What just happened?

We could simply recalculate the View matrix every time the Camera class was asked for it, but doing so would incur a small performance penalty. Because we do not have a lot of action happening in Cube Chaser, this penalty would not impact our game, but we can avoid it altogether. We are building a caching mechanism into the camera code in the event our game develops to the point that this optimization is helpful. Any time the View matrix is calculated, we will store it in cachedViewMatrix and simply return that matrix if the View matrix is requested without the underlying camera information having been modified.

In order to create the View matrix, we use the convenient Matrix.CreateLookAt() method, which accepts the camera position, the look at point we calculated previously, and a vector indicating what direction is considered to be up for the camera. In our case, we are using the pre-defined Vector3.Up, which translates to (0, 1, 0), or up along the positive Y axis.

That is enough of the camera to get us started. We will return to the Camera class later when we implement movement. For now, let's get on with actually drawing something to the screen!