The role of cardiac computed tomography (CT) in clinical practice is constantly evolving. Early machines were only capable of measuring coronary calcification. Advances in temporal and spatial resolution, especially the introduction of 64-detector rows, now mean that high-quality non-invasive angiograms are possible in most patients. This review will outline the capabilities and limitations of coronary artery imaging with CT, and also highlight areas that differentiate CT from X-ray angiography, including direct plaque visualisation and potential vulnerable plaque identification. Development of cardiac computed tomography The concept of 'computerised transverse axial scanning' was first demonstrated by Godfrey Hounsfield nearly 30 years ago. 1 Initial computed tomography (CT) scanners required up to 300 seconds for the acquisition of a single image. With such poor temporal resolution they were only suitable for imaging static structures such as the brain. 2 The coronary arteries move throughout the cardiac cycle, although their velocity decreases in diastole. 3 This underlies the concept of 'gating' the scan with the electrocardiogram (ECG), so that data are acquired preferentially during diastole. 4 The advent of multi-detector CT (MDCT) has allowed simultaneous acquisition of multiple slices of imaging data. Current CT scanners can deliver a temporal resolution of 75 ms (due to very rapid gantry rotation 5) at a spatial resolution of under 400 µm. 5