OBJECTIVE Automated perimetry is both one of the most important diagnostic procedures in ophthalmology and the most difficult for the patient who is required to comply with strict and tiring conditions. This paper examines the use of a moving fixation target and a strategy for full threshold determination only in those data points found to be abnormal. METHODS 142 eyes in 71 patients were subjected to two types of visual field tests: the Dicon TKS 4000 Autoperimeter program 5 and the Humphrey field analyser program 30/2. The first procedure was compared with a commonly used instrument and strategy that is usually employed for the management of glaucoma patients. RESULTS Such a strategy was used in patients presenting with neuro-ophthalmic problems taking an average testing time of 14.9 minutes. The alternative test described here took an average of 3.9 minutes, without loss of diagnostic value. CONCLUSIONS The shorter testing time results in greater patient acceptance, fewer fatigue induced artefacts, and the possibility of completing a perimetry test with patients whose ability might seriously be challenged by a longer test. The reduced time required and the more natural condition of not having to fixate on an unmoving position were found not to reduce the diagnostic value of the data produced in patients presenting for various reasons at a neuro-ophthalmology clinic.