Rendering is an image processing technique that takes 3D digital objects or scenes with their assigned physical-based material properties, lighting conditions and a camera with a chosen field of view.
The 3D model, besides containing geometric data, may also contain other information regarding material properties of each object and the lighting within the scene. To render a photo-realistic image, more information is needed, such as the object's color, texture, and lighting. For a glass object, a degree of transparency, edge density, and reflections can be applied. This extra information is commonly referred to as specifying object attributes.
Assigning a material to an object is similar to the specification provided to the building contractor. In order to duplicate the object's surface quality, it is necessary to obtain this information in a digital format. One way to create a texture map is by scanning a physical sample material provided by the designer or to scan a photograph. A texture may be created within a paint program like photoshop.
This technique is referred to as texture bit mapping. The surface of the object is based on the bit map image and looks as if the image was painted onto the object. You can assign bit maps to change other properties of a material, such as bump maps, specular maps, reflection maps, and opacity maps.
In addition to bit maps, some high-end rendering software provides a procedural method of applying a 3D texture. This texture is written in a computer code language and, once mastered, can be modified. Third-party developers provide libraries of procedural volume shaders or solid shaders.
This method provides a somewhat better quality of texture since it generates a solid 3D pattern during the rendering process and it does not require mapping coordinates to properly map textures to a complex object.
Lighting Objects or Scenes
This process is the most difficult and time consuming as digital light behaves somewhat differently than in a true life, and is one of the most challenging and least understood tasks in the synthetic rendering process.
There are a variety of types of light sources available to illuminate a scene: spotlights, ambient light, radial lights, parallel lights, tube lights, or camera light. Having studied architectural photography some years ago, I have tried to apply my lighting knowledge to the world of digital lighting. But I soon learned that digital lighting behaves much differently than traditional lighting techniques. The only way to create well-lighted scenes is by trial and error. A parallel light source behaves much like the sun. A low radial light added to a parallel light will lighten the hard shadows, while ambient light will add or subtract to the overall lighting of the scene. In a complex scene over a hundred light sources of different types could be placed over the site and many tests must be conducted before a final satisfying rendering is produced.
Many architects would argue that the traditional method of hand rendering gives the drawing a human touch, whereas the synthetic rendering tends to look somewhat plastic. This argument has some validity, and it takes an experienced person to produce electronic renderings that will satisfy an experienced traditional renderer. The greatest advantage of electronic rendering is that the digital rendering can be changed relatively painlessly and in a timely manner. Gone are the
days of scraping away, cutting, and pasting with paper and pen. The computer is much more forgiving than watercolor paper. Paint in some trees, and if you don't like the way they look, simply "revert" back to the original image. Try doing that with gouache!
In a simple shading rendering technique, only the light coming directly from the source is considered in the shading process. In more realistic and accurate images, it is important not only to take into account the light source itself, but also how all the surfaces and objects in the scene interact with the light. Lighting a 3D scene is a complex and time-consuming process, both in terms of required computing power and the time it takes to place and experiment with light sources.
What is more economical: investing in a tool that is fast, accurate, and easy to use but costly, or a tool that is much less expensive but takes more time? I personally find it more rewarding to be able to concentrate on lighting techniques and previewing the rendered image in a matter of seconds or minutes while still concentrating on possible improvements, than waiting for hours for the image to appear. Creating beautiful, true-to-life images is like playing music; one must
experiment and practice until satisfied with the results. We all have limits as to the time we can spend on a project in order to compete in this fast-growing marketplace and remain in business. Clients have budgets and once they realize the economic benefits of having high-quality projects produced in digital media, they will be willing to increase their spending. It won't be long before this form of visualization becomes the standard practice within the design profession.