A new look at zooming

Over the past several decades, "zooming" methods have been used by some as a means of producing computational results with a reduced computational cost. Unfortunately, zooming methods have not gained widespread acceptance both because of the perceived difficulty associated with setting the problem up, together with a lack of accuracy; the latter is addressed herein through the introduction and development of "virtual areas", which is a new concept. A key to using virtual areas is to predict them; a method for empirically predicting the virtual areas is described. Results for several two-dimensional, steady, inviscid duct-like configurations are shown.