Computational physics is the study of numerical algorithms to solve physics problems in which quantitative theory already exists. Computerized mathematical models provide predictions on how systems will behave. This was the first application of computers in science.
In absence of computers, certain mathematical models cannot be solved. For instance, when quantitate theory does not have a closed-form expression or the calculations are too complicated for human calculations, numerical approximations are required. Computational physics is used to find the approximate solution and respective error for numerical algorithms.
Computational physics can be used for a wide range of problems and is therefore divided into categories amongst the different mathematical problems it solves. The mathematical problems computational physics can solve are: protein structure predictions, matrix eigenvalue problems in quantum physics, partial differential equations, simulating physical systems, ordinary differential equations and integration.
The areas in which computational physics can be applied are vast. It is currently being used in, astrophysics, lattice field theory, accelerator physics, fluid mechanics, solid state physics, plasma physics and soft condensed matter physics.