Littmann's formula relating the size of a retinal feature to its measured image size on a telecentric fundus camera film is widely used. It requires only the corneal radius, ametropia, and Littmann's factor q obtained from nomograms or tables. These procedures are here computerized for practitioners' convenience. Basic optical principles are discussed, showing q to be a constant fraction of the theoretical ocular dimension k', the distance from the eye's second principal point to the retina. If the eye's axial length is known, three new methods of determining q become available: (a) simply reducing the axial length by a constant 1.82 mm; (b) constructing a personalized schematic eye, given additional data; (c) ray tracing through this eye to extend calculations to peripheral retinal areas. Results of all these evaluations for 12 subjects of known ocular dimensions are presented for comparison. Method (a), the simplest, is arguably the most reliable. It shows good agreement with Littmann's supplementary procedure when the eye's axial length is known.