Pointers are useful when you don''t know the size of something.. like strings. Strings can be any length.

You might use them for reading file data.. ie

char *data;

//Get filesize somehow

data = new char[filesize];

//read filedata into data


Someone else can add on to that

quote:Original post by asdasd12345
So far Ive been reading my C++ book and looking up tutorials online and I really dont understand the point of pointers. I mean what kind of situations would you really want to use them in. As far as I understand them pointers are variables that hold the address of another variable..or something like that. Any help would be helpful, apart from help that isnt helpful.

The worst part of most pointer lessons just involve pointers to int and stuff. That''s useful in some cases but really doesn''t help beginners understand why to use them.

Not sure if you know much about structs or classes but in a gaming situation if you have a Fighter struct or class it probably will store all sorts of info... name, hit points, strength, experience points, turn-ons... Lots of things basically. The size of a Fighter (all the info it stores) could be big, lets say its maybe 100 bytes total in this case.

Pointers are most useful when you want do something in a function with a big object. So then, in your game, you want your Fighter to do stuff like get stronger and have more hit points. So you have a function that calculates how much experience he gains after a fight. Since the calculations could be based on other current stats you have two basic options... pass by value or pass by pointer.

If you pass by value, you have your obese 100 byte Fighter make a 100 byte copy of itself for your function to work with and return that copy to whoever called it.

If you pass as a pointer, you are just telling your program that oh, my Fighter is over here (in memory), if you want to give him a gun, he''s right there, go hand it to him. Instead of passing 100 bytes, you pass 4 bytes (the size of a pointer) and you can mess with your Fighter''s stats directly. That''s 25 times less information your program has to shuffle around when you want to do a simple calculation with something big.

There are other uses but stuff like the example above hopefully explains the power of using pointers in actually programming. Ask more questions if this didn''t make any sense.

I've found the best way to learn things is to learn their applications. So try making some linked lists, and implement them in a practical program. Pointers are also useful for C-style strings. Dynamic arrays are allocated using pointers. All kinds of things, really...

my best advice is just to keep coding and coding. As your program becomes increasingly sophisticated, the methods you use to code it will too. Practice makes perfect - if you can't figure out a reason for using pointers in your current project then that's fine. Don't try to come to pointers, make them come to you - you will need them eventually, and when you do, you will know it.

edit: don't any of you know how to pass variables as reference parameters? Well I'm sure you do but i just haven't seen it mentioned... Pretty much eliminates the need for passing your big object into the function as a pointer.

[edited by - Odoacer on October 6, 2003 12:18:42 AM]

I use pointers mainly for dynamic memory, but thats a whole nother ball park I suppose.

Here''s an other example of pointer use. While there''s an alternative to pretty much all above examples (though quite a painful one, often!), I can''t think of any way to accomplish this without a pointer.

Writing directly into memory.

There are many, many applications of the above. Suppose you want an ultra-fast drawing routine, you can''t just rely on the GDI''s SetPixel() or SetPixelV() functions. You''d have to get down and dirty and write directly into memory, which is only possible either via a function provided by an API (see little note on how SetPixel just won''t do the trick for our example above) or via a pointer to the video ram. You can''t store the whole video ram into an array because the changes won''t be reflected anywhere: you need to tell your program WHERE to write stuff as well as WHAT to change. Hence the need for a pointer.