Comparative examinations were carried out under controlled conditions in 101 eyes of 66 open-angle glaucoma patients with field defects (Stages I-III according to Aulhorn) with the Octopus and Competer computerized perimeters. The results of the Octopus perimetry were selected as a reference system to evaluate the sensitivity of the Competer in identifying early glaucomatous field defects. The interpretation of qualitative and quantitative discrepancies between the two instruments took into account the different sizes of the test fields. Of the glaucomatous field defects identified with the Octopus perimeter, 84.1% could be recognized with the Competer as well; 4% were questionably pathologic and 11.9% appeared to be false-negative in the Competers. The 11.9% false-negative results of the Competer were all due to the fact that they were scotomas beyond the test eccentricity of this perimeter. In 16.9% of the positive Competer findings the scotomas were outside the Competer test field by more than 50% of their total extent, whereas 83.1% were within the Competer test field by more than 50% of their total extent. In the light of these results it may be concluded that the Competer provides good sensitivity to detect early glaucomatous field defects once these are within 20 degrees eccentricity, while more peripherally situated scotomas will be missed. It would be advisable to enlarge the test field of the Competer to 30 degrees eccentricity. The relation of early glaucomatous field changes to disk morphology revealed that in 42.5% of the eyes definite glaucomatous disk changes were not recognized.