Starting game programming

Originally posted by JoshG
What else does c++ add besides classes?

• Function prototyping
• Anonymous unions
• Constructors and destructors
• Exceptrions with try/catch blocks
• External function linkages
• Function overloading
• Inline functions
• Namespaces
• Bit Field types
• New and Delete operators
• Operator overloading
• Different STL
• Template classes
• Template functions

C++ is not just "C with classes"

Originally posted by JoshG
What else does c++ add besides classes?

I think that C++ is object-oriented programming.

a lot of those things you mentioned you can do in C. Like prototyping, or external functions.

Can you do type casting in C??

Originally posted by frenchfry164
a lot of those things you mentioned you can do in C. Like prototyping, or external functions.

heh, ok, well this was true for old-style C... it's not applicable anymore

Originally posted by Ruski

I think that C++ is object-oriented programming.

OOP is a methodology, or a way to structure and write a program; since an object is an intangible thing, you can create an "object" with just about any language you want, there is no constant "this is an object; this is not". C++ is considered an object-oriented language because it has built-in support for creating, managing, and using user-defined objects through the use of classes and all the goodies thereof (operator overloading)

make sense?

you can do OOP with C, just wasn't as good as what C++ has made it

I don't think you can do typecasting in C. You might be able to, correct me if I am wrong.

Anyone?? I asked a simple question

I'm fairly sure that you *can* do type casting in C...out of "C For Dummies, Volume I"

[QUOTE]
You can type cast any variable. Often the

Code:

(float)

is used as shown here when you're manipulating integer values. Another common case is when you're dealing with C language functions that require specific types of values. For example, the

Code:

sin

function computes the sine of an angle (yawn), but it requires a

Code:

double

value. If you only passed it an integer variable

Code:

x

, you would have to uyse the following statement:

Code:

result=sin((double)x);

Here, the integer variable

Code:

x

is converted into a double for processing by the

Code:

sin

function. The result is saved in the

Code:

result

varaible.[/PHP]

So, yes, you CAN do type casting in C.

ack, sorry guys, that looks ugly, I didn't realize it put the monospaced font on a seperate line every time ):

You can do type casting in C. I am just curious can you do type casting in c++ with an object? Say you have a class fu and an object thefu that is 12 bytes. Then say you have a class bar (12 bytes also) and a object thebar, could you (fu)thebar ?

BTW, I have no idea why you would want to do this, just wondering.

well try it yourself.

oh, and a reason why you would want to do it, is if you made a function that returns a fu, and you want to pass a thebar to it.

[QUOTE]Originally posted by blackrat364
I'm fairly sure that you *can* do type casting in C...out of "C For Dummies, Volume I"

You can type cast any variable. Often the

Code:

(float)

is used as shown here when you're manipulating integer values. Another common case is when you're dealing with C language functions that require specific types of values. For example, the

Code:

sin

function computes the sine of an angle (yawn), but it requires a

Code:

double

value. If you only passed it an integer variable

Code:

x

, you would have to uyse the following statement:

Code:

result=sin((double)x);

Here, the integer variable

Code:

x

is converted into a double for processing by the

Code:

sin

function. The result is saved in the

Code:

result

varaible.[/PHP]
Thanks For The Answer

So, yes, you CAN do type casting in C.

wow. thanks a lot for all the replies !

You can type cast object in C++. The typecasted object will have only the members that are shared between the 2, so it will be typecast as the "lesser" of the 2.

I do this in my current game. I have 3 similar objects (they're all actually derived from a single base class, but for simplicity here we'll treat them as wholly unique; still works the same).

Object 1 - CTrooper
Object 2 - CBot
Object 3 - CWall

All 3 classes have these members in common:
[code]
//variables
RECT rcSrc;
LPDIRECTDRAWSURFACE7 sfSurface;
POINT ptBlt;
//functions
void Clip(RECT rcViewable);
void Blt(LPDIRECTDRAWSURFACE7 sfDest);
[code]

CTrooper adds a bunch of variables and functions to allow for adjust health, speed, firing accuracy, courage, getting input, etc.

Code:

//some examples of CTrooper's unique members
int iCourage;
void AdjustCourage(int iAdjust);
int GetInput(long lBtn, long lLastBtn, POINT ptMouse);

CBot adds a bunch of variables and functions to allow for determining how "smart" it is, get firing accuracy, determining actions base on AI, etc.

Code:

//a few examples
int iIntelligence;
int GetAction(int iIncomingCommands);

CWall only has a few additional items like wall "health" etc (the game will allow for misplaced shots to damage terrain, etc.

Code:

//ditto
int iDamage;
void UpdateSrcForDamage();

I have an array of pointers to the lowest level object, in this case CWall. So, if I have a total of 3 of these objects in the game (that would be one of each) then I have an array of 3 CWall pointers.

Code:

CTrooper ctrprSam;
CBot cbotFred;
CWall cwallWall;
CWall *pcwall[3];

In my blitting code I first copy the addresses of each of my original objects into the array of CWall pointers.

Code:

pcwall[0]=&ctrprSam;
pcwall[1]=&cbotFred;
pcwall[2]=&cwallWall;

Then I loop through the array of CWall pointers and sort them based on their ptBlt.y value (the higher up the screen they are the sooner they should be blitted, so they don't cover up someone/thing who's in front of them), and then run the Blit routine.

Code:

pcwallWall->Blt(sfBack);

Here's a more simple, generic example using structs (same principles apply):

Code:

struct num1
{
    int i;
    long l;
    char c;    
};

struct num2
{
    int i;
    long l;
};

num1 nm1;
num2 nm2;
num2 *pnm2;

nm1.i=10;
nm1.l=3600001;
nm1.c='a';

pnm2=&nm1;

cout<<pnm2->i<<endl; //would output 10
cout<<pnm2->l<<endl; //would output 3600001