我已经阅读了所有的答案(超过30个)，并没有找到一个简单的原因:如果你读过并练习过Intel®64和IA-32架构优化参考手册，汇编程序比C更快，所以汇编程序可能更慢的原因是编写这种慢汇编程序的人没有阅读优化手册。

In the good old days of Intel 80286, each instruction was executed at a fixed count of CPU cycles. Still, since Pentium Pro, released in 1995, Intel processors became superscalar, utilizing Complex Pipelining: Out-of-Order Execution & Register Renaming. Before that, on Pentium, produced in 1993, there were U and V pipelines. Therefore, Pentium introduced dual pipelines that could execute two simple instructions at one clock cycle if they didn't depend on one another. However, this was nothing compared with the Out-of-Order Execution & Register Renaming that appeared in Pentium Pro. This approach introduced in Pentium Pro is practically the same nowadays on most recent Intel processors.

Let me explain the Out-of-Order Execution in a few words. The fastest code is where instructions do not depend on previous results, e.g., you should always clear whole registers (by movzx) to remove dependency from previous values of the registers you are working with, so they may be renamed internally by the CPU to allow instruction execute in parallel or in a different order. Or, on some processors, false dependency may exist that may also slow things down, like false dependency on Pentium 4 for inc/dec, so you may wish to use add eax, 1 instead or inc eax to remove dependency on the previous state of the flags.

如果时间允许，您可以阅读更多无序执行和注册重命名。因特网上有大量的信息。

There are also many other essential issues like branch prediction, number of load and store units, number of gates that execute micro-ops, memory cache coherence protocols, etc., but the crucial thing to consider is the Out-of-Order Execution. Most people are simply not aware of the Out-of-Order Execution. Therefore, they write their assembly programs like for 80286, expecting their instructions will take a fixed time to execute regardless of the context. At the same time, C compilers are aware of the Out-of-Order Execution and generate the code correctly. That's why the code of such uninformed people is slower, but if you become knowledgeable, your code will be faster.

除了乱序执行之外，还有很多优化技巧和技巧。请阅读上面提到的优化手册:-)

However, assembly language has its own drawbacks when it comes to optimization. According to Peter Cordes (see the comment below), some of the optimizations compilers do would be unmaintainable for large code-bases in hand-written assembly. For example, suppose you write in assembly. In that case, you need to completely change an inline function (an assembly macro) when it inlines into a function that calls it with some arguments being constants. At the same time, a C compiler makes its job a lot simpler—and inlining the same code in different ways into different call sites. There is a limit to what you can do with assembly macros. So to get the same benefit, you'd have to manually optimize the same logic in each place to match the constants and available registers you have.

Actually you can build large scale programs in a large model mode segaments may be restricted to 64kb code but you can write many segaments, people give the argument against ASM as it is an old language and we don't need to preserve memory anymore, If that were the case why would we be packing our PC's with memory, the only Flaw I can find with ASM is that it is more or less Processor based so most programs written for the intel architecture Most likely would not run on An AMD Architecture. As for C being faster than ASM there is no language faster than ASM and ASM can do many thing's C and other HLL's can not do at processor level. ASM is a difficult language to learn but once you learn it no HLL can translate it better than you. If you could only see some of the things HLL's Do to you code, and understand what it is doing, you would wonder why More people don't use ASM and why assembers are no longer being updated ( For general public use anyway). So no C is not faster than ASM. Even experiences C++ programmers still use and write code Chunks in ASM added to there C++ code for speed. Other Languages Also that some people think are obsolete or possibly no good is a myth at times for instance Photoshop is written in Pascal/ASM 1st release of souce has been submitted to the technical history museum, and paintshop pro is written still written in Python,TCL and ASM ... a common denominator of these to "Fast and Great image processors is ASM, although photoshop may have Upgraded to delphi now it is still pascal. and any speed problems are comming from pascal but this is because we like the way programs look and not what they do now days. I would like to make a Photoshop Clone in pure ASM which I have been working on and its comming along rather well. not code,interpret,arange,rewwrite,etc.... Just code and go process complete.

我不能给出具体的例子，因为那是很多年前的事情了，但是在很多情况下，手工编写的汇编程序可以胜过任何编译器。原因:

您可以偏离调用约定，在寄存器中传递参数。 您可以仔细考虑如何使用寄存器，避免将变量存储在内存中。 对于跳转表之类的东西，可以避免检查索引的边界。

基本上，编译器在优化方面做得很好，这几乎总是“足够好”，但在某些情况下(如图形渲染)，你要为每一个周期付出高昂的代价，你可以走捷径，因为你知道代码，而编译器不能，因为它必须在安全的方面。

事实上，我听说过一些图形渲染代码，其中一个例程，如直线绘制或多边形填充例程，实际上在堆栈上生成了一小块机器代码并在那里执行，以避免关于线条样式、宽度、模式等的连续决策。

也就是说，我想让编译器为我生成好的汇编代码，但又不太聪明，它们通常都是这样做的。事实上，我讨厌Fortran的一个原因是它为了“优化”而打乱代码，通常没有什么重要的目的。

通常，当应用程序出现性能问题时，都是由于浪费的设计造成的。这些天，我永远不会推荐汇编程序的性能，除非整个应用程序已经在它的生命周期内进行了调优，仍然不够快，并且把所有的时间都花在了紧凑的内部循环中。

补充:我见过很多用汇编语言编写的应用程序，与C、Pascal、Fortran等语言相比，汇编语言的主要速度优势是因为程序员在用汇编语言编码时要谨慎得多。他或她每天要写大约100行代码，不管哪种语言，在编译器语言中，这将等于3或400条指令。

几乎任何时候编译器看到浮点代码，如果你使用的是旧的糟糕的编译器，手写的版本会更快。(2019年更新:对于现代编译器来说，这并不普遍。特别是在编译x87以外的东西时;编译器更容易使用SSE2或AVX进行标量数学运算，或任何具有平面FP寄存器集的非x86，不像x87的寄存器堆栈。)

主要原因是编译器不能执行任何健壮的优化。关于这个主题的讨论，请参阅来自MSDN的这篇文章。下面是一个例子，其中汇编版本的速度是C版本的两倍(用VS2K5编译):

#include "stdafx.h"
#include <windows.h>

float KahanSum(const float *data, int n)
{
   float sum = 0.0f, C = 0.0f, Y, T;

   for (int i = 0 ; i < n ; ++i) {
      Y = *data++ - C;
      T = sum + Y;
      C = T - sum - Y;
      sum = T;
   }

   return sum;
}

float AsmSum(const float *data, int n)
{
  float result = 0.0f;

  _asm
  {
    mov esi,data
    mov ecx,n
    fldz
    fldz
l1:
    fsubr [esi]
    add esi,4
    fld st(0)
    fadd st(0),st(2)
    fld st(0)
    fsub st(0),st(3)
    fsub st(0),st(2)
    fstp st(2)
    fstp st(2)
    loop l1
    fstp result
    fstp result
  }

  return result;
}

int main (int, char **)
{
  int count = 1000000;

  float *source = new float [count];

  for (int i = 0 ; i < count ; ++i) {
    source [i] = static_cast <float> (rand ()) / static_cast <float> (RAND_MAX);
  }

  LARGE_INTEGER start, mid, end;

  float sum1 = 0.0f, sum2 = 0.0f;

  QueryPerformanceCounter (&start);

  sum1 = KahanSum (source, count);

  QueryPerformanceCounter (&mid);

  sum2 = AsmSum (source, count);

  QueryPerformanceCounter (&end);

  cout << "  C code: " << sum1 << " in " << (mid.QuadPart - start.QuadPart) << endl;
  cout << "asm code: " << sum2 << " in " << (end.QuadPart - mid.QuadPart) << endl;

  return 0;
}

和一些数字从我的PC运行默认版本*:

  C code: 500137 in 103884668
asm code: 500137 in 52129147

出于兴趣，我用dec/jnz交换了循环，它对计时没有影响——有时更快，有时更慢。我想内存有限的方面使其他优化相形见绌。(编者注:更可能的情况是，FP延迟瓶颈足以隐藏循环的额外成本。对奇数/偶数元素并行进行两个Kahan求和，并在最后添加它们，可能会加快2倍的速度。)

哎呀，我正在运行一个稍微不同的代码版本，它输出的数字是错误的(即C更快!)修正并更新了结果。

不需要给出任何具体的示例或分析器证据，当您比编译器知道的更多时，您可以编写比编译器更好的汇编程序。

In the general case, a modern C compiler knows much more about how to optimize the code in question: it knows how the processor pipeline works, it can try to reorder instructions quicker than a human can, and so on - it's basically the same as a computer being as good as or better than the best human player for boardgames, etc. simply because it can make searches within the problem space faster than most humans. Although you theoretically can perform as well as the computer in a specific case, you certainly can't do it at the same speed, making it infeasible for more than a few cases (i.e. the compiler will most certainly outperform you if you try to write more than a few routines in assembler).

另一方面，有些情况下编译器没有那么多的信息——我想说主要是在使用不同形式的外部硬件时，编译器不知道这些信息。主要的例子可能是设备驱动程序，其中汇编程序结合人类对相关硬件的熟悉知识可以产生比C编译器更好的结果。

其他人提到了特殊用途指令，这就是我在上面一段中所说的——编译器可能对这些指令了解有限或根本不了解，这使得人类可以编写更快的代码。

在Amiga上，CPU和图形/音频芯片会为了访问特定区域的RAM(具体来说是前2MB的RAM)而争斗。因此，当你只有2MB RAM(或更少)时，显示复杂的图形加上播放声音会杀死CPU的性能。

在汇编程序中，你可以巧妙地交错你的代码，使CPU只在图形/音频芯片内部繁忙时(即当总线空闲时)才尝试访问RAM。因此，通过重新排序指令，巧妙地使用CPU缓存，总线定时，你可以实现一些使用任何高级语言都不可能实现的效果，因为你必须为每个命令定时，甚至在这里或那里插入nop，以使不同的芯片不受彼此的雷达影响。

这也是为什么CPU的NOP (No Operation -什么都不做)指令实际上可以让你的整个应用程序运行得更快的另一个原因。

当然，这种技术取决于特定的硬件设置。这就是为什么许多Amiga游戏无法适应更快的cpu的主要原因:指令的计时错误。