Calcomp Inc.

Robert L Hammond

Bala R Vatti

Three terms (X,Y,'Error') are combined in a single floating-point number in an advanced microprocessor to compute pixel addresses. The floating-point number is reconverted to integer form in multiplexers operating in a mode determined by slope, relative to unity, of the line to be defined by the pixels. The output of the multiplexers is enhanced by a look-up table to produce color data, which are converted to analog form for display. Speed of operation is greatly augmented by the use of floating-point arithmetic to compute the pixel addresses.

Rapid line drawing in computer graphics employing floating-point

This method and apparatus are suitable for inclusion in the 'graphics circuit board' of a personal computer. The digital output of the computer is used to plot a series of 'pixel pairs' which, taken together, closely approximate the desired path of a line to be drawn on a display device such as a color cathode-ray tube. The addresses and relative intensities of the pixels of each pair and of the series are computed so as to produce a line that is substantially free from the 'jaggies' resulting from 'aliasing' in prior-art apparatus. Computation of addresses relies upon both integer and floating-point numbers.

Growing community of inventors

Nashua, NH, United States of America

Robert L Hammond