use of a 2-d array

[tubby123]

guys, we use like a 2-d array very often.
But what exactly is the use of a 2-d array ?

Now when i look at it, i feel everything can be done using a 1-d array and incrementing the pointer appropriately.
Is the only advantage of a 2-d array , matrix representation ? (By the way, matrices are very important, i understand cuz its matrix multiplication that happens in GPU's)

But can someone give me another example, or induce a different kind of thought into this.

Thanks.

[Narue]

Now when i look at it, i feel everything can be done using a 1-d array and incrementing the pointer appropriately.

Just because you can do something doesn't mean it's always the best way.

[tubby123]

Just because you can do something doesn't mean it's always the best way.

@Narue: So can u give an example where you compulsarily have to use a 2-d array.

[Narue]

There's no such situation. You can always simulate the effect of a 2D array without actually using one. At this point, I have no idea what you're looking for aside from imaginary guarantees that save you from making an informed decision in your own code.

[tubby123]

There's no such situation. You can always simulate the effect of a 2D array without actually using one. At this point, I have no idea what you're looking for aside from imaginary guarantees that save you from making an informed decision in your own code.

I have a couple of doubts and trying to solve these gave rise to those.
1)Why is the syntax for function definition like this ?

int fun_name(int array[])
{ 

}

You are passing the address of the first character of the array. How is that stored in int array[] ?

2)If you can receive a 1-d array like mentioned above(in 1), then why cant you receive a 2-d array as

int fun_name(int array[][])
{

}

I have found in text-books and resources that this is rather the syntax
int fun_name(int array[])

{

}

My argument : All that you need to access any element of an array is just its starting address cuz array is contiguous memory allocation. So y shud u pass the size then ?

[Narue]

1) Array syntax for the first dimension is sugar for a pointer. These two are equivalent:

int fun_name(int array[])

int fun_name(int *array)

So array isn't actually an array, it's a pointer. Personally, I think arrays and pointers would be somewhat less confusing if this particular feature didn't exist.

2) Because only the first dimension enjoys the conversion to a pointer. The syntax falls out of the pointer conversion of an array name used in value context. Passing a bare array name as an argument is value context, so the conversion for this:

int fun_name(int array[][20]);
int array[10][20];

fun_name(array);

is conceptually like so:

int fun_name(int array[][20]);
int array[10][20];
int (*__temp)[20] = &array[0];

fun_name(__temp);

The usefulness of applying that conversion recursively to all dimensions is dubious at best.

So y shud u pass the size then ?

That's easy: because you can't not pass the size. This is a syntactical rule of C. Whether you agree with the decision or not, it's set in stone at this point.

[rockerjhr]

I used a 2 d array of characters to make the board of my four in a row program

made a state struct that had different fields including a brd field which was a board a 2d array then that had the configuration of every move in it and stored each state in an N-ary tree so that my heuristic function could make a good calculation of what the user's move was going to be