PostScript Halftones

Because of the device independence, PostScript halftones are created in two stages. First the tonal values captured as gray-scale levels. These are then rendered (mapped) to the particular output device.

• During the rendering stage, transfer functions may be applied to adjust for distortions in tonal values. Then halftone screen frequency, angle and dot shape are determined and the image finally rasterized.

• The specification of gray-scale levels is independent of the rendering stage, i.e. it remains device independent.

• Dot shapes are determined by a spot function, which specifies the sequence of device pixels that make up the specified dot.

• PostScript allows an application to define the spot function with complete freedom, thus allowing use of alternative dithering methods.

• Halftone parameters may be stored as part of the file description or in a halftone dictionary.

• PostScript allows different objects to have different halftone parameters.

• Digital halftone screens are based on a grid of device pixels. Thus, there is a relationship between the screen ruling and the number of gray levels. For a given ruling and spot size, the cell size is fixed and the number of gray levels is one more than the number of spots in the cell.

• Thus for a given device, as the halftone frequency changes, the number of gray levels (device pixels in the cell) varies.

• Hence, to achieve 256 gray levels (really 257, i.e. 0-256) the halftone cell must be based on a 16x16 grid. A 2400 dpi imagesetter can then produce a 150-lpi screen with sufficient gray levels.

• Correspondingly, a 300 dpi laser printer would only allow a 18-lpi screen. In this case, it is usually necessary to reduce the number of gray levels and/or utilize FM screening.