The language of computing to describe physical processes has become popular in a number of scientific fields. However, without a clear definition of computing outside very narrow domains, such usage fails to add content to our understanding of physical reality. In this paper I explore how the theory of these specific engineered devices can possibly help us understand fundamental science, by close consideration of the connection between abstract computational theory and physical computing devices. Using the recently developed formalism of Abstraction/Representation Theory, I show what it means for a physical system to be acting as a computer, and give the conditions for a system to be capable of supporting a computational representation. A computational representation gives nontrivial information about the underlying physical system; but not every system under every physical theory is necessarily capable of supporting such a representation. In the cases where it is possible to represent a system computationally, this then becomes a new language and logic in which to describe, understand, and investigate the fundamental processes of physical reality.