1. Although we perceive a pixel as being white, it is actually comprised of 3 colored sub-pixels. The sub-pixels can be separately addressed in order to triple the potential resolution in the horizontal orientation. Sub-Pixel Text Rendering – Preference, Legibility and Reading Performance James Sheedy, Yu-Chi Tai, Manoj Subbaram, Sowjanya Gowrisankaran, John Hayes College of Optometry, Pacific University, Forest Grove, Oregon ABSTRACT ClearType (CT) is a text rendering technique in which the red, green, and blue sub-pixels are separately addressed to increase text legibility. However, it results in colored borders on characters that can be bothersome. This paper describes five experiments measuring subject preference, text legibility, reading performance, and discomfort symptoms for five implementation levels of CT rendered text. MATERIALS & METHODS CLEARTYPE EXPLAINED FIVE LEVELS OF CLEARTYPE TESTED IN THIS STUDY CONCLUSIONS CONTACT INFORMATION ACKNOWLEDGEMENTS Supported by a grant from the Microsoft Corporation Advanced Reading Technologies Group to Jim Sheedy. Principal Investigator Jim Sheedy T: 503.352.2884 E: jsheedy@pacificu.edu CT level does not affect legibility as measured by threshold sized text. Individual variance in visual acuity and color vision ability do not affect individual preference for CT level. CT level does not affect reading speed. CT level does not affect vision symptoms. Figure 1. Mean preference ratings for CT levels for both 10- and 12-point font. Values greater than 50 indicate preference and values less than 50 indicate non-preference for the correspondent CT level. Statistical differences from neutral value of 50 are indicated (* p <.05; # p <.0001). The result shows that lower CT levels are preferred but not higher CTs, and CT preference (for lower levels) is higher for smaller (10-pt) font, probably due to the better legibility of larger (12-pt) font even without CT rendering. Subjects preferred CT level 1 for 10-pt font and CT level 2 for 12-pt font. Higher CT levels are selected against because of too much perceived color Figure 2. Average rating (0~100) of color, contrast, and clarity as reason for NOT choosing a particular CT level. Statistical testing compared the difference between color/contrast and color/clarity for each pair and significance is indicated (* p < 0.05; ** p < 0.01) next to the contrast and clarity ratings respectively. The results shows that higher CT levels are less preferred because of too much fringing colors and the more preferred lower CT levels, if not chosen, are bothered by lower contrast and clarity. OTHER RESULTS The primary advantage of CT over grayscale as measured is that subjects prefer the appearance of the text, even though functional improvements were not identified. Subjects preferred CT level 1 for 10-pt font and CT level 2 for 12-pt font. The higher preferred level for a larger font type is likely related to the greater number of pixels allocated to each character in a larger font. The reasons for preferring CT are improved perceived contrast and clarity. Despite the perceived improvement in clarity and contrast, CT had no significant effect upon threshold legibility, reading speed, or visual comfort. In summary, subjects preferred low to moderate levels of CT; however no functional improvements were measured. The same thirty subjects (ages 18-38) served in all 5 experiments: 1.Preference for CT level and the reasons for NOT choosing a CT level 2.Does subject perceive color in CT? 3.Does individual visual acuity or color discrimination ability affect CT preference? 4.Does CT affect threshold legibility 5.Does CT affect reading speed and vision symptoms RESULTS Sheedy J, Tai Y, Subbaram M, Gowrisankaran S, Hayes JR. ClearType sub-pixel text rendering: Preference, legibility and reading performance. Displays, 2008;29(2):138-151. Figure 3. Mean rating (SEM bars) of perceived color (“yes” = 1, “no” = 0) as function of CT level. Data are separated into ratings before (naïve) and after (informed) the fringing “color” in CT level 4 was demonstrated. The result confirms the conclusion on Figure 2 that subjects perceive color at the two highest CT levels regardless of the font size. While color is perceived at higher level CTs for both 10- and 12-pt fonts (Figure 3), it is more bothersome for smaller (10-pt) than for larger (12-pt) font, which leads to differnet attitude to CT level 3 in Figure 1. Subjects see color at the 2 highest CT levels