Information about spatial distribution of rock true resistivity (Rt) throughout an oil or gas field is always desired. Since seismic survey is the only widespread source that provides information for inter-well locations, considerable efforts have been devoted to extract as much data as possible from it. It is therefore desirable to be able to extract Rt from seismic attributes normally represented by acoustic impedance (AI). This paper presents a field trial of an approach that is basically an application of artificial intelligence (artificial neural network, ANN) on well-log and seismic data based on a proven theoretical relationship that relates Rt to AI. The approach itself, which has been successfully verified through a series of laboratory trials, includes training of the ANN using relevant well-log data, Rt prediction using the trained ANN, and blind tests as means of result validation. An oil field located in East Java is chosen for the trial. It has been shown that there is a certain correlation between log-derived resistivity and log-derived AI. As the approach is applied to map the resistivity and water-saturation, comparisons between conventional/ deterministic water-saturation (Sw) map and the corresponding map resulted from the trial has shown the superiority of the method in presenting inter-well variations in water-saturation. It is also found that the new method has provided a high level of flexibility in interpreting and distributing the inter-well Sw values.