The assessment of cardiac function, particularly cardiac output (Q) during heavy exercise is essential for the evaluation of cardiovascular factors that might limit oxygen transport. A series of invasive and noninvasive techniques has been developed for the assessment and monitoring of Q during resting and submaximal exercise conditions. However, very few techniques have been found to give accurate and reliable determinations of Q during vigorous to maximum exercise. For exercise physiologists and sport cardiologists, maximal exercise data are of primary importance. The 'gold standard' measures of cardiac function are considered to be the direct Fick and dye-dilution methods. These have been widely shown to give accurate and reliable determinations of Q during resting and submaximal exercise conditions; however, their use during maximal exercise conditions is debatable due to the inherent risks involved with each and their increasing inaccuracy during the later stages of vigorous exercise. Thermodilution has also been considered to be a relatively good method for the determination of Q during rest and exercise conditions, but recent authors have questioned its use due to the nature of the measure and its inaccuracy during strenuous exercise. Various noninvasive measures of cardiac function have been developed to overcome the problems associated with the 'gold standard' measures. The first part of this article discusses conventional techniques used in exercise physiology settings. The majority of these provide accurate and reliable determinations of Q during rest and submaximal exercise. However, very few techniques are suitable for maximal exercise conditions. Perhaps only the foreign gas rebreathe using acetylene (C2H2) meets all the criteria of being noninvasive, simple to use, reliable over repeated measurements, accurate and useful during maximal exercise.