Stephen J. Dain,1,*
David A. Atchison,2
Jeffery K. Hovis,3
and Mei-Ying Boon1
1School of Optometry and Vision Science, University of New South Wales, Sydney, NSW, Australia
2School of Optometry and Vision Sciences and Institute of Health and Biomedical Innovation, Queensland University of Technology, Kelvin Grove, QLD, Australia
3School of Optometry and Vision Science, University of Waterloo, Waterloo, Ontario N2L 5Z5, Canada
Stephen J. Dain, David A. Atchison, Jeffery K. Hovis, and Mei-Ying Boon, "Lighting for color vision examination in the era of LEDs: the FM100Hue Test," J. Opt. Soc. Am. A 37, A122-A132 (2020)
Lighting conditions nominated for color vision testing are many and
varied. The recommendation of CIE color rendering index (CIE CRI) $ {\ge} {90}$ and correlated color temperature of
close to 6500 K is widely made for color vision testing generally.
With the demise of incandescent and fluorescent lighting and their
replacement by light-emitting diodes (LEDs), this is an opportune time
to revisit the recommendation. In this paper, we consider the current
sources, acceptable and unacceptable, and improvements to the
recommendation as it applies to the Farnsworth–Munsell 100 Hue Test
(FM100Hue Test). We conclude that there is no need to treat LEDs as a
special case but propose a modified CRI measure.
You do not have subscription access to this journal. Cited by links are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Figure files are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
You do not have subscription access to this journal. Article tables are available to subscribers only. You may subscribe either as an Optica member, or as an authorized user of your institution.
Change in Total Chromaticity
Area Enclosed Compared with D65
4%
2%
1%
0%
−1%
−2%
−2%
−3%
−3%
−4%
1%
The Color Rendering Index
Calculated Using the 85 FM100Hue Transformed
Chromaticities and D65 as a Fixed Reference
FM100 CRI
89.4
93.7
97.1
100
97.6
95.5
93.8
92.2
90.9
89.7
98.0
Luminance clue is the change in luminous reflectance as a
function of cap number. The lower, the better. In
principle, it should be 0.0%.
Raw chromaticity spacing data are the metrics of the
untransformed cap chromaticity spacing. The closer to the
D65 value, the better.
Transformed chromaticity spacing data are the metrics of
the cap chromaticity spacing transformed to the same
reference white. The closer to the D65 value,
the better.
Table 2.
Correlation Coefficients and Significances (23DF, )
CCT
CIE CRI
FM100 CRI
r
p
r
p
r
p
Luminance Clue
Normals
0.23
ns
−0.60
−0.37
ns
Protanopes
0.73
0.62
0.81
Deuteranopes
−0.69
−0.65
−0.80
Raw Data
Mean
0.41
ns
0.80
0.85
SD
0.78
−0.22
ns
−0.45
SD %
0.85
−0.43
−0.65
Transformed Data
Mean
0.88
0.16
ns
−0.16
ns
SD
0.86
−0.29
ns
−0.52
SD %
0.70
−0.56
−0.69
Area enclosed
−0.76
0.44
0.12
ns
Table 3.
Sources and Illuminants Included in This Study
Name
Type
Measured CIE CRI
Measured CCT K
Macbeth lamp
Filtered incandescent
92.6
6549
Duratest
Halophosphate fluorescent
tube
90.0
7094
GE artificial daylight
92.9
6851
GE BS950 compliant
89.6
7003
Osram 965
Triphosphor/modified
triphosphor fluorescent tube
96.2
6369
Phillips 950
93.5
5429
Thorn D62
76.4
6589
NEC Super cool white
87.2
4384
Kenco daylight
82.6
5962
Sylvania 840
83.4
4220
Phillips 940
92.0
4145
Thorn 860
80.8
6189
Table 4.
Part 1. Data for the Sources: Traditionally Acceptable
Sources
D65
Yuki LED
Duratest
CIE C
Richmond W1
Osram LED Globe
CCT K
6500
7283
7094
6774
6717
6672
CIE CRI
100
95
90
97.5
97.2
86.9
Luminance Clue
Normals
7.10%
7.30%
7.80%
6.90%
7.10%
7.30%
Protanopes
11.90%
12.20%
12.20%
11.70%
12.00%
11.00%
Deuteranopes
5.50%
5.60%
6.30%
5.30%
5.50%
5.90%
Raw Data
Mean
0.0041
0.0042
0.004
0.0042
0.0042
0.004
SD
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
SD %
32%
32%
32%
32%
32%
32%
Transformed Data
Mean
0.0041
0.0041
0.0039
0.0041
0.0041
0.0039
SD
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
SD %
32%
32%
32%
32%
32%
32%
Change in the Area
Enclosed Compared with D65
0%
−1%
−5%
1%
1%
−5%
D65 CRI
100
94.2
88.4
97.5
98.1
86.3
FM100 CRI
100
95.9
87.7
98
98.6
86.5
GE AD
Macbeth
Osram 965
Richmond W2
Xenon arc
Phillips 950
CCT K
6851
6549
6369
6093
5518
5429
CIE CRI
92.9
92.6
96.2
96.9
96.6
93.5
Luminance Clue
Normals
7.40%
7.20%
7.30%
7.00%
6.40%
7.40%
Protanopes
11.60%
11.40%
12.10%
11.50%
11.00%
10.90%
Deuteranopes
5.90%
5.30%
5.60%
5.40%
5.30%
5.90%
Raw Data
Mean
0.0041
0.004
0.0041
0.0041
0.0041
0.0041
SD
0.0013
0.0013
0.0013
0.0013
0.0013
0.0015
SD %
32%
32%
32%
32%
32%
38%
Table 5.
Part 2. Data for the Sources: Traditionally Unacceptable
Sources
GE BS950
Phillips LED Globe
Thorn D62
Richmond F
LED Batten
Thorn 860
CCT K
7003
6777
6589
6358
6272
6189
CIE CRI
89.6
83.9
76.4
75.6
83.8
80.8
Luminance Clue
Normals
7.70%
7.60%
7.60%
7.70%
7.30%
8.80%
Protanopes
11.80%
11.00%
10.40%
10.30%
10.50%
11.30%
Deuteranopes
6.20%
6.20%
6.30%
6.30%
5.90%
7.00%
Raw Data
Mean
0.0041
0.0039
0.0038
0.0038
0.0039
0.0041
SD
0.0013
0.0013
0.0013
0.0013
0.0013
0.0016
SD %
33%
33%
34%
34%
32%
38%
Transformed Data
Mean
0.004
0.0039
0.0038
0.0038
0.0039
0.0042
SD
0.0013
0.0013
0.0013
0.0013
0.0013
0.0016
SD %
32%
33%
34%
34%
33%
38%
Change in the Area
Enclosed Compared with D65
−3%
−6%
−9%
−9%
−5%
−2%
D65 CRI
88.7
83.1
76
76.2
84.5
81.2
FM100 CRI
88.4
82.8
75.8
75.5
84.5
81.8
Table 6.
Kenco daylight
NEC Super cool white
Sylvania 840
Phillips 840
Phillips 940
CCT K
5962
4384
4220
4191
4145
CIE CRI
82.6
87.2
83.4
82.6
92
Luminance Clue
Normals
8.70%
6.90%
7.30%
7.60%
6.80%
Protanopes
11.50%
10.20%
10.30%
10.60%
10.00%
Deuteranopes
7.00%
6.80%
7.80%
7.80%
7.30%
Raw Data
Mean
0.004
0.0038
0.0039
0.0039
0.004
SD
0.0015
0.0014
0.0017
0.0017
0.0017
SD %
38%
36%
43%
44%
42%
Transformed Data
Mean
0.0042
0.0044
0.0045
0.0046
0.0046
SD
0.0016
0.0015
0.0017
0.0018
0.0017
SD %
37%
33%
38%
39%
37%
Change in the Area
Enclosed Compared with D65
−1%
5%
6%
6%
9%
D65 CRI
83.3
75.9
75.7
73.4
74.3
FM100 CRI
83.9
82.3
79
77.4
78.2
Tables (6)
Table 1.
Luminance Clue, Chromatic Spacing, and Area Enclosed for the
FM100Hue Caps as a Function of Illuminant
Change in Total Chromaticity
Area Enclosed Compared with D65
4%
2%
1%
0%
−1%
−2%
−2%
−3%
−3%
−4%
1%
The Color Rendering Index
Calculated Using the 85 FM100Hue Transformed
Chromaticities and D65 as a Fixed Reference
FM100 CRI
89.4
93.7
97.1
100
97.6
95.5
93.8
92.2
90.9
89.7
98.0
Luminance clue is the change in luminous reflectance as a
function of cap number. The lower, the better. In
principle, it should be 0.0%.
Raw chromaticity spacing data are the metrics of the
untransformed cap chromaticity spacing. The closer to the
D65 value, the better.
Transformed chromaticity spacing data are the metrics of
the cap chromaticity spacing transformed to the same
reference white. The closer to the D65 value,
the better.
Table 2.
Correlation Coefficients and Significances (23DF, )
CCT
CIE CRI
FM100 CRI
r
p
r
p
r
p
Luminance Clue
Normals
0.23
ns
−0.60
−0.37
ns
Protanopes
0.73
0.62
0.81
Deuteranopes
−0.69
−0.65
−0.80
Raw Data
Mean
0.41
ns
0.80
0.85
SD
0.78
−0.22
ns
−0.45
SD %
0.85
−0.43
−0.65
Transformed Data
Mean
0.88
0.16
ns
−0.16
ns
SD
0.86
−0.29
ns
−0.52
SD %
0.70
−0.56
−0.69
Area enclosed
−0.76
0.44
0.12
ns
Table 3.
Sources and Illuminants Included in This Study
Name
Type
Measured CIE CRI
Measured CCT K
Macbeth lamp
Filtered incandescent
92.6
6549
Duratest
Halophosphate fluorescent
tube
90.0
7094
GE artificial daylight
92.9
6851
GE BS950 compliant
89.6
7003
Osram 965
Triphosphor/modified
triphosphor fluorescent tube
96.2
6369
Phillips 950
93.5
5429
Thorn D62
76.4
6589
NEC Super cool white
87.2
4384
Kenco daylight
82.6
5962
Sylvania 840
83.4
4220
Phillips 940
92.0
4145
Thorn 860
80.8
6189
Table 4.
Part 1. Data for the Sources: Traditionally Acceptable
Sources
D65
Yuki LED
Duratest
CIE C
Richmond W1
Osram LED Globe
CCT K
6500
7283
7094
6774
6717
6672
CIE CRI
100
95
90
97.5
97.2
86.9
Luminance Clue
Normals
7.10%
7.30%
7.80%
6.90%
7.10%
7.30%
Protanopes
11.90%
12.20%
12.20%
11.70%
12.00%
11.00%
Deuteranopes
5.50%
5.60%
6.30%
5.30%
5.50%
5.90%
Raw Data
Mean
0.0041
0.0042
0.004
0.0042
0.0042
0.004
SD
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
SD %
32%
32%
32%
32%
32%
32%
Transformed Data
Mean
0.0041
0.0041
0.0039
0.0041
0.0041
0.0039
SD
0.0013
0.0013
0.0013
0.0013
0.0013
0.0013
SD %
32%
32%
32%
32%
32%
32%
Change in the Area
Enclosed Compared with D65
0%
−1%
−5%
1%
1%
−5%
D65 CRI
100
94.2
88.4
97.5
98.1
86.3
FM100 CRI
100
95.9
87.7
98
98.6
86.5
GE AD
Macbeth
Osram 965
Richmond W2
Xenon arc
Phillips 950
CCT K
6851
6549
6369
6093
5518
5429
CIE CRI
92.9
92.6
96.2
96.9
96.6
93.5
Luminance Clue
Normals
7.40%
7.20%
7.30%
7.00%
6.40%
7.40%
Protanopes
11.60%
11.40%
12.10%
11.50%
11.00%
10.90%
Deuteranopes
5.90%
5.30%
5.60%
5.40%
5.30%
5.90%
Raw Data
Mean
0.0041
0.004
0.0041
0.0041
0.0041
0.0041
SD
0.0013
0.0013
0.0013
0.0013
0.0013
0.0015
SD %
32%
32%
32%
32%
32%
38%
Table 5.
Part 2. Data for the Sources: Traditionally Unacceptable
Sources