3D graphics: normal display vs bump mapping?

Question

I know that normal display describes the process of adding details to grids without increasing the number of polygons, and this is achieved by using certain normal textures to control the way light is applied to the object. Good.

• But what is bump mapping? Is this just another term for normal display?
• How are visual results compared? Is it possible to combine both methods?

Answer 1

Bump Mapping describes a general technique for modeling bumps and wrinkles on the surface of an object. This is usually achieved by manipulating surface normals in lighting calculations.

Normal mapping is a variation of Bump Mapping in which surface normals are provided through a texture, with normals embedded in the RGB image channels.

Other methods, such as Parallax Mapping, are also Bump Mapping methods, as they distort normal surfaces.

To answer the second part of the question, they could be easily combined. The normals of the base surface can be determined from the normal display, and then modified using another relief mapping method.

Answer 2

Mapping Bump was originally proposed by Jim Blinn in 1978. His system basically works, disturbing the normal on the surface, using the height of this texel and the height of the surrounding texels.

This is very similar to Bumpmapping DUDV (you can recall the original BMS mapping mapping that was introduced in DX6, which was DUDV). This works by precomputing the derivatives from above so that you can skip the first step of the computation (since it does not change every frame).

Normal mapping is a very similar method that works by simply replacing the normal at each texel position. Conceptually its much simpler.

There is another method that gives “similar” results. It is called embossed relief mapping. This method works using multi-pass rendering. Basically you end up subtracting a gray scale height map from the last pass, but compensating for it with a small amount depending on the direction of the light.

There are other ways to emulate surface topology.

Elevation mapping uses the elevation map as an alpha texture, and then displays several fragments through this texture with a different alpha value to simulate a change in height. However, if they are not performed correctly, the slices can be very noticeable.

Displacement mapping works by creating a three-dimensional grid based on texture. This will obviously significantly increase the number of your vertices.

Steep parallax, bump mapping, etc. - latest methods. They work by casting a beam through a height map until it crosses. This has the big advantage that if the piece should block the texture, now performed when the beam does not hit the DEM behind where it initially hits, the “nearest” texel is always displayed.