Understanding Pointers in C++

Pointers have the fame of being hard to understand. However, this is not true. Pointers are simple to comprehend. What makes them confusing is how powerful and flexible they can be.

Imagine your computer's entire memory as a long row of containers; where each container has a unique address and can store a value.

A pointer is just container that has a unique address but instead of storing a value it stores the address of a variable. For example, the image below shows the address of the variable myChar stored inside the container myPointer.

The key point is this: A variable stores a value. A Pointer stores the Address of the variable.

Defining a Pointer

A Declaration announces a variable or function to the compiler so that it knows the data type before it compiles it. A Definition is a declaration that reserves storage.

So in the following code:

int main(){

char myChar='b'; //define variable myChar

return 0;
}

The variable myChar is being declared and defined. That is, memory space has been reserved for the variable, and it contains the value: 'b.'

To define a pointer, you specify the data type your pointer will be holding the address of and include the asterik symbol "*" before the pointer's name.

For example:

int main(){

char myChar='b'; //1. define variable myChar

char *myPointer;  //2. pointer definition

return 0;
}

Assigning an address to a pointer

To assign the address of a variable to a pointer, you use the "&" symbol as shown in line 3 below. In C++, the symbol "&" means "the address of."

int main(){

char myChar='b'; //1. define variable myChar

char *myPointer;  //2. pointer definition

myPointer=&myChar; //3. assigning the address of myChar to the pointer myPointer

return 0;
}

If you were to print the value of "myPointer" you will get the address of the variable "myChar."

Dereferencing a Pointer

You can retrieve the value a pointer points to by dereferencing it.

Let's go through a quick example:

int main(){

char myChar='b'; //1. define variable myChar

char myOtherChar; //2. define a variable myOtherChar

char *myPointer;  //3. pointer definition

myPointer=&myChar; //4. assigning the address of myChar to the pointer myPointer

myOtherChar=*myPointer;  //5. dereferencing myPointer

//myOtherChar now holds the value 'b'

return 0;
}

In line 5, we dereference the pointer using the dereferencing operator "*".

In other words, we accessed what the pointer points to, i.e., "myChar" and assigned the value of "myChar" to the variable "myOtherChar."

The dereferencing operator "*" is the same asterisk used to declared a pointer. However, its behavior differs depending on the position of the asterisk in a program. When the asterisk is not used in a declaration, it means "access what the pointer points to." I know this is confusing, but that is how it works in C++.

Hope this helps

Understanding Data Types in C++

How you represent an integer, a letter or a word in C++ differs. For example, representing the integer 23 requires the keyword int before the variable name:

int myNumber=23;

Representing the letter "L" requires the keyword char before the variable name:

char myLetter='L';

The keywords char and int along with float, bool and double are known as Data Types. A Data Type defines the set of possible values allowed in a variable and its memory size. These five data types constitute the fundamental data types in C++.

The Basic Integer Type: int

To represent an integer such as 201, -33, 1024, etc., C++ requires the keyword int before the variable name. For example:

int numberA=201;
int numberB=-33;
int numberC=1024;

The Floating Number Type: float

To represent a decimal number such as 21.3, 0.42, 0.10, etc., you must use the keyword float before the variable name. For example:

float numberA=21.3;
float numberB=0.42;
float numberC=0.10;

The Extended Precision Type: double

The double data type is similar to float but can store roughly twice as many significant digits than a float data type. To represent numbers such as -243.16, 2453345.2, C++ requires the keyword double before the variable name. For example:

double numberA=-243.16;
double numberB=2453345.2;

The Single Character Type: char

To represent a character such as 'a,' 'B,' '9', etc., use the keyword char. For example:

char myCharacterA='a';
char myCharacterB='B';
char myCharacterC='9';

The Boolean Data Type: bool

To represent the value of 0 or 1. Or true or false, use the keyword bool. For example:

bool myValueA=false;
bool myValueB=true;

Also, a data type specifies the memory allocated for the variable. The memory allocated is machine dependent, and it varies from machine to machine. For example, on my Mac:

• An int takes 4 bytes of memory
• A char takes 1 byte of memory
• A float takes 4 bytes of memory
• A double takes 8 bytes of memory
• A bool takes 1 byte of memory

Hope this helps.

Understanding how Functions work in C++

In computer science, a function is a set of statements that together perform an action. The most important function in a C++ program is main().

int main(){}

When you run an app, the computer automatically looks for main() and executes the statements within its body. For example, main() can have several functions such as run, walk, stop that will execute when the app is open:

 

int main(){
    run(); //call function run
    walk(); //call function walk
    stop(); //call function stop
}

Note: The double slash, //, defines the beginning of a comment. A comment is for readability only. Your program ignores them.

In C++, a function consists of a:

• Return type: The value a function may return. For example, it can return an integer, string, etc. A function may also return Nothing.
• Arguments: Input data to the function. The input data can be an integer, string, boolean, etc.
• Function Name: The actual name of the function.
• Function Body: A collection of statements that define what the function does.

Function Declaration

Before main() can call a function, the function must be declared. A declaration introduces a function to the program. A function declaration provides the function's name, return type, and parameters.

In a function declaration, the return type comes before the function's name. The parenthesis after the name encloses the function's arguments.

For example:

a) The return type of the function below is an int (integer). The name of the function is add. And it has two arguments x and y of type int.

int add(int x, int y);

b) The return type of the function below is void. Meaning the function does not return any value. The name of the function is print. And its argument is of type string.

void print(string x);

c) The return type of the function below is void. Its name is run, and it does not contain any arguments:

void run();

Function Definition

Whereas a declaration provides the function's return type, name, and arguments, a definition provides the set of tasks in a function's body. For example:

int add(int x, int y){
    int z=x+y;  //add x and y and store the value in z
    return z;  //return the value in z
}

Putting it all together

As stated before, a main() function will execute the statements within its body. Before a function is called by main(), the function must be declared.

For example, in the code snippet below, the printHello() function is declared in line 1. The main() function starts at line 2 and it calls the printHello() function (see line 3). The printHello function is defined in line 4.

//1. function declaration comes before the main() function
void printHello();

//2. main function starts
int main(){

    printHello(); //3. call function printHello()

    return 0;
}

//4. function definition
void printHello(){
    std::cout<<"Hello World!"; //print "Hello World!"
}

The code snippet above provides the general template of any program in C++. Next, you will learn about data types and variables in C++.

The plan for the game engine

My plan for the engine has always been to release it as open source. With the current release of beta version 3, my plan is becoming a reality. However, the engine is not ready to be released into the wild yet.

I believe half-baked Open Source projects should never be released. In my opinion, an Open Source project should be released when is stable, its basic features work, its documentation is complete and support channels are ready.

My game engine has the essential features to implement a game but still crashes in certain scenarios and lacks 1/3 of the documentation. On top of that, I don't have the time to provide technical support to users and collaborators.

Therefore, I will not release the game engine as an Open Source yet. Instead, I will use the game engine as an educational resource. I plan to teach Game Development in C++ using my game engine.

So, from today to the end of the year I will work on creating game development tutorials for you to learn.

Game Engine Beta v0.0.3

The last time I showed the progress of the engine was back in August where I showcased the first demo of the game engine. Since then I have been working on implementing new features and fixing several issues with the engine. Here is a video showing the progress of the engine:

In this beta version v0.0.3 of the engine, animations and collision detection can work simultaneously. The BHV algorithm was improved helping the engine make better decisions when pairing up 3D models for collision detection. The MVC (Model-View-Controller) flow of information was also improved.

Improvements

The beta version v0.0.3 of the engine can handle animations and collision simultaneously. The game engine can run an animation on a character and at the same time detect a collision. Furthermore, the engine allows a developer to apply action at any keyframe during the animation. For example, the engine allows you to apply an upward force during the second keyframe of an animation. You have direct control over the animation and which action to apply.

In this new version, I improved the flow of information between the Controller and the Model class. In previous versions, I had kept the flow of information in the MVC (Model-View-Controller) simplistic. In this new version, the model receives any actions on buttons and relays the information to the appropriate game character.

This version also improves the BVH algorithm. In the previous version, the BVH paired up incorrect models. This issue led several instances of missed collision detections. In this new version, all models are correctly paired up.

Issues

There is a major issue with the Convex Hull algorithm. For the most part, it works well with simple game characters. However, as soon as you model a character with complex modeling techniques, the algorithm fails. The problem is not a bug in the algorithm, but that it requires clean geometry to compute the hull.

It makes no sense to put a restriction on how to model a character. So, instead, I'm going to develop an external plugin to remove any restrictions. It will allow collision on any 3D character regardless of the modeling techniques used.

Thanks for reading.