direct x c++ camera movement

Question

I'm fairly new to directx so this may sound really basic.

I have started working on a first person game where you can walk through rooms, the language i am coding in is c++ and I'm using directx to help me create my game.

So far i have all the rooms drawn with doors etc but im a bit stuck how to make a first person camera and allow the user move forwards, backwards and side to side using the arrow keys on a keyboard.

The simpler the better as i am a beginner.

Could anyone help me out with this or point me in the right direction?

Thanks in advance

Answer 1

There's a lot of tutorials in web covering this topic, so Google will certainly help you.

As for the basics: You will want to store your position and camera rotation. Assuming Z is your up-axis, you should use the arrows to change only the X and Y.

Let's also say, that you will store your camera orientation is stored as a composition of rotations along Z-axis (movement direction) and X axis (looking up-down).

Simple class:

class Player 
{
protected:
    float3 Position; // Z-up
    float2 CameraRotation; // X for turning, Y for up-down
public:
    void MoveForward()
    {
        Position.X += -cosf(CameraRotation.X) * PLAYER_SPEED;
        Position.Y += -sinf(CameraRotation.X) * PLAYER_SPEED;
    }
    // when using any other arrow, just add a multiply of PI/2 to the camera rotation
    // PI for backwards, +PI/2 for left strafe and -PI/2 for right strafe.
    // If you don't want to use mouse, use left and right arrow to modify camera rotation
    // and MoveForward and Backward will look the same, having different signs.
};

The '-' signs in front of sinf and cosf functions are there because you will probably want this kind of behavior, feel free to change them.

As for camera, you will have to implement mouse delta between frames. In every frame compare the mouse position with the previous one. Then multiply it by turning and looking speed and set directly to camera value.

Hope this helped.

Answer 2

I am more of an OpenGL guy so I cannot help you with the technical side, what I can do is give you a direction.

In general, a 3D camera has:

Translation - where the camera is (x, y, z)

Rotation - angle of the camera around each axis

What you want do is related to the translation part only:

1. Monitor the user input, and capture when he presses one of the arrow keys.
2. When an arrow key is down, you want to start changing the camera translation. If the user presses the up key, you would add a constant value to the corresponding camera translation component (x) each iteration of your main game loop until he releases the key. If he presses the down key, you would want to subtract that value instead of adding it.
3. When he releases the key, your code needs to stop adding that value to the camera translation.

Let's assume that your game runs at 60Hz, and you add, say, 1/60 units to the camera translation each iteration for each direction the user wants to go. If the user held up the up arrow key for 2 seconds, the camera would have moved forward 2 units.

This is the "theory" in general, now I can only point you to web pages I found that may be useful for solving the technical side of your problem:

DirectX camera movement - I'm guessing this article has a lot more than you need, but it looked pretty good and I think that you should read it anyway... But you can just skip to the View Transformation part.

Input handling - nothing much to say, regular Win32 input handling. If you are not familiar with win32 input handling I think that you should take an hour or two to learn that first.

Alright that's it, hope I helped