This paper presents a review on common sources of inaccuracy related to solving the non-convex economic dispatch problem. The inaccuracies originate from using invalidated cost function, inaccurate calculation of transmission losses, violating equality or inequality constraints, comparing the performance of different algorithms based on different modeling equations, and comparing the results with different variants of benchmark systems. Numerous observations of unintended inaccuracies reported in previous published research work are used to illustrate the each argument; however, only few variants of benchmark test systems were sufficient to demonstrate and explain the existence of inaccuracies. The observed inaccuracies presented in this paper are explained, and the corresponding accurate results are presented to be used as standards in the future research.