Thread: Confused about Pointers

I am taking a beginners C course and we are starting to learn about Pointers. I am somewhat confused as to the value of them. My instructor said they make programming "fun", though they don't seem very fun to me I can't quite grasp it.
From what I can tell a "pointer" points to a address in the memory as part of it's "use". Ok, I got that. But for coding why use a Pointer, when you can just use an Array? Etc etc.

I am probably missing something obvious here, but Pointers just seem like a different way to go about it, albeit not necessary. I am sure I am wrong though. Can someone break it down into layman's terms? What is the value? And where would it be most value to use a Pointer?

This isn't for homework, it's for my own knowledge!

Thanks

Or if someone can point me to a good article about the use of Pointers, that would be great too.

Thanks!

An C array is implemented as a pointer; without the idea of pointers you could only return a single value from a function. With pointers/pass by reference you can change multiple non-global variables (outside the function) in a single function.

Tim S.

Originally Posted by stahta01

An C array is implemented as a pointer; without the idea of pointers you could only return a single value from a function. With pointers/pass by reference you can change multiple non-global variables (outside the function) in a single function.

Tim S.

Ahh ok, that sort of makes sense now. It makes your program more flexible without changing a specific function.

Thanks

There are many situations where "why not just use an array" is in fact the right answer.

Some of the reasons to use pointers...
1) You need to create and destroy arrays, structures, lists etc. "on the fly" inside your program. Linked lists are a good example of this.

2) You need to make room for something REALLY BIG like loading a 30 megabyte file. This would almost certainly crash your program if you tried to do it with char array[30000000];

3) You don't know in advance how big something needs to be. In these cases you use malloc() and free() to get around the problem of not knowing the size of an array at compile time.

Does that help?

somebody might be able to correct me if i am wrong but here is a basic understanding of it

computer memory is cmos technology( which is a type of transistor) which we could get into a whole debate about cmos technology but that is not the point of this topic

computer memory gets saved in stacks...as i am sure you teach has explained or drew the picture. so each variable gets saved into a specific memory location, and enless you make that variable static or pass by reference it will take what is previous in that memory location delete it

so if you had a variable called test and set that equal to zero so the code would be...

int test=0

the variable test is being saved into a spacific memory location, and lets say for the the conversation it is saved in memory location oxo1(hex 1) or ooo1 in binary so if you had a pointer in your code say *test it would point directly to that memory location(oxo1) and you could read the value of what it( for our case it is zero, or the aski equivalent of zero) otherwise you would have a bunch of garbage in that memory location

Originally Posted by begginer

somebody might be able to correct me if i am wrong but here is a basic understanding of it

It makes no difference, to a programmer, what technology memory uses. From the keyboard, we simply don't care about that. (but fwiw, ddr2 and ddr3 memory actually uses a matrix of capacitors that store tiny charges to represent the content of memory. Google "computer memory types" for more complete info.)

When you launch a program it is assigned a block of memory out of the system's pool of memory. This is enough to load the program image, create a stack and do any initial memory allocations for things where the size is known at compile time.

The stack is actually a region within the program's allocation set aside for temporary storage used by the CPU as it executes the code. When a procedure begins, the parameters and internal variables are "pushed" onto the stack where the procedure's code can have the CPU operate on them. When the function exits, it's space on the stack is relinquished and made available to the next function.

The "heap" is a global pool of memory shared by all programs. When we call malloc() or realloc() we are asking for a chunk of that memory for our own use. When we call free() we are returning that memory back to the heap.

A pointer can point to either the stack or the heap... depending on what kind of software legerdemain we are trying to pull at the moment. But the important thing to remember is that in modern systems we don't actually know the hardware address of that memory. It is "virtual memory" assigned as an offset from the program's entry point, not as a harware address.