Math与StrictMath的区别

求一个角度的正弦函数，我们一般会想到Math类，其实还有一个类也可以实现这个我们的需求StrictMath

Math.sin(30.0d);
StrictMath.sin(30.0d);

通过两个方法都可以得到相同的结果

查看源代码 这两个实现方法有什么不一样呢？

StrictMath源代码

public static native double sin(double a);

sin方式是个本地方法，是靠底层代码实现

再来看看Math的sin方法

 public static double sin(double a) {
    return StrictMath.sin(a); // default impl. delegates to StrictMath
    }

通过源代码可以清晰的看到，其实Math类实现sin方法就是通过调用StrictMath类的native方法.

引用一段

Unlike   some   of   the   numeric   functions   of   class   StrictMath,   all   implementations   of   the   equivalent   functions   of   class   Math   are   not   defined   to   return   the   bit-for-bit   same   results.   This   relaxation   permits   better-performing   implementations   where   strict   reproducibility   is   not   required.  

By   default   many   of   the   Math   functions   simply   delegate   to   the   equivalent   functions   in   StrictMath   for   their   implementations.   Code   generators   are   encouraged   to   use   platform-specific   native   libraries   or   microprocessor   instructions,   where   available,   to   provide   higher-performance   implementations   of   Math   functions.   Such   higher-performance   implementations   still   must   conform   to   the   specification   for   Math.

所以要实现Math中现有的基本算术运算方法都可以用StrictMath类中的方法来替换