The goal of this project is to produce a system capable of creating systems of angular and linear springs, which in turn are simulated with the goal of producing natural looking patterns. Both the rest angles of these angular springs and the initial placement of these springs will determine the output image. The project involves testing various methods of setting rest angles (rest angle set functions, or RASFs) with various methods of producing starting systems to find combinations which result in the most natural looking patterns. Various graphs were used as starting patterns such as Voronoi diagrams and different kinds of fractals. This experimentation resulted in a few properties of starting graphs and RASFs which are desirable for creating natural looking patterns. Generally, continuity in the angles from one edge is important, especially in strongly connected graphs. In fractal trees, asymmetricity is key for making natural looking trees, and a randomized RASF can produce convincing branches. Overall, it’s good to consider the uniformity of any starting graph or RASF, and non-uniformity is almost always preferable when creating natural looking patterns. The system ends up being a good and highly expandable method of creating natural and non-natural patterns.