memcpy 보다 std::copy가 더 빠르게 나오는데요

#include "stdafx.h"
 
#include <Windows.h>
#include <Mmsystem.h>
 
#include <vector>
#include <algorithm>
 
#include <iostream>
 
using namespace std;
 
const int MAX_LOOP = 1000000;
 
class CCGQueryPerformance 
{
public:
	   void CheckTimeStart()
	   {
	   	   _StartTime = timeGetTime();
	   }
 
	   void CheckTimeEnd()
	   {
	   	   _EndTime = timeGetTime();
	   }
 
	   double GetGapTime()	
	   {
	   	   return double(_EndTime - _StartTime)*0.001;
	   }
 
private:
	   DWORD _StartTime;
	   DWORD _EndTime;
};
 
 
int main()
{
	   typedef int TYPE; 
	   ///////////////////////////////
 
	   /////////////////////////////////
	   TYPE* source	       = new TYPE[MAX_LOOP];
	   TYPE* ArrayDst	     = new TYPE[MAX_LOOP];
 
	   vector<TYPE> VectorDst(MAX_LOOP);
 
	   for (int i = 0; i < MAX_LOOP; i++)
	   	   source[i] = 100;
 
	   /////////////////////////////////
 
 
	   {
	   	   CCGQueryPerformance	perfTest;
	   	   // Start) Timer 시작!
	   	   perfTest.CheckTimeStart();
 
	   	   for (int i = 0; i < 100; i++)
	   	   {
	   	   	   memcpy(ArrayDst,source, sizeof(TYPE)*MAX_LOOP);
	   	   }
 
	   	   // End) Timer 끝!
	   	   perfTest.CheckTimeEnd();
 
	   	   cout << "integer copy " << perfTest.GetGapTime() << endl;
	   }
 
 
	   {
	   	   CCGQueryPerformance	perfTest;
	   	   // Start) Timer 시작!
	   	   perfTest.CheckTimeStart();
 
	   	   for (int i = 0; i < 100; i++)
	   	   {
	   	   	   std::copy(source,source+MAX_LOOP,VectorDst.begin());
	   	   }
 
	   	   // End) Timer 끝!
	   	   perfTest.CheckTimeEnd();
 
	   	   cout << "vector copy : " << perfTest.GetGapTime() << endl;
	   }
}

   CCGQueryPerformance	perfTest;
	   	   // Start) Timer 시작!
	   	   perfTest.CheckTimeStart();
004012C2  call        dword ptr [__imp__timeGetTime@0 (4020A8h)] 
004012C8  mov         dword ptr [esp+10h],eax 
004012CC  mov         eax,64h 
 
	   	   for (int i = 0; i < 100; i++)
004012D1  dec         eax  
	   	   {
	   	   	   memcpy(ArrayDst,source, sizeof(TYPE)*MAX_LOOP);
004012D2  mov         ecx,0F4240h 
004012D7  mov         esi,ebx 
004012D9  mov         edi,ebp 
004012DB  rep movs    dword ptr [edi],dword ptr [esi] 
004012DD  jne         main+71h (4012D1h) 
	   	   }
 
	   	   // End) Timer 끝!
	   	   perfTest.CheckTimeEnd();
004012DF  mov         esi,dword ptr [__imp__timeGetTime@0 (4020A8h)] 
004012E5  call        esi  

CCGQueryPerformance	perfTest;
	   	   // Start) Timer 시작!
	   	   perfTest.CheckTimeStart();
00401335  call        esi  
00401337  mov         edi,dword ptr [esp+24h] 
0040133B  mov         dword ptr [esp+18h],eax 
0040133F  lea         edx,[ebx+3D0900h] 
00401345  mov         dword ptr [esp+10h],64h 
0040134D  lea         ecx,[ecx] 
 
	   	   for (int i = 0; i < 100; i++)
	   	   {
	   	   	   std::copy(source,source+MAX_LOOP,VectorDst.begin());
00401350  cmp         ebx,edx 
00401352  mov         eax,ebx 
00401354  je          main+10Ch (40136Ch) 
00401356  mov         ecx,edi 
00401358  sub         ecx,ebx 
0040135A  lea         ebx,[ebx] 
00401360  mov         ebp,dword ptr [eax] 
00401362  mov         dword ptr [ecx+eax],ebp 
00401365  add         eax,4 
00401368  cmp         eax,edx 
0040136A  jne         main+100h (401360h) 
0040136C  dec         dword ptr [esp+10h] 
00401370  jne         main+0F0h (401350h) 
	   	   }
 
	   	   // End) Timer 끝!
	   	   perfTest.CheckTimeEnd();
00401372  call        esi