- Animation Techniques
- DirectX Objects
- DirectX Techniques
DirectX makes use of early animation technologies, but has added a few tricks of its own. This section describes DirectX objects and concepts used throughout the computer animation community.
Frame rates are basically the speed at which frames or cels of animation are shown, expressed in number of frames per second of time. A normal frame rate for animated motion pictures is 24 frames per second (fps). With computer animation, frame rates look much better at about 40fps-50fps.
When working with DirectX programming, frame rates are one of the most important factors to keep in mind. If the frame rate of the animation you're producing is too slow, the user will be bored by your animation. If it's too fast, it will look unnatural, perhaps making the user think something is wrong with the machine. Therefore, you need to make sure the frame rate is a constant one that's pleasing to the eye.
Frame rates are implemented in DirectX by how you send data to the video card. You want to make sure you send enough frames of your animation to the card to give you the results you need. However, not all video cards have the same capabilities, so you need to make sure you don't overload the card. For example, if you know that your frame rate is going to be 50fps, but the average card specification for your users doesn't support more than 30fps under this load, you could have a problem. A good rule of thumb when working with video cards is to look at trade magazines and online sites to determine what video cards are popular. Then you can choose from the bottom and top of that price range to determine the cards you should test. Also, using programs such as 3D Mark can help you determine the capabilities of various video cards. 3D Mark can be found at http://MadOnion.com. Using timers can help regulate frame rates. Chapter 2, "Multimedia Programming in the Visual Basic Environment," discusses using timers to control frame rates.
Blit, an acronym for "bit block transfer," is the process of moving sections of data from one memory space into another. After you assemble the frame of animation you want to show, you must then move it from the back drawing surface to the front drawing surface, where the user can see it. To do that, you use blits. Blits can transfer a full screen of information from one memory buffer to the other. They can also transfer a small rectangle of information. After the information is blitted to a surface, the surface can be flipped with another surface to change the image the user sees. Figure 3.4 shows how Blits are performed on an offscreen drawing surface. The blit process and surfaces are explained in more detail in Chapter 5, "Creating Basic Animations with Surface Objects."
Figure 3.4 Blits being performed on an offscreen drawing surface.
Sprites are rectangles of data that contain an image. These small portions of the overall image are used to create the animation effect.
Sprites can be used in the foreground to give you animation. For example, if you have a kiosk program that shows animated tickets moving across the screen, you would need to create a series of images for each unique frame of the animation. You could then save all the sprite images as one image, which is called a sprite strip. The sprite strip in Figure 3.5 shows a spinning donut in all four possible poses. You would then move the sprite around the screen by changing its coordinates and loading a different image every frame. Chapter 5 discusses sprites and how they are animated in a DirectDraw application.
Figure 3.5 A sprite strip.
The drawing surface is where you assemble your picture for blitting; it can be thought of as the staging area for each frame of animation. The drawing surface can be made up of a background image and one or more sprites. During the process of blitting the image, this surface is transferred from the back buffer to the front buffer. Buffers and surfaces are explained in more detail in Chapter 5. There are three types of drawing surfaces used in animation:
Primary: The primary drawing surface is the one currently displayed on the screen. It's what the user sees at all times.
Secondary: The secondary drawing surface is where you set up for the next frame of animation. When it's time for the next animation frame to be displayed, it becomes the primary surface and the old primary surface becomes the new secondary. In essence, you force them to switch identities.
Offscreen: The offscreen drawing surface is where you assemble the content for the primary and secondary surfaces. This part of DirectDraw is always working to keep the primary and secondary surfaces full of information.
An offscreen surface is used only if your design for flipping uses three surfaces. Three surfaces are used in more complex animations, especially since the memory on the video card can help speed this process up. However, for simple animations you need to use only a primary and secondary surface.
Palettes are simply a list of colors available for the display, and you choose from this list when rendering the animation. The color depth required for your application determines the number of colors available in a palette. Just as it does for a painter, the palette defines the colors you have to work with in your project. When working with palettes, it's a good idea to choose a color depth that gives you the best results for your animation work.
It's possible to change the palette at runtime when producing animations, however. This animation technique is good for full-screen animations that require fading, such as a sunset scene or movie credits.
Rectangles are square areas of the screen that you define with an x1, y1, x2, y2 coordinate system. Many of the methods in DirectX need to know the area of the display that's being modified. This is accomplished by passing a rectangle object that provides the coordinates of these areas. For example, if you have a small bug on the screen and you want to change its color, it's easier to just pass in a rectangle object with the bug's dimensions and location on your screen.
You can tell the DirectDraw object where the bug is by passing in the Rectangle object that contains the location of the rectangle and its dimensions. In Microsoft Visual Basic, the RECT type has the following definition:
Type RECT Left As Long ` Top-left. X coordinate. Top As Long ` Top-left. Y coordinate. Right As Long ` Bottom-right. X coordinate. Bottom As Long ` Bottom-right. Y coordinate. End Type
In the preceding example, the Left and Top members are the x- and y-coordinates of a bounding rectangle's top-left corner. Similarly, the Right and Bottom members make up the coordinates of the bottom-right corner. The following diagram in Figure 3.6 illustrates how you can visualize these values by showing how a rectangle (RECT) is measured on the screen.
Figure 3.6 A rectangle.
Clipping is the process of defining areas of the display that will show the animation. By marking off these areas, you are designating where it is okay for DirectDraw to write(or blit) on the destination surface. An example of this technique can be seen in a shooting gallery. The duck targets move from one side of the screen to the other; however, you don't want them to appear on the left and right sides of the shooting area because that would ruin the effect of your shooting gallery. You would need to define a clipper rectangle that encompasses the "shooting area" of the display. This technique gives the illusion that the ducks are going behind the curtain.
Figure 3.7 shows a screen made up of clipper rectangles. Each of the four rectangles is part of a clipper list. The clipper list is associated with a clipper object, which in turn defines the area that can be used for animation. Any blitting that occurs outside this area will not show up on the screen. If a sprite crosses this boundary, only the part of the sprite that is within the clipper area is displayed. In the previous example of the shooting gallery, you would want the ducks to appear only in the "shooting area," not in front of the curtains.
Figure 3.7 A screen with clipper rectangles.
Billboards are objects used to create the illusion of 3D objects without having to spend the resources on creating complex objects. This effect is accomplished by creating a plane object that's moved in relation to the person viewing it, which produces the illusion that the object has many sides, even though it actually has only one side. The one side keeps moving in relation to the person viewing it.
For example, if you were creating a driving game that had many trees along the sides of the course, you don't want to create complex trees made up of thousands of polygons. The memory needed for that operation would be enormous, and your program's performance would suffer. So you simply create a 2D image of a tree, attach it to a billboard, and move the billboard in relation to the user. The user doesn't know the difference, and you save on performance. Figure 3.8 shows how billboards work.
Figure 3.8 A billboard animation.
Billboards work well with objects that have a vertical nature, such as trees and signs. Make sure your users can't view the objects from the top, however; otherwise, they'll see that the objects are a facade, and your illusion will be ruined.