11 min read

(For more resources related to this topic, see here.)

To put some graphics on the screen; the first step for us would be creating a new game window class that will use Direct2D. This new game window class will derive from our original game window class, while adding the Direct2D functionality.

Open Visual Studio. Add a new class to the project called GameWindow2D. We need to change its declaration to:

public class GameWindow2D : GameWindow, IDispoable

As you can see, it inherits from the GameWindow class meaning that it has all of the public and protected members of the GameWindow class, as though we had implemented them again in this class. It also implements the IDisposable interface, just as the GameWindow class does. Also, don’t forget to add a reference to SlimDX to this project if you haven’t already.

We need to add some using statements to the top of this class file as well. They are all the same using statements that the GameWindow class has, plus one more. The new one is SlimDX.Direct2D. They are as follows:

using System.Windows.Forms; using System.Diagnostics; using System.Drawing; using System; using SlimDX; using SlimDX.Direct2D; using SlimDX.Windows;

Next, we need to create a handful of member variables:

WindowRenderTarget m_RenderTarget; Factory m_Factory; PathGeometry m_Geometry; SolidColorBrush m_BrushRed; SolidColorBrush m_BrushGreen; SolidColorBrush m_BrushBlue;

The first variable is a WindowRenderTarget object. The term render target is used to refer to the surface we are going to draw on. In this case, it is our game window. However, this is not always the case. Games can render to other places as well. For example, rendering into a texture object is used to create various effects. One example would be a simple security camera effect. Say, we have a security camera in one room and a monitor in another room. We want the monitor to display what our security camera sees. To do this, we can render the camera’s view into a texture, which can then be used to texture the screen of the monitor. Of course, this has to be re-done in every frame so that the monitor screen shows what the camera is currently seeing. This idea is useful in 2D too.

Back to our member variables, the second one is a Factory object that we will be using to set up our Direct2D stuff. It is used to create Direct2D resources such as RenderTargets. The third variable is a PathGeometry object that will hold the geometry for the first thing we will draw, which will be a rectangle. The last three variables are all SolidColorBrush objects. We use these to specify the color we want to draw something with. There is a little more to them than that, but that’s all we need right now.

The constructor

Let’s turn our attention now to the constructor of our Direct2D game window class. It will do two things. Firstly, it will call the base class constructor (remember the base class is the original GameWindow class), and it will then get our Direct2D stuff initialized. The following is the initial code for our constructor:

public GameWindow2D(string title, int width, int height,
  bool fullscreen)     : base(title, width, height, fullscreen) {     m_Factory = new Factory();     WindowRenderTargetProperties properties = new
      WindowRenderTargetProperties();     properties.Handle = FormObject.Handle;     properties.PixelSize = new Size(width, height);     m_RenderTarget = new WindowRenderTarget(m_Factory,
      properties); }

In the preceding code, the line starting with a colon is calling the constructor of the base class for us. This ensures that everything inherited from the base class is initialized. In the body of the constructor, the first line creates a new Factory object and stores it in our m_Factory member variable. Next, we create a WindowRenderTargetProperties object and store the handle of our RenderForm object in it. Note that FormObject is one of the properties defined in our GameWindow base class. Remember that the RenderForm object is a SlimDX object that represents a window for us to draw on. The next line saves the size of our game window in the PixelSize property. The WindowRenderTargetProperties object is basically how we specify the initial configuration for a WindowRenderTarget object when we create it. The last line in our constructor creates our WindowRenderTarget object, storing it in our m_RenderTarget member variable. The two parameters we pass in are our Factory object and the WindowRenderTargetProperties object we just created. A WindowRenderTarget object is a render target that refers to the client area of a window. We use the WindowRenderTarget object to draw in a window.

Creating our rectangle

Now that our render target is set up, we are ready to draw stuff, but first we need to create something to draw! So, we will add a bit more code at the bottom of our constructor. First, we need to initialize our three SolidColorBrush objects. Add these three lines of code at the bottom of the constructor:

m_BrushRed = new SolidColorBrush(m_RenderTarget, new Color4(1.0f,
  1.0f, 0.0f, 0.0f)); m_BrushGreen = new SolidColorBrush(m_RenderTarget, new
  Color4(1.0f, 0.0f, 1.0f, 0.0f)); m_BrushBlue = new SolidColorBrush(m_RenderTarget, new Color4(1.0f,
  0.0f, 0.0f, 1.0f));

This code is fairly simple. For each brush, we pass in two parameters. The first parameter is the render target we will use this brush on. The second parameter is the color of the brush, which is an ARGB (Alpha Red Green Blue) value. The first parameter we give for the color is 1.0f. The f character on the end indicates that this number is of the float data type. We set alpha to 1.0 because we want the brush to be completely opaque. A value of 0.0 will make it completely transparent, and a value of 0.5 will be 50 percent transparent. Next, we have the red, green, and blue parameters. These are all float values in the range 0.0 to 1.0 as well. As you can see for the red brush, we set the red channel to 1.0f and the green and blue channels are both set to 0.0f. This means we have maximum red, but no green or blue in our color.

With our SolidColorBrush objects set up, we now have three brushes we can draw with, but we still lack something to draw! So, let’s fix that by adding some code to make our rectangle. Add this code to the end of the constructor:

m_Geometry = new PathGeometry(m_RenderTarget.Factory); using (GeometrySink sink = m_Geometry.Open()) {     int top = (int) (0.25f * FormObject.Height);     int left = (int) (0.25f * FormObject.Width);     int right = (int) (0.75f * FormObject.Width);     int bottom = (int) (0.75f * FormObject.Height);     PointF p0 = new Point(left, top);     PointF p1 = new Point(right, top);     PointF p2 = new Point(right, bottom);     PointF p3 = new Point(left, bottom);     sink.BeginFigure(p0, FigureBegin.Filled);     sink.AddLine(p1);     sink.AddLine(p2);     sink.AddLine(p3);     sink.EndFigure(FigureEnd.Closed);     sink.Close(); }

This code is a bit longer, but it’s still fairly simple. The first line creates a new PathGeometry object and stores it in our m_Geometry member variable. The next line starts the using block and creates a new GeometrySink object that we will use to build the geometry of our rectangle. The using block will automatically dispose of the GeometrySink object for us when program execution reaches the end of the using block.

The using blocks only work with objects that implement the IDisposable interface.

The next four lines calculate where each edge of our rectangle will be. For example, the first line calculates the vertical position of the top edge of the rectangle. In this case, we are making the rectangle’s top edge be 25 percent of the way down from the top of the screen. Then, we do the same thing for the other three sides of our rectangle. The second group of four lines of code creates four Point objects and initializes them using the values we just calculated. These four Point objects represent the corners of our rectangle. A point is also often referred to as a vertex. When we have more than one vertex, we call them vertices (pronounced as vert-is-ces).

The final group of code has six lines. They use the GeometrySink and the Point objects we just created to set up the geometry of our rectangle inside the PathGeometry object. The first line uses the BeginFigure() method to begin the creation of a new geometric figure. The next three lines each add one more line segment to the figure by adding another point or vertex to it. With all four vertices added, we then call the EndFigure() method to specify that we are done adding vertices. The last line calls the Close() method to specify that we are finished adding geometric figures, since we can have more than one if we want. In this case, we are only adding one geometric figure, our rectangle.

Drawing our rectangle

Since our rectangle never changes, we don’t need to add any code to our UpdateScene() method. We will override the base class’s UpdateScene() method anyway, in case we need to add some code in here later, which is given as follows:

public override void UpdateScene(double frameTime) {     base.UpdateScene(frameTime); }

As you can see, we only have one line of code in this override modifier of the base class’s UpdateScene() method. It simply calls the base class’s version of this method. This is important because the base class’s UpdateScene() method contains our code that gets the latest user input data each frame.

Now, we are finally ready to write the code that will draw our rectangle on the screen! We will override the RenderScene() method so we can add our custom code. The following is the code:

public override void RenderScene() {     if ((!this.IsInitialized) || this.IsDisposed)     {         return;     }     m_RenderTarget.BeginDraw();     m_RenderTarget.Clear(ClearColor);     m_RenderTarget.FillGeometry(m_Geometry, m_BrushBlue);     m_RenderTarget.DrawGeometry(m_Geometry, m_BrushRed, 1.0f);     m_RenderTarget.EndDraw();

}

First, we have an if statement, which happens to be identical to the one we put in the base class’s RenderScene() method. This is because we are not calling the base class’s RenderScene() method, since the only code in it is this if statement. Not calling the base class version of this method will give us a slight performance boost, since we don’t have the overhead of that function call. We could do the same thing with the UpdateScene() method as well. In this case we didn’t though, because the base class version of that method has a lot more code in it. In your own projects you may want to copy and paste that code into your override of the UpdateScene() method.

The next line of code calls the render target’s BeginDraw() method to tell it that we are ready to begin drawing. Then, we clear the screen on the next line by filling it with the color stored in the ClearColor property that is defined by our GameWindow base class. The last three lines draw our geometry twice. First, we draw it using the FillGeometry() method of our render target. This will draw our rectangle filled in with the specified brush (in this case, solid blue). Then, we draw the rectangle a second time, but this time with the DrawGeometry() method. This draws only the lines of our shape but doesn’t fill it in, so this draws a border on our rectangle. The extra parameter on the DrawGeometry() method is optional and specifies the width of the lines we are drawing. We set it to 1.0f, which means the lines will be one-pixel wide. And the last line calls the EndDraw() method to tell the render target that we are finished drawing.

Cleanup

As usual, we need to clean things up after ourselves when the program closes. So, we need to add override of the base class’s Dispose(bool) method. We’ve already done this a few times, so it should be somewhat familiar and is not shown here.

Our blue rectangle with a red border

As you might guess, there is a lot more you can do with drawing geometry. You can draw curved line segments and draw shapes with gradient brushes too for example. You can also draw text on the screen using the render target’s DrawText() method. But since we have limited space on these pages, we’re going to look at how to draw bitmap images on the screen. These images are something that make up the graphics of most 2D games.

Summary

In this article, we first made a simple demo application that drew a rectangle on the screen. Then, we got a bit more ambitious and built a 2D tile-based game world.

Resources for Article:


Further resources on this subject:


LEAVE A REPLY

Please enter your comment!
Please enter your name here