A Half Century of Monte Carlo, from Slide Rule and Mechanical Desk Calculator to the Laptop Supercomputer

Abstract

Some early examples of Monte Carlo simulations of radiation transport, prior to the general availability of automatic electronic computers, are recalled. In particular, some results and details are presented of a gamma ray albedo calculation in the early 1950s by Hayward and Hubbell using mechanical desk calculators (+, -, x, ÷ only), in which 67 trajectories were determined using the RAND book of random numbers, with three random numbers at each collision being used to determine (1) the Compton scatter energy loss (and thus the deflection angle), (2) the azimuthal angle and (3) the path length since the previous collision. Successive angles were compounded in three dimensions using a two-dimensional grid with a rotating arm with a slider on it, the device being dubbed an “Ouija Board”. Survival probabilities along each path segment were determined analytically according to photoelectric absorption exponential attenuation in each of five materials, using a slide rule. For the five substances, H₂O, Al, Cu, Sn and Pb, useful number and energy albedo values were obtained for 1 MeV photons incident at 0° (normal), 45° and 80° angles of incidence. Advances in the Monte Carlo method following this and other early-1950s computations, up to the present time with high-speed all-function automatic computers, are briefly reviewed. A brief review of advances in the input cross section data, particularly for photon interactions, over that same period, is included.