Thread: Can anyone show me examples of these two questions.

One approach is to subscript the array in a loop over the elements of the array.




The other approach is to use a combination of advancing the pointer ptr thru the array and dereferencing it.

This is a typical performance technique, which avoids the calculation required to locate an element in an array through a subscript.

Code:

int a[1000];

//...assume a is populated with something

for( int n=0; n < 1000; ++n )
    {
      int z = a[ n ];
    }

In this case, n is referencing an element in the array, and in theory (that is, not accounting potential compiler optimization by recognizing this simple loop), there is a calculation of n * sizeof( int ) required for each reference in the array.

This is the same result, but the reference won't calculate n * sizeof( int ) for each access in the array;

Code:

int * ptr = a;
int * lim = ptr + 1000;

while( ptr < lim )
     {
      int z = *ptr;

      ++ptr;
     }

In the case the array refers to a struct, then the form ptr->m is often used.