The ongoing de-regulation of the U.S. electric power industry is to build open-access, non-discriminative and competitive electricity markets. In 2002 the Federal Energy Regulatory Commission (FERC) proposed Standard Market Design (SMD), in which the key is the Locational Marginal Pricing (LMP) methodology to determine the energy price and to manage the network congestion. LMP has been adopted by or is under implementation at a number of ISO's such as PJM, NYISO, ISO-New England, etc. This paper presents an iterative DC Optimal Power Flow (DCOPF) algorithm to calculate LMP. Iteration is needed in a linearized DC network because the loss factor is dependent on generation dispatch, while generation dispatch is affected by loss factor as well. The results from DCOPF are compared with ACOPF algorithm at different loading levels and transmission line resistances. It is interesting to observe that the LMP results from two models are close for most of the loading levels, but may be significantly different in a few scenarios when the error in the linearized DC model causes the change of marginal units. It is also interesting to observe the increasing trend of the difference between two models when line resistance grows.