Question

<代码>Long.MaxValue>的价值9,223,372,036,854,775,807。达到单一精确值后,即为9.223372E18。然而, cast向一条龙卷土重来:<条码>。尽管独家企业的价值实际上低于长期可能的最高价值,但这种情况仍然存在。观察窗口中的实验表明,事实上,价值是

8.070450532247929344e18

is the largest double precision value that will safely ctype to a Long. further

8.070450807125836288e18

is the largest double precision value that after conversion to single precision, will safely convert to a Long.

这似乎同非常奇怪的行为一样。

为什么?

Answer 1

你们的问题主要是“在转换为浮动点和重新注入愤怒时,有些愤怒者没有四舍五入?” 而且,由于ISO 754格式固有的精准问题。 NET用于其浮动。

如果您,则该代码为。 (C#,但在VB中运行。) NET或其中任何一种。互联网支持的语言:

using System;

public class Program
{
    public static void Main()
    {
        Console.WriteLine((ulong)long.MaxValue);
        Console.WriteLine((ulong)(float)long.MaxValue);
    }
}

最后,

9223372036854775807
9223372036854775808

通知第二号是怎样的。如何做到这一点?

您可以安全地储存在浮动点号中的最大可能分类为:2^24,float和2^53,两倍(即其红外线的借方数目,加标线的1)。这符合你能够达到的最高精确度。不能准确代表更多的人,因此,你可以接近。事实上,根据:Wikipedia article on the subject:

Integers between 2^n and 2^(n+1) round to a multiple of 2^(n-23)

<代码>long.MaxValue过于庞大(其编号为<2^63-1<>>),因此没有准确储存。而是四舍五入到<代码>2^39/代码>的多个。页: 1 而且,这太大,难以长期适应,因此,在检查的情况下,你会例外。

If you put in 9223372036854775807 in a tool like this one, you ll see the value actually being stored is 9223372036854775808. Just as expected.

如果这确实是一个问题,你可以使用<代码>decimal ,该编码是四级的精度编号,可以安全地安装在<2>^-1<>>>>><113/code>以下的立体:

// This Just Works (tm)
Console.WriteLine((long)(decimal)long.MaxValue);

Or you could just not use floating point numbers.

Floating point numbers aren t broken, they re just built different.

Answer 2

当达到单一精确值时,<代码>Long.MaxValue即成为9.223372E18。

Although 9.223372E18 is the default string representation of the result, this is misleading. The exact result is not 9.223372 × 10¹⁸ but rather 2⁶³ = 9,223,372,036,854,775,808, which is the closest possible Single to the original value of 2⁶³ − 1. (The difference of 1 is due to the inability of Single to precisely represent the original value.)

由于2<⁶³超过Long.MaxValue,试图将其重新编为>。

友情链接