The conceptual developments of x-ray dose calculation algorithms followed an evolutionary progression that mirrors what led to a deeper understanding of visible light. For higher megavoltage energies, photon scattering was less prevalent but electron ranges were longer and could disrupt equilibrium inside the body. Algorithms had to be improved, particularly in regions of low density tissue where the ranges were further elongated. Tracking of charged particle migration away from photon interaction sites became much more important. The chapter introduces three dose modelling approaches used currently in clinical radiation oncology. Empirical methods all solve the same radiation propagation problem, but view it from a different vantage and invoke different mathematical tools: convolution-superposition of Green’s functions, stochastic Monte Carlo simulation, and deterministic solution of Boltzmann equations. All algorithms indeed had access to the computed tomography image matrix for display, but only made partial use of the density information for dose computation purposes.