Computer-generated holography (CGH) is a technique for converting a three-dimensional (3D) object scene into a two-dimensional (2D), complex-valued hologram. One of the major bottlenecks of CGH is the intensive computation that is involved in the hologram generation process. To overcome this problem, numerous research works have been conducted with the aim of reducing arithmetic operations involved in CGH. In this paper, we shall review a number of fast CGH methods that have been developed in the past decade. These methods, which are commonly referred to as point-based CGH, are applied to compute digital Fresnel holograms for an object space that is represented in a point cloud model. While each method has its own strength and weakness, trading off conflicting issues, such as computation efficiency and memory requirement, they also exhibit potential grounds of synergy. We hope that this paper will bring out the essence of each method and provide some insight on how different methods may crossover into better ones.