SOME SAMPLE PROBLEM COMPARISONS BETWEEN THE DSA CELL MODEL AND THE QUASI-DETERMINISTIC METHOD

Variance reduction methods have become increasingly more automated wit h respect to the selection of parameters required by the methods. Each automatic or semi-automatic procedure is typically published showing the variance reduction achieved by the procedure on a different set of test problems from every other procedure. This makes it extremely dif ficult to understand the relative strengths and weaknesses of the diff erent procedures. This paper compares the DSA cell model and the quasi -deterministic methods (implemented in modified versions of MCNP) for obtaining splitting and Russian Roulette parameters on the same test p roblems.