Conventional and digital mammographic images obtained with storage phosphors were compared. The digital images were acquired with high-resolution 3rd-generation (HR III) screens and specifically adapted algorithms. The experience was made both on a phantom and in vivo. The phantom study was carried out by comparing conventional with digital radiographs acquired with: a) same kV (28) and same mAs; b) same kV (24) and same mAs; c) 28 kV with 30% mAs reduction, in digital images only. The results obtained upon counting the amount of recognizable details per image demonstrated slight loss of information in digital radiographs only when mAs was reduced by 30%. Two hundred patients were studied; they were divided into four groups according to the type of breast (medium inherent contrast or solid breast) and to exposure factors. After conventional mammography, an additional digital radiograph was performed using one of the three techniques previously employed on the phantom. Separately, 10 ductogalactographies and 10 magnification radiographs of microcalcifications were compared; the same exposure factors were used in these images too. The radiographs were viewed by three expert mammographic radiologists; contrast quality and spatial resolution were investigated and a score was given to each image on a 3-grade scale (insufficient, sufficient, good). The comparison of the mean values obtained showed higher contrast and better enhancement of nodular lesions on digital images, on which, however, the depiction of microcalcifications was worse than on conventional radiographs, especially with lower radiation doses, even though the number of detected microcalcifications was the same.