Stochastic All-to-All Propagators for Baryon Correlators

Abstract

The effectiveness of various dilution schemes in the evaluation of baryonic two-point functions is compared. The error of a representative set of observables as a function of the number of Dirac matrix inversions is used as a basis for comparison. To achieve an equivalent reduction in error, we demonstrate that an increase in the number of dilution projectors on a single noise source usually requires fewer inversions than the use of multiple noise sources. This exploratory study was performed on 100 quenched gauge configurations and will be applied to the calculation of low-lying hadron spectra.