I'm noticing a trend with the results I receive from various flow simulation runs, and wondering if anyone else has seen the same. I have a model with cooling water flow, 2 heat sources, and surfaces with a relatively large span (~ 5 ft X 6 ft) that need to be maintained at a certain temperature. There are several points of interest in the model, and on some runs, I have specified a fairly large number of goals. What I am finding however is that when fewer goals are selected, the results come out differently for the same points than when a greater number of goals are selected. I know that when goals are specified, more computer resources are allocated to them than other areas not specified with goals. The concern I have is that there are a number of critical points in the model that need accurate results. I presume the best approach would be to do a few runs specifying a minimum number of goals in each run for the sake of accuracy. I'm guessing too that this is all a function of computer capacity as well. If anyone else has run into this and found the best approach after a number of trials, I would be interested in hearing about what you found. Thanks.