153 lines
8.3 KiB
Plaintext
153 lines
8.3 KiB
Plaintext
|
|
|
|
GUIDE LINES FOR EFFECTIVE COLOR TERMINAL USAGE
|
|
|
|
Based on research done at Tektronix
|
|
and reported in Handshake, 1985.
|
|
|
|
PHYSIOLOGICAL GUIDELINES:
|
|
|
|
Avoid simultaneous display of highly saturated, spectrally extreme
|
|
colors. Reds, oranges, yellows, and greens can be viewed without refocusing,
|
|
but cyan and blues cannot be easily viewed at the same time at as red. To
|
|
avoid frequent refocusing and visual fatigue, extreme color pairs such as red
|
|
and blue or yellow and purple should avoided. However, desaturating
|
|
spectrally extreme colors will reduce the need for refocusing.
|
|
|
|
Pure blue should be avoided for text, thin lines, and small shapes.
|
|
Our visual system is just not set up for detailed, sharp short-wavelength
|
|
stimuli. However, blue does make a good background color and is perceived
|
|
clearly out into the periphery of the visual field. In fact, due to the eye's
|
|
reduced sensitivity to blue, the raster scan pattern in a blue background
|
|
display is less visible then when other background colors are used.
|
|
|
|
Avoid adjacent colors differing only in the amount of blue. Edges
|
|
differing only in the amount of blue appear indistinct.
|
|
|
|
Older viewers need higher brightness levels to distinguish colors.
|
|
|
|
Colors change appearance as ambient light level changes. Displays
|
|
change color under different kinds of light - fluorescent, incandescent, or
|
|
daylight. Appearance also changes as the light level is increased or
|
|
decreased. On the one hand, a change due to an increase or decrease in
|
|
contrast occurs, and on the other, due to the shift in the sensitivity of the
|
|
eye.
|
|
|
|
The magnitude of a detectable change in color varies across the
|
|
spectrum. Small changes in extreme reds and purples are more difficult to
|
|
detect than changes in other colors such as yellow and blue-green. Also, our
|
|
visual system does not readily perceive changes in green.
|
|
|
|
Difficulty in focusing results from edges created by color alone. Our
|
|
visual system depends on a brightness difference at an edge to effect clear
|
|
focusing. Multi-colored images then, should be differentiated on the basis of
|
|
brightness as well as color.
|
|
|
|
Avoid red and green in the periphery of large scale displays. Due to
|
|
the insensitivity of the retinal periphery to red and green, these colors in
|
|
saturated form should be avoided, especially for small symbols and shapes.
|
|
Yellow and blue make good peripheral colors.
|
|
|
|
Opponent colors go well together. Red and green or yellow and blue are
|
|
good combinations for simple displays. The opposite combinations - red with
|
|
yellow or green with blue - produce poorer images.
|
|
|
|
For color deficient observers, avoid single-color distinctions.
|
|
|
|
|
|
PERCEPTUAL GUIDELINES:
|
|
|
|
Not all colors are equally discernible. Perceptually, we need a large
|
|
change in wavelength to perceive and identify a color difference in some
|
|
portions of the spectrum and a small one in other portions.
|
|
|
|
Luminance does not equal brightness. Two equal luminance colors will
|
|
probably appear to have different brightnesses. The deviations are most
|
|
extreme for colors towards the ends of the spectrum (red, magenta, blue).
|
|
|
|
Different hues have inherently different saturation levels. Yellow in
|
|
particular always appears less saturated than other hues.
|
|
|
|
Lightness and brightness are confounded on a color display, but not on
|
|
a printed hard copy. The nature of a display does not allow lightness and
|
|
brightness to be varied independently.
|
|
|
|
Not all colors are equally readable or legible. Extreme care should
|
|
be exercised with text color relative to background colors. Besides a loss in
|
|
hue with reduced size, inadequate contrast frequently results when the
|
|
background and test color are similar. As a general rule, the darker,
|
|
spectrally extreme colors such as red, blue, magenta, brown, etc., make good
|
|
backgrounds while the brighter, spectrum center and desaturated hues produce
|
|
more legible test.
|
|
|
|
Hues change with intensity and background color. In grouping elements
|
|
on the basis of color, be sure that backgrounds or nearby colors do not change
|
|
the hue of an element in the group. Limiting the number of colors and making
|
|
sure they are widely separated in the spectrum will reduce confusion.
|
|
|
|
Avoid the need for color discrimination in small areas. Hue
|
|
information is lost for small areas. Hue information is lost for small
|
|
areas. In general, two adjacent lines of a single pixel width will merge to
|
|
produce a mixture of the two. Also, the human visual system produces sharper
|
|
images with achromatic colors. Thus, for fine detail it's best to use black,
|
|
white, and gray while reserving chromatic color for larger panels or
|
|
attracting attention.
|
|
|
|
COGNITIVE GUIDELINES:
|
|
|
|
Don't overuse color. Perhaps the best rule is to use color sparingly.
|
|
The benefits of color as an attention getter, information grouper, and value
|
|
assigner are lost if too many colors are used. Cognitive scientists have
|
|
shown that the human mind experiences great difficulty in maintaining more
|
|
than 5 to 7 elements simultaneously. So it's best to limit displays to about 6
|
|
clearly discriminable colors.
|
|
|
|
Be aware of the nonlimear color manipulation in video and hard copy.
|
|
At this time algorithms do not exist for translating the physical colors of an
|
|
imaging device into a perceptually structured color set. Video or hard copy
|
|
systems cannot match the human perception and expectations on all fronts.
|
|
|
|
Group related elements by common background color. Cognitive science
|
|
has advanced the notion of set and preattentive processing. In this context,
|
|
you can prepare or set the user for related events by using a common color
|
|
code. A successive set of images can be shown to be related by using the same
|
|
background color.
|
|
|
|
Similar colors connote similar meaning. Elements related in some way
|
|
can convey that message through the degree in similarity of hue. The color
|
|
range from blue to green is experienced as more similar that the gamut from
|
|
red to green. Along the same lines, saturation level can also be used to
|
|
connote strength of the relationship.
|
|
|
|
Brightness and saturation draw attention. The brightest and most
|
|
highly saturated area of a color display immediately draws the viewer's
|
|
attention.
|
|
|
|
Link the degree of color change to event magnitude. As an alternative
|
|
to bar charts or tic marks on amplitude scales, displays can portray magnitude
|
|
changes with the progressive steps of changing color. A desaturated cyan can
|
|
be increased in saturation as the graphed elements increase in value.
|
|
Progressively switching from one hue to another can be used to indicate
|
|
passing critical levels.
|
|
|
|
Order colors by their spectral position. To increase the number of
|
|
colors on a a display requires imposing a meaningful order on the colors. The
|
|
most obvious order is that provided by the spectrum with the mnemonic ROY G
|
|
BIV (red, orange, yellow, green, blue, indigo, violet).
|
|
|
|
Warm and cold colors should indicate action levels. Traditionally,
|
|
the warm (long wavelength) colors are used to signify action or the
|
|
requirement of response. Cool colors, on the other hand, indicate status or
|
|
background information. Most people also experience warm colors advancing
|
|
towards them - hence forcing attention - and cool colors receding or drawing
|
|
away.
|
|
|
|
While these guidelines offer some suggestions, they certainly
|
|
shouldn't be taken as binding under all circumstances. There are too many
|
|
variables in color display, color copying, human perception, and human
|
|
interpretation to make any hard and fast roules. So, by all means, experiment.
|
|
|
|
|
|
|
|
|
|
|