floating-point operations, Java will select the corresponding operation rules based on the types of operators and operands. For example, the addition operation aligns the mantissas of two floating point numbers and compensates for the difference in exponent bits before adding them. This processing method can ensure the validity and correctness of the operation results.
4. Java 浮点运算的使用示例
下面是一个简单的 Java 浮点运算示例：
double a = 0.1;
double b = 0.2;
double c = a + b;
System.out.println(c);
输出结果为：
0.30000000000000004
上述代码中，由于 0.1 和 0.2 无法精确表示，所以在进行加法运算时会引入一定的误差，导致最终结果不是 0.3。
5. Java 浮点运算的优点
-
能够处理大范围和高精度的数值计算需求。 -
提供了标准化的浮点数表示方式和运算规则，保证了跨平台的兼容性。
6. Java 浮点运算的缺点
-
精度有限，可能存在舍入误差。 -
对于要求精确计算的场景（如金融领域），需要使用 BigDecimal 等其他数据类型来替代浮点数。
7. Java 浮点运算的使用注意事项
-
避免直接比较浮点数是否相等，应该使用误差范围判断。 -
在涉及到累加或累减操作时，尽量避免多次运算，可以先将所有操作数累加或累减后再进行运算，以减少舍入误差的积累。
8. 总结
Java 浮点运算不精确主要是由于浮点数的内部表示方式以及计算机硬件的限制所导致的。虽然存在一定的精度损失，但 Java 提供了标准化的浮点数表示方式和运算规则，能够满足大多数数值计算需求。在需要精确计算的场景下，可以使用 BigDecimal 等其他数据类型来替代浮点数。
参考资料
首发博客地址: https://blog.zysicyj.top/
[2]面试题手册: https://store.amazingmemo.com/chapterDetail/1685324709017001
[3]系列文章地址: https://blog.zysicyj.top/categories/技术文章/后端技术/系列文章/面试题精讲/
本文由 mdnice 多平台发布