INSTRUMENT MARKING 



79 



legibility of such design factors as stroke 

 width and any variations there, of, the area 

 of white space included within figure out- 

 hnes, and the spacing of grouped digits. It 

 remained, however, for Berger (7, 8) to make 

 the first clearly systematic studj^ of the 

 whole problem. Berger's experimental var- 

 iables included stroke width, numeral shape, 

 spacing and lighting. His measure of leg- 

 ibility was the maximum distance at which 

 digits of a given size could be distinguished. 



8DI29 



34576 



B 



12 345 



12345678^0 



Fig. 1. Three sets of digit forms 



A. Berger's digits (7, 8) showing optimum 

 stroke thicknesses for white on black displays 

 seen by reflected light and by transmitted light 



B. U.S. Army-Navy standard instrument dial 

 numbers (91) 



C. Craik's digits (19) intended for white on 

 black displays 



The forms which Berger arrived at are 

 shown with two comparison sets of numer- 

 als in Fig. 1. 



Berger began by studying stroke '^ddth. 

 He proposed to find what width would per- 

 mit discrimination between the numerals 2, 

 5, and 8 at the greatest distance. Experi- 

 ments were conducted out of doors under 

 three conditions of lighting and contrast: 

 (1) with diffuse light, black figures on a white 

 ground; (2) wdth diffuse hght, white figures 

 on a dark ground; and (3) with bright figures 



produced by transmitted hght. These tests 

 showed that optimum stroke width is a func- 

 tion of the specific illumination and contrast 

 conditions. Expressed as a fraction of num- 

 eral height, the optimum widths found by 

 Berger were as follows: l/13th letter height 

 for diffusely lit white figures on a black 

 background; l/8th for diffusely lit black 

 figures on a white background; and l/40th 

 for bright figures presented by transmitted 

 hght. His results for the diffusely lit white 

 on black and black on white figures are 



-0\ WHITE ON BLACK 



</) f 30 - 



\*^ 



BLACK ON WHITE 



1,40 i:i6 no i:8 i:6 1:5 



RATIO OF STROKE WIDTH TO NUMBER HEIGHT 



Fig. 2. Influence of ratio of stroke to height 

 and of direction of contrast upon the legibility of 

 Berger's digits. Data for two subjects who ob- 

 served approaching 80-mm. numerals until they 

 could be correctly read (7). 



shown in Fig. 2. These data, based on the 

 observations of tw^o subjects, show neatly 

 the nature of the interaction between con- 

 trast direction and stroke ^^^dth. Each type 

 of figure had its own optimum stroke thick- 

 ness. Note that the best white on black 

 figures were readable at greater distances 

 than the best black on white ones. 



When Berger investigated digit shape, he 

 did so by varying the form of each digit 

 until its legibility was equivalent to that of 

 the number 8 used as standard. The letter 



