Display Monitor Measurement Methods

Under Discussion by
EIA (Electronic Industries Association) Committee JT-20

Part 2: Color CRT Monitor Performance

Version 2.0 mSYBMUflCm gfÄflMElT Jf

July 12, 1995

Submitted to EIA JT-20 by:

National Information Display Laboratory

P.O. Box 8619
Princeton, NJ 08543-8619
TEL (609) 951-0150
FAX (609) 734-2313
19980415 115
Publication No. 171795-037

The U.S. Government has contract rights in this report through the
National Information Display Laboratory (NIDL)

jyncqoALmra^EüiBDa

Approved for Public Release: Distribution is Unlimited

 National Information Display Laboratory
Publication No. 171795-037

FileName: NIDL95Col717037.MCW Printed: 3/31/98 11:25 AM

National Information Display Laboratory Part 2: Color CRT Display Measurement Methods

July 12,1995

Dear Monitor Professional,

The National Information Display Laboratory would like to present for your consideration Display
Monitor Measurement Methods under discussion by EIA (Electronic Industries Association)
Committee JT-20 Part 2: Color CRT Monitor Performance Version 2.0. We would also like to
express our gratitude to members of the SID Standards and Definitions Committee, the U.S.
Government, members of VESA, NIST, and the display industry, who have provided many highly
constructive and valuable suggestions for improving our earlier version of this document. We
encourage you to review this document and let us know how we can further improve it to meet your
needs.

The goal of this document is to provide practical, tested procedures for obtaining and reporting
consistent and repeatable performance measurements of color CRT displays, especially those used
in imagery analysis. The reported results of these measurements can be used to determine whether
a candidate display can meet the needs of Image Analysts and other critical users, such as
radiologists in the medical imaging community. This document enables users to easily make an
objective comparison when choosing between candidate displays. The goal is not to reinvent, but to
identify, incorporate, and revise as appropriate, measurement standards that other standards-
generating bodies have established.^ The establishment of such measurement and reporting
standards, and promotion of their use, will provide a common language by which display users can
communicate their needs to display manufacturers, better enabling the manufacturers to recognize
the critical performance criteria which must be met in order to successfully provide increasingly
higher quality, more cost-effective color CRT displays to the user.

We believe this document, along with its companion NIDL publication 171795-036 Display
Monitor Measurement Methods under discussion by EIA (Electronic Industries Association)
Committee JT-20 Part 1: Monochrome CRT Monitor Performance Version 2.0, presents a practical
and sufficiently detailed set of measurement procedures for evaluating the performance of color CRT
monitors for imagery applications. It owes much to existing standards documents and to the people
and organizations who created them. Existing standards documents provide established methods for
evaluating color CRT displays targeted for text or graphics applications. This document is intended
to be complementary to existing standards bv addressing the performance issues which effect
whether the Tmage Analyst can reliably and efficiently perceive low contrast detail in color images.

A standard set of measurement evaluation procedures and method of reporting the results for high-
resolution displays benefits the user because it:
• provides a quantifiable means to judge the performance of the display in an objective and
meaningful way;
• eliminates the need for display evaluators to spend time identifying and gathering other test
procedure documents generated by a variety of organizations;

• results in information which is useful to help reduce the time required to select a display that
optimally meets the critical needs of the user;

1
Kelley, E. F., et al., A Survey of the Components of Display-Measurement Standards, SID 95 DIGEST, pp.
637 - 440.
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

• avoids excessive costs of purchasing over specified displays when standardized

measurements are used to provide the exact performance criteria for selecting a display;

• promotes a common language to describe the exact requirements of the user, thus enabling
the display industry to develop monitors that better meet users' needs.

Applying the previous version of the NIDL measurement procedures, NIDL and NIST successfully
completed the first-known round-robin measurements on one CRT display monitor using two very
different sets of measurement equipment but working from the same set of measurement
procedures. The results, identical to within small measurement errors, were presented at SID PI
where manufacturers were encouraged to conduct round-robins with NIST to confirm that their
equipment or monitor data sheets provide information based on standardized measurement practices.

Again, we thank NIST, the members of the SID Standards and Definitions Committee, the U.S.
Government, and contributors from VESA and the display industry who provided many highly
constructive and valuable suggestions for improving our earlier version of this document. We
especially want to state our appreciation to Edward Kelley, Danny Gross, Carlo Infante and Howard
Okamoto. We would also like to acknowledge the dedicated efforts of NIDL retirees, Peter
Wojtowicz and Arthur Miller, for their contributions to the conception and development of this
document.
As we work towards gaining acceptance of each of these sets of test procedures as national and
world standards through established channels, including: SID, EIA, ANSI, ISO and VESA. We
look forward to receiving comments. Please send them in as soon as possible, either by email to
<[email protected]>, or by surface mail to:

Michael Grote
National Information Display Laboratory
P.O. Box 8619
Princeton, NJ 08543-8619

Sincerely,
Dennis Bechis Philip Heyman David Bortfeld Michael Grote

Tel (609) 951-0150 Fax (609) 734-2313 email [email protected]

2
Bechis, D. J., et al., Display-Measurement Round-Robin, SID 95 DIGEST, pp. 641 - 644.
National Information Display Laboratory Preface

Preface
Display Monitor Measurement Methods under discussion by EIA (Electronic Industries
Association) Committee JT-20 Part 2: Color CRT Monitor Performance Version 2.0, along with its
companion document Display Monitor Measurement Methods under discussion by EIA (Electronic
Industries Association) Committee JT-20 Part 1: Monochrome CRT Monitor Performance Version
2.0, are intended to satisfy a need expressed by many users and evaluators of displays in the
government, medical, and commercial communities, as well as by manufacturers of displays and
display systems, for a self-contained, comprehensive document of evaluation procedures for
characterizing the photometric and electrical response related to the performance of monochrome and
color CRT display monitors. These docu-ments enable evaluators in different laboratories to conduct
standardized measurements of aspects of CRT monitor performance and to attain the same results
when measuring the same display.

In turn, this benefits purchasers and manufacturers of CRT monitors by providing a means of
specifying and measuring display performance in a common, meaningful way. By providing
standardized procedures that can be consistently repeated, the documents also benefit users who need
to track the performance of their displays over time and perform quality control adjustments. By
being self-contained and as comprehensive as possible, the documents enable users and
manufacturers to rapidly access, read, and implement tests of display systems, and eliminate the
need of display evaluators to spend time identifying and gathering test procedure documents
generated by a variety of organizations.
The procedures are intended to be applicable to CRT monitor displays of all performance levels,
covering the range from low resolution CGA displays, to high resolution color displays with greater
than 2500 x 2000 pixels resolution.

Each issue mentioned below represents an important need that evaluators and specifiers of display
systems have. Some of these needs are addressed by existing standards documents for particular
display user tasks. Other issues stated here represent needs that call for additional scientific and
human factors studies, and the generation of a separate standards document.

It is important to state what this document is not intended to achieve. The document Display Monitor
Measurement Methods under discussion by EIA (Electronic Industries Association) Committee JT-
20 Part 2: Color CRT Monitor Performance Version 2.0 does not:

1. specify which procedures to use to evaluate the capability of a display system to enable
its user to accomplish a particular task with the maximum of efficiency and reliability;

2. specify or recommend the values or the ranges of values which the display system
must achieve in a specific test procedure or set of test procedures for the display
system to "qualify" for use in a particular task;

3. address which measured performance parameters can be traded off against one
another and by how much and still allow the display to achieve a level of performance
that is acceptable for a particular task;
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

4. rule out or judge alternative test procedures that may have been developed to measure
the same physical effect. Such test procedures may employ different laboratory
measurement equipment, different measurement procedures, and different analysis
techniques. However, since the goal is to provide a common basis for comparison of
monitor performance parameters, discrepancies in measured performance parameters
obtained by different test procedures are unacceptable and need to be resolved.

These test procedures for evaluating critical performance parameters, including resolution, luminance
uniformity, and geometric distortions, have been condensed from an earlier draft of a different NIDL
document Test Procedures for Evaluation of CRT Display Monitors Version 3.1 dated 6/15/92.
Additional test procedures for measuring parameters such as spot size and video amplifier bandwidth
are useful as diagnostics and will be developed for inclusion in a future document.

11
National Information Display Laboratory Sample Evaluation Datasheet

SAMPLE EVALUATION DATASHEET

MANUFACTURER'S DATA
Manufacturer Name . Company ABC
Model # 1A
Monochrome or Color Color
Screen Diagonal 21 inches
Horizontal Scan Rate 89.71 kHz
Vertical Scan Rate 72.00 Hz
Image Size (H x V) 380.0 mm x 284.5 mm (14.96 x 11.20 inches)
Addressable Pixel Number 1600 x1200
Pixel Size 0.237 x 0.237 mm (9.35 x 9.33 mils)
Dot or Stripe Pitch 0.28mm (11.0 mils)

H. MEASURED PERFORMANCE
A. Performance Related to Luminance
Warmup Time 20 minutes to ±1%
Full Screen Luminance 103 cd/m2 (30 fL)
Luminance Uniformity 76.67 - 96.13 cd/m2
Color Coordinates x = 0.282, y = 0.295
Color Uniformity 2.9% in x, 4.1% in y
System Gamma W=2.45 R=2.35 G=2.53 B=2.40
Luminance Stability <12% .
B. Performance Related to Geometry
Waviness < 0.4%
Linearity < 2.6%
Raster Size Stability < 0.1%
Jitter _^__^_ < 0.13 mm (< 5 mils)
Performance Related to Resolution
50% Linewidth (HxV):
center 0.328 x 0.287 mm (12.9 x 11.3 mils)
average periphery 0.340 x 0.284 mm (13.4 x 11.2 mils)
worst location (@ 10:00) 0.399 x 0.290 mm (15.7 x 11.4 mils)
Convergence (HxV):
center 0.061 x 0.109 mm (2.4 x 4.3 mils)
average periphery 0.163 x 0.168 mm (6.4 x 6.6 mils)
worst location (@ 8:00) 0.371 x 0.229 mm (14.6 x 9.0 mils)
Faceplate Reflectivity specular 20%
diffuse 3%
Contrast Ratio 75:1
Halation < 5.6%
1-on/l-off Contrast Modulation (HxV):
center 43 x 31%
average periphery 16 x 38%
worst location (@ 8:00) 6 x 46%
Resolvable Pixels (HxV) (screen average;
@ Cm = 25% 1412x1174
@ Cm = 50% 1047 x 970

111
 Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

D. Reliability and Life Performance

MTBF 10,000 h
Cathode life at 100 cd/m^ luminance 10,000 h

E. Evaluator

Organization Name Testing Lab XYZ

Address Tucson, AZ
Phone (___)___-
Evaluation Dates 3/1/93 to 4/1/93
Equipment Used Photo Research PR-704, Microvision SS 100

Additional Performance Measurements Available: (Y_x_ / N— )

IV
National Information Display Laboratory Measurement Procedure Cross-Reference

CROSS-REFERENCE TO APPLICABLE MEASUREMENT PROCEDURE

A. Performance Related to Luminance

Warm-up Time 4.1 Warm-up Characteristics, Page 11
Full Screen Luminance 3.0 Initial Monitor Setup, page 9
Luminance Uniformity 4.4 Luminance and Color Uniformity, page 12
Color Coordinates 3.0 Initial Monitor Setup, page 9
Color Uniformity 4.4 Luminance and Color Uniformity, page 12
System Gamma 4.2 System Gamma, Page 11
Luminance Stability 4.3 Luminance Stability vs. Fill Factor, page 11

B. Performance Related to Geometry

Waviness 6.1 Waviness, page 35

Linearity 6.2 Linearity, page 35
Raster Size Stability 6.3 Raster Size Stability, page 35
Jitter 6.4 Scan Variability With Time, Page 36

Performance Related to Resolution

50% Linewidth (HxV): 5.1 Linewidth, page 17

center
average periphery
worst location (@ 10:00)
Convergence (HxV): 5.3 Convergence, page 27
center
average periphery
worst location (@ 8:00)
Faceplate Reflectivity specular, diffuse 4.5 Reflectance, page 15
Contrast Ratio 4.7 Contrast Ratio, page 15
Halation 4.6 Halation, page 15
1-on/l-off Contrast Modulation (HxV): 5.2 Contrast Modulation, page 21
center
average periphery
worst location (@ 8:00)
Resolvable Pixels (HxV)(screen average 5.2 Contrast Modulation, page 21
@ Cm = 25%
@ Cm = 50%

D. Reliability and Life Performance

MTBF
Cathode life at 100 cd/m2 luminance 4.8 Lifetest, page 15
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

This page intentionally left blank.

VI
National Information Display Laboratory Contents

CONTENTS

Section 1.0 Introduction 1

Section 2.0 Required Equipment 3

2.1 Examples of Required Equipment 3
2.1.1 Video Generator 3
2.1.2 Spatial Luminance Measurement Equipment 5
2.1.3 Luminance Measurement Equipment 6
2.1.4 Color Measurement Equipment 7
2.2 Equipment Calibration 8
2.2.1 Luminance Calibration 8
2.2.2 Color Calibration 8
2.2.3 Spatial Luminance Calibration 8
9
Section 3.0 Initial Monitor Setup

Section 4.0 Photometric Characterization H

4.1 Warm-up Characteristics H
4.2 System Gamma H
4.3 Luminance Stability vs. Fill Factor 11
4.4 Luminance and Color Uniformity 12
4.5 Reflectance **
15
4.6 Halation
15
4.7 Contrast Ratio
15
4.8 Lifetest 17
Section 5.0 Resolution Characterization I7
5.1 Line Width ■ 17

21
5.2 Contrast Modulation
5.3 Convergence 27
5.4 Moire 31
35
Section 6.0 Geometric Characterization
6.1 Waviness 35
6.2 Linearity 35
6.3 Raster Size Stability 35
6.4 Scan Variability With Time: Jitter, Swim, Drift 36
Jl
Section 7.0 Reporting

vn
 Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

39
References

Appendix A: Definitions of Measurement Terms and Acronyms A-l

Appendix B: Screen Test Points: Color CRT Displays B-l

Appendix C: Colorimetry and the CIELuv Diagram C-l

Vlll
National Information Display Laboratory Introduction

1.0 INTRODUCTION

To be unambiguous, a standardized test procedure for characterizing the photometric response of

a CRT display monitor not only must state what needs to be done, but also describe a particular
method for accomplishing the task and an example including data analysis. As stated in the
Preface, alternative methods of accomplishing the same task using different equipment, different
procedures, and/or different analysis techniques are acceptable if they are equivalent, that is, they
yield the same measured performance parameter(s) when applied to the same CRT display.

To achieve this purpose, each procedure is structured in the following manner:

• Objective states not only what physical performance parameters are to be measured,
but also provides a rationale for the measurement (why this is an important
measurement for evaluating a CRT monitor).

• References states what standardized measurement procedures may already exist for
this performance metric. Furthermore, References states whether the measurement
procedure contained herein is identical to the standardized existing procedure or
whether the existing measurement procedure has been modified or abandoned.
Reasons for modifications or abandonment are provided.

• Equipment provides a list of necessary generic measurement devices. Specific

examples of commercial measurement equipment are provided in Section 2.0 of this
document, without qualification or endorsement, and without the intent of providing a
complete list.
• Procedure describes a procedure for generating the data necessary for characterizing
this particular aspect of display performance.

• Data provides a description of the output of fully implementing the Procedure, and in
many cases, a representative sample sheet for tabulating the data is included.

• Analysis explains how the data is treated mathematically and presents the equations
used to arrive at the final measured performance parameter or set of parameters.

• Output presents a sample of the Analysis results.

This document is divided into the following sections:

• Section 2.0 lists the generic equipment required to conduct the measurements,
provides examples of the required photometric and electrical equipment, and
provides instructions on the calibration of the photometric equipment. Section 2.0
also provides optional measurement procedures to characterize the electrical
performance of the video signal generator required in nearly all measurements of the
CRT display monitor performance.

• Section 3.0 provides a standardized procedure for setting up the monitor to be

evaluated.
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

• Section 4.0 provides additional procedures to characterize the photometric

performance of the CRT display monitor system.

• Section 5.0 contains procedures for performing measurements to determine the most
important performance characteristic of a display monitor, its resolution.

• Section 6.0 provides additional procedures to characterize the geometric performance

of the CRT display monitor system.

• Section 7.0 contains a sample of a proposed standard monitor specification sheet for
reporting measurement results in a format which enables the display user to make an
objective and consistent comparison between candidate monitors.
National Information Display Laboratory Required Equipment

2.0 REQUIRED EQUIPMENT

2.1 Examples of Required Equipment

2.1.1 Video Generator

Function: The video generator drives the monitor with input signals for displaying a wide
variety of test patterns.

Test patterns: In addition to the test patterns specified in Section 2.1.1 Video Generator in
NIDL Publication No. 171795-036 Display Monitor Measurement Methods
under discussion by Elk (Electronic Industries Association) Committee JT-20
Part 1: Monochrome CRT Monitor Performance, Version 2.0, the
programmable video generator used for color displays should also be capable
of displaying the following test patterns using multiple-pixels wide lines to
increase shadowmask sampling, thus providing better accuracy when
determining the position of the center of luminance (centroid) of the line.

5% screen
height

5% screen
width 3 to 7 pixels wide lines
to increase shadow
mask sampling

V-grillefor measuring horizontal linearity H-grillefor measuring vertical linearity

Figure 2.1.1-1

Three-line test patterns with 3 to 7-pixels wide lines to increase

shadowmask sampling for measuring waviness and misconvergence.
Figure 2.1.1-2
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

Crosshatch pattern of single-pixel wide lines for visually assessing misconvergence.

Figure 2.1.1-3

V-grille and H-grille patterns with variable width lines 1 to 5-pixels wide
and variable width spacings 1 to 5-pixels wide for assessing moire.

Figure 2.1.1-4
National Information Display Laboratory Required Equipment

2.1.2 Spatial Luminance Measurement Equipment

Function: Spatial luminance measurements are required to evaluate contrast modulation,

convergence, waviness, linearity, jitter, and raster size of an image displayed on a
structured screen such as a shadowmask CRT. Spatial luminance measurements
may be performed by slowly moving the image past a slit or round aperture of a
photometer either: 1) by mechanically moving the photometer past the image
(scanning slit method), 2) by electrically moving the image past the photometer
(moving beam method), or 3) by focusing the image onto a spatially calibrated
photodiode or CCD linear array [ARP1782]. Some measurement equipment
companies provide systems that enable one to move the beam across the raster in
sub-pixel increments using programmable time delay circuits to alter the
relationship between vertical and horizontal synchronization pulses.

References: ARP1782, Photometric and Colorimetric Measurement Procedures for

Airborne Direct View CRT Displays, SAE, January 1989.

TEPAC 105-9, Line Profile Measurements in Shadow Mask and Other

Structured Screen Cathode Ray Tubes, ETA, January 1987.

Kawakami, Y., and Palmer, W., High-Accuracy Convergence Measurements,

SID Seminar Lecture Notes, Vol. I: May 6,1991, pp. M5/1 - M5/26.

Procedure: For color CRT displays, follow recommendations outlined in Section 2.1.2
Spatial Luminance Measurement Equipment in NIDL Publication No. 171795-
036 Display Monitor Measurement Methods under discussion by EIA
(Electronic Industries Association) Committee JT-20 Part 1: Monochrome CRT
Monitor Performance, Version 2.0.

Accuracy: In High-Accuracy Convergence Measurements, Kawakami and Palmer discuss

the relationship between phosphor pitch and image magnification, and the
consequential effects on measurement accuracy when using multi-element
photosensors such as CCD arrays Improved repeatability and accuracy are
gained hv adjusting the sensor optir.s so that the magnification of the image is
such that the phosphor pitch, when imaged onto the sensor, is an integer multiple
of the CCD element pitch.
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

2.1.3 Luminance Measurement Equipment

2.1.3.1 Photometer

Function: Required to measure average area luminance on the display for evaluating raster
luminance, system gamma, luminance stability, luminance uniformity, and
halation.

Procedure: For color CRT displays, follow recommendations outlined in Section 2.1.3.1
Photometer in NIDL Publication No. 171795-036 Display Monitor Measurement
Methods under discussion by EIA (Electronic Industries Association) Committee
JT-20 Part 1: Monochrome CRT Monitor Performance, Version 2.0.

2.1.3.2 Micro-Photometer

Function: The micro-photometer is required for measuring luminance of peaks and valleys
of grille patterns used for evaluating display contrast modulation. For color
CRTs, the micro-luminance measurement field must be less than 1/2 phosphor
stripe width, or less than 1/2 the diameter of a phosphor dot.

References: TEPAC 105-9, Line Profile Measurements in Shadow Mask and Other
Structured Screen Cathode Ray Tubes, EIA, January 1987.

Procedure: For color CRT displays, follow recommendations outlined in Section 2.1.3.2
Micro-Photometer in NIDL Publication No. 171795-036 Display Monitor
Measurement Methods under discussion by EIA (Electronic Industries
Association) Committee JT-20 Part 1: Monochrome CRT Monitor Performance,
Version 2.0.

2.1.3.3 Luminance Standards

Function: Provide a standard luminance source traceable to NIST for calibrating photometer
and spectroradiometer.

Procedure: For color CRT displays, follow recommendations outlined in Section 2.1.3.3
Luminance Standards in NIDL Publication No. 171795-036 Display Monitor
Measurement Methods under discussion by EIA (Electronic Industries
Association) Committee JT-20 Part 1: Monochrome CRT Monitor Performance,
Version 2.0.
National Information Display Laboratory Required Equipment

2.1.4 Color Measurement Equipment

Function: Color measurement equipment may include preferably a spectroradiometer or

colorimeter for measuring chromaticity coordinates of display. For color
displays, chromaticity coordinates of white and individual red, green, and blue
primaries are measured to evaluate color spatial uniformity and variation of white
purity as a function of luminance. Use a spectroradiometer with minimum 400
to 700 nm wavelength range. If a colorimeter is used, it should be calibrated to a
spectroradiometer.

Procedure: For color CRT displays, follow recommendations outlined in Section 2.1.4
Color Measurement Equipment in NIDL Publication No. 171795-036 Display
Monitor Measurement Methods under discussion by EIA (Electronic Industries
Association) Committee JT-20 Part 1: Monochrome CRT Monitor Performance,
Version 2.0.

2.2 Equipment Calibration

2.2.1 Luminance Calibration

Objective: Calibrate luminance measurement equipment using a NIST (NBS) traceable

luminance standard as a reference.

Procedure: For color CRT displays, follow recommendations outlined in 2.2.1 Luminance
Calibration in NIDL Publication No. 171795-036 Display Monitor Measurement
Methods under discussion by EIA (Electronic Industries Association) Committee
JT-20 Part 1: Monochrome CRT Monitor Performance, Version 2.0.

2.2.2 Color Calibration

Objective: Calibrate color measurement equipment using a luminance standard

as a reference.

Procedure: For color CRT displays, follow recommendations outlined in 2.2.2 Color
Calibration in NIDL Publication No. 171795-036 Display Monitor Measurement
Methods under discussion by EIA (Electronic Industries Association) Committee
JT-20 Part 1: Monochrome CRT Monitor Performance, Version 2.0.
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

2.2.3 Spatial Luminance Calibration

Objective: Calibrate spatial luminance measurement equipment using a Ronchi grating as a

reference. For the particular optical setup of a CCD or photodiode array, establish
the effective size of a single module element in the object plane.

Procedure: For color CRT displays, follow recommendations outlined in 2.2.3 Spatial
Luminance Calibration in NIDL Publication No. 171795-036 Display Monitor
Measurement Methods under discussion by EIA (Electronic Industries
Association) Committee JT-20 Part 1: Monochrome CRT Monitor Performance,
Version 2.0.

8
National Information Display Laboratory Initial Monitor Setup

3.0 INITIAL MONITOR SETUP

Objective: BRIGHTNESS, CONTRAST, FOCUS, and CONVERGENCE controls are

adjusted following the recommendations of the monitor manufacturer.

Procedure: Display three-beam white to perform the procedures outlined in Section 3.0 Initial
Monitor Setup in NIDL Publication No. 171795-036 Display Monitor
Measurement Methods under discussion by EIA (Electronic Industries
Association) Committee JT-20 Part 1: Monochrome CRT Monitor Performance,
Version 2.0.

Convergence
Leave convergence as factory configured or follow factory procedures if they are
provided.

Color temperature
Leave color temperature as factory configured or follow factory procedures if they
are provided. Color displays should be measured for color temperature for an
output luminance level of 50% Lmax as measured at screen center using a test
pattern consisting of a square box 1% total screen area in size. The input count
necessary to achieve 50% Lmax must be determined at center screen using the
same target. The x, y coordinates are measured directly using a spectroradiometer
or colorimeter for comparison with manufacturers specifications.
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

This page intentionally left blank

10
National Information Display Laboratory Photometric Characterization

4.0 PHOTOMETRIC CHARACTERIZATION

4.1 Warm-up Characteristics

Objective: Determine the transient photometric characteristics of the display device. Find the
minimum time required for the display to stabilize to a predetermined luminance
level.

Procedure: Follow the recommendations outlined in Section 4.1 Warm-up Characteristics in

NIDL Publication No. 171795-036 Display Monitor Measurement Methods
under discussion by EIA (Electronic Industries Association) Committee JT-20
Part 1: Monochrome CRT Monitor Performance, Version 2.0.

4.2 System Gamma

Objective: Measure the nonlinear photometric characteristic (output luminance versus input
drive) of the display as it spans L^ to Lmax. Tonal variations such as intensity
and shades of gray are essential for imagery tasks. Typical values for gamma
range between 1.5 to 3. The gamma value determines the resolution required for
the digital to analog converter and look up table for digital to luminance mapping.

Procedure: Display individual red, green, and blue beams and three-beam white to perform
the procedures outlined in Section 4.2 System Gamma in NIDL Publication No.
171795-036 Display Monitor Measurement Methods under discussion by EIA
(Electronic Industries Association) Committee JT-20 Part 1: Monochrome CRT
Monitor Performance, Version 2.0.

4.3 Luminance Stability as a Function of Fill Factor

Objective: The purpose of this test procedure is to determine the luminance stability of the
display as a function of the fill factor. Halation, phosphor saturation, video
amplifier low frequency response and thermal effects may contribute to
luminance instability.

Procedure: Display three-beam white to perform the procedures outlined in Section 4.3
Luminance Stability as a Function of Fill Factor in NIDL Publication No.
171795-036 Display Monitor Measurement Methods under discussion by EIA
(Electronic Industries Association) Committee JT-20 Part 1: Monochrome CRT
Monitor Performance, Version 2.0.
11
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

4.4 Luminance and Color Uniformity

Objective: Measure the variability of luminance and colorimetry (chromaticity coordinates)

of the white point and of individual primaries (red, green and blue) as a function
of time, intensity, and spatial position on the screen. Variability of luminance and
color purity impacts the total number of discriminable gray steps. Color CRTs
may exhibit shifts in luminance and color purity caused by nonuniformities in the
shadowmask as a result of design tolerances, doming and blister. Individual
phosphor components may saturate or exhibit a change in persistence as a
function of beam intensity.

References: ASTM El 336 - 91, Standard Test Method for Obtaining Colorimetric Data
from a Video Display Unit by Spectroradiometry.

ASTM El 341 Standard Practice for Obtaining Spectroradiometric Data from

Radiant Sources for Colorimetry.

MPR 1990:8 Test Methods for Visual Display Units, Swedish National Board for
Measurement and Testing, December, 1990.

TEP105-11-A, Measurement of the Color of CRT Screens, EIA, 1988.

TEP116-B, Optical Characteristics of CRTs, EIA, 1989.

ISO/TC159/SC4 N201 Ergonomics Requirements for Office Work with Visual

Display Terminals (VDTs). Part 8, Requirements for Displayed Colours,
December, 1990.

CIE No. 38, Radiometrie and Photometric Characteristics of Materials and

Their Measurement, 1977.

CIE No. 63, The Spectroradiometric Measurement of Light Sources, 1984.

CIE No. 69, Methods of Characterizing Illuminance Meters and Luminance

Meters, Publication, 1987.

Tchen, H., et al., Photocolorimetric Measurements of TV and HDTV Display

Devices, Application Notes, SID'92, pp. 75-78.

VESA Standard: Display Specifications and Test Procedures, Version 1.0, Rev.
1.0, 3 October 1994.

Equipment: Video generator

Photometer
Spectroradiometer or Colorimeter
12
National Information Display Laboratory Photometric Characterization

Test pattern: Full screen flat field as shown in Figure 4.4-2.

20 mm

Full screen
flat field test pattern Nine screen test locations

Figure 4.4-1 Figure 4.4-2

Procedure: Measure temporal variation of luminance and CIE chromaticity coordinates at

screen center as a function of intensity for a time span of 2 hours for each video
input count level. Display a full screen flat field shown in Figure 4.4-1 for four
different brightness levels with input counts corresponding to 25%, 50%, 75%,
and 100% of Lmax as determined in Section 3.0. Make measurements every
minute during the first 30 minutes, then at longer intervals when changes occur
more gradually. For example, the measurement interval should be every 1 minute
for the first 30 minutes, then every 5 minutes up to the first hour, and then every
10 minutes after the first hour. Between each set of measurements return the
display to Lmin, and let the display stabilize and the shadow mask cool down for
20 minutes before proceeding to the next set of measurements.

Measure the temporal variation of luminance and color as a function of spatial

position on the screen by repeating these measurements at each of the locations
depicted in Figure 4.4-2. The locations of corner screen test points are arbitrarily
defined. Guidance for selecting alternative test point locations is provided in
Appendix B. In addition, measure variations at any other critical points on the
screen as deemed by the experimenter.
The procedures described above should be carried out in a darkened environment
such that the stray luminance Lament diffusely reflected by the screen in the
absence of electron-beam excitation is less than 0.003 cd/m2 (1 mfL). In any
case, the value of Lambient should be measured and recorded.

Data: Tabulate the luminance and 1931 CIE chromaticity coordinates (x, y) of red,
green blue, and white for each intensity level as a function of time at each of the
nine locations depicted in Figure 4.4-2. Additionally, note the location of any
additional points that are measured along with the corresponding luminance
values.

13
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

Sample data:
Table 4.4-1. Sample Data for Spatial Uniformity of Luminance and
Chromaticity Coordinates of the White Point for Four Luminance
Settings Taken at Nine Screen Positions (in % of Lmax).
100% 7S%
j -«max
max
POSITION X y cd/m 2
X y cd/m2
center 0.275 0.289 101.8 0.274 0.290 76.5
2 0.272 0.285 91.4 0.272 0.288 69.1
3 0.277 0.292 97.7 0.280 0.291 73.7
4 0.271 0.280 64.2 0.272 0.282 49.1
6 0.272 0.282 73.7 0.273 0.285 55.6
8 0.273 0.288 79.8 0.275 0.290 60.8
9 0.277 0.289 88.8 0.277 0.290 68.1
10 0.274 0.292 87.9 0.272 0.292 66.2
12 0.273 0.287 83.7 0.271 0.288 62.8

50% 25%
Lmax Lmax
center 0.275 0.291 51.3 0.276 0.296 25.2
2 0.273 0.287 45.6 0.272 0.289 22.4
3 0.277 0.293 48.7 0.278 0.297 23.9
4 0.275 0.286 30.6 0.278 0.289 14.7
6 0.274 0.288 37.8 0.276 0.292 18.8
8 0.277 0.295 41.4 0.276 0.300 20.9
9 0.277 0.293 46.3 0.277 0.298 23.4
10 0.272 0.290 44.2 0.270 0.293 22.0
12 0.270 0.289 41.4 0.269 0.292 20.1

Analysis: Determine the time required for each display to achieve white stability at the
specified intensity levels at each spatial location.
Determine the color gamut for the display at the each of the four intensities.
Compute luminance and color variation over the screen relative to screen center
for each target luminance level.
The variation may be expressed as chroma difference in terms of CIELuv Aw ,
Av * units [Defined in Appendix C].
AC*MV =[(A«*)2+(Av*)2]1/2
Optionally, the variation may be expressed in terms of CffiLab or CIELuv AE
units which quantify the combined perceived lightness and color of the display
[Appendix C].

Accuracy: Luminance measurement accuracy is ± 10% [MPR 1990:8]. Photometer

accuracy is at least ±5%. Use a measurement field covering at least 10 scanning
lines for raster luminance measurement [TEPAC Publ. 105].
Luminance uniformity measurement results obtained by different laboratories
agree to within 2%. Uncertainty of the luminance uniformity measurement is 2%.
Color uniformity measurement results obtained by different laboratories agree to
within 1%. Uncertainty of the color uniformity measurement is 1%.
14
National Information Display Laboratory Photometric Characterization

4.5 Reflectance
Objective: Determine the reflection coefficients of the display screen to enable the user to
calculate whether the display will provide contrast modulation required for
performing tasks in a particular ambient light.

Procedure: Measure the display reflectance according to the procedures outlined in Section
4.5 Reflectance in NIDL Publication No. 171795-036 Display Monitor
Measurement Methods under discussion by EIA (Electronic Industries
Association) Committee JT-20 Part 1: Monochrome CRT Monitor Performance,
Version 2.0.

4.6 Halation
Objective- Measure the contribution of halation to contrast degradation. Halation is a
phenomenon in which the luminance of a given region of the screen is increased
by contributions from surrounding, more luminous, areas caused by light
scattering within the phosphor layer and internal reflections inside the glass
faceplate. Determine what percentage of light is piped into the dark areas as a
function of the extent of the surrounding light areas. The results can be used in
combination with the measured values of the reflection factor to obtain estimates
of contrast ratio that include the effects of both halation and diffuse reflection of
ambient light.
Procedure: Follow the recommendations outlined in Section 4.6 Halation in NIDL
Publication No. 171795-036 Display Monitor Measurement Methods under
discussion by EIA (Electronic Industries Association) Committee JT-20 Part 1:
Monochrome CRT Monitor Performance, Version 2.0.

4.7 Contrast Ratio

Objective: Calculate large area contrast ratio using measured results for halation and screen
reflectance. Measure the contribution of halation to contrast degradation.

Procedure: Follow the recommendations outlined in Section 4.7 Contrast Ratio in NIDL
Publication No. 171795-036 Display Monitor Measurement Methods under
discussion by EIA (Electronic Industries Association) Committee JT-20 Part 1 :
Monochrome CRT Monitor Performance, Version 2.0.

4.8 Lifetest
Objective: Over time, measure the changes in nonlinear photometric characteristic (output
luminance versus input drive) of the display as it spans Lmin to Lmax. Determine
changes in gamma value over time.
Procedure: Follow the recommendations outlined in Section 4.8 Lifetest in NIDL Publication
No. 171795-036 Display Monitor Measurement Methods under discussion by
EIA (Electronic Industries Association) Committee JT-20 Part 1: Monochrome
CRT Monitor Performance, Version 2.0.
15
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

This page intentionally left blank

16
National Information Display Laboratory Resolution Characterization

5.0 RESOLUTION CHARACTERIZATION

5.1 Line Width

Objective: Characterize line width profile of the CRT display. For color displays, characterize
line width profile in the presence of the shadow mask for individual beams and for
combined beams to include effects of misconvergence. This measurement is used
for the determination of resolution. The relationship between vertical and horizontal
line width is an indication of the beam spot (pixel) shape. Luminance calibration is
not required for this measurement unless absolute luminance is measured.

References: TEP105-7-A, Line Profile Measurements in Monochrome CRTs, EIA, 1987.

TEP105-9, Line Profile Measurements in Shadowmask and Other Structured

Screen CRTs, EIA, 1987.

TEP192, The Glossary of CRT Terms and Definitions, EIA, 1984.

TEB25, A Survey of Data Display CRT Resolution Measurement Techniques,

EIA, 1985.

TEB27, Relating Display Resolution and Addressability, EIA, 1988.

Beaton and Farley, Display Measurement Issues in the ANSI/HFS100-1988

Standard, SID'91 Digest, p. 648.

Farrell, Richard J., and Booth, John M., Design Handbook for Imagery
Interpretation Equipment, Boeing Aerospace Co., 1984.

ARP1782, Photometric and Colorimetric Measurement Procedures for

Airborne Direct View CRT Displays, SAE, January 1989.

ISO/TC159/SC4 WG2/N219 Final Text for IS 9241, Part 3—Visual Displays,

December 1990.

VESA Standard: Display Specifications and Test Procedures, Version 1.0, Rev.
1.0,3 October 1994.

Equipment: Video generator

Spatially calibrated CCD or diode optics
Photoptic filter adapter

Test pattern: Use the three-line pattern in Figure 5.1-1 to display vertical or horizontal lines
each 1-pixel wide. Position lines in video to each test location as shown in Figure
5.1-2. Clock positions reference test positions which are symmetrically spaced 20
mm inside the addressable screen edges. The locations of corner screen test
points are arbitrarily defined to be severe enough to adequately evaluate the
resolution capabilities of CRTs used to display high pixel-density imagery.

17
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

1-pixel wide lines for monochrome and color CRTs

Three-line grille test patterns

Figure 5.1-1

20 mm

20 mm

«-8D

Nine screen test locations

Figure 5.1-2

Procedure: Use moving beam method, scanning slit method, or photodiode (or CCD) array to
measure full-width at half-maximum (FWHM) vertical and horizontal white line
widths as displayed using single-line video test pattern. Use CCD or diode optic
module with photopic filter to measure width at 5% and 50% peak luminance of
line profiles for horizontal and vertical lines displayed at 50% Lmax at nine (9)
screen locations: center, ends of major and minor axes, and four corners.
Optionally, measure individual red, green, and blue lines.

18
National Information Display Laboratory Resolution Characterization

Screen Sampling

The measurement of linewidth on a structured color CRT screen is complicated by

the sampling of the mask and phosphor pattern. In order to account for the varia-
tion in apparent linewidth caused by the sampling (and the phase differences be-
tween image information and phosphor pattern), and to represent what a viewer
actually sees on the screen, linewidth measurements at any screen position are taken
on single-pixel wide lines successively moved by one-pixel width to seven different
locations. Use a programmable video generator to reposition lines over a range of at
least ±3 pixel locations nearest the test point to obtain line width mea-surements
over a full cycle of shadow-mask sampling.

Curve fitting

For color displays, the luminance profile of a line measured through the apertures
of a shadowmask consists of luminance peaks like those shown in Figure 5.1-3
with spatial frequency equal to the phosphor pitch. For this kind of data, use
straight line segments to connect the luminance peaks, then determine the FWHM
linewidth on the reconstructed profile [ISO 9241, Part 3]. Optionally, a Gaussian
curve fitted to the data points along the measured luminance distribution profile
may be used, but with extreme caution!

Optional measure of line width with brightness: Obtain line width measurement
for individual beams and white at screen center only for four video input levels
corresponding to: 25%, 50%, 75%, and 100% Lmax as determined in Section 3.0
Initial Monitor Set-up.

Data: Sample deflected Line Width Measurements (measure both horizontal and vertical lines)

Table 5.1-1 Linewidth (FWHM) in mm for White Lines at 50% Lmax, Measured with
Photopic Filter.

H AH V AV H AH AV H AH AV
w 0.399 0.0641 0.290 0.03?| 0.249 0.058 0.325 0.0181 | 0.348 0.0381 0.315 0.058

w 0.348 0.069 0.254 0.043 0.328 0.117 0.287 0.041 0.373 0.079 0.254 0.048

w 0.368 0.0461 0.262 0.023 [ | 0.246 0.0891 0.310 0.051 | 0.394 0.061 0.259 0.051 "|

Screen positions are indicated by position in sub-table.

H = Horizontal width of vertical lines (mean of seven runs).
V = Vertical height of horizontal lines (mean of seven runs).
AH, AV = range = max-min of seven runs. W = White (with photopic filter).

19
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

Analysis: Quantify line width uniformity by reporting vertical and horizontal line widths
separately (H x V). Report line width data in same format as Table 5.1-II below.

50% Linewidth (HxV):

center 0.328 x 0.287 mm (12.9 x 11.3 mils)
average periphery 0.340 x 0.284 mm (13.4 x 11.2 mils)
worst location (@ 10:00) 0.399 x 0.290 mm (15.7 x 11.4 mils)

Table 5.1-II Sample reported linewidth data

Worst location is defined as the test location on the screen where the maximum
combined horizontal and vertical linewidth occurs. The combined linewidth is the
magnitude calculated using square-root of the sum of the squares:
(H2 + v2) 1/2
where H is horizontal linewidth and V is vertical linewidth of white lines.
Optional line width measurements at 50% level in RAR ITEB271
Report horizontal and vertical line widths as Resolution-Addressability Ratio (RAR)
at nine screen positions:
RAR = R/A
where:
R = full width at half maximum of the line.
A = inter-pixel distance

Output: See sample Figure 5.1-3.

Full Width
at Half Max

L /M Lk,
Sample display of line width measurement using diode optics and
gaussian curve fit on color screen
Figure 5.1-3

Accuracy: For shadowmask CRTs, a reported linewidth value ±10% is acceptable [ISO
9241, Part 3]. Linewidth measurement results obtained by different laboratories
agree to within 0.05mm. Uncertainty of the linewidth measurement is 9%.

20
National Information Display Laboratory Resolution Characterization

5.2 Contrast Modulation

Objective: Quantify luminance spatial frequency effects as a function of spatial frequency
and screen position through measurements of luminance profiles. Measure
contrast modulation in both horizontal and vertical directions as a function of
spatial frequency over a range of luminance levels of the CRT display. This
measurement is required for the determination of resolution, and provides
information on the large and small signal handling capabilities of the display
system. Spatial resolution capabilities of the display may or may not be closely
correlated with the addressability. Spatial resolution in color displays is a function
of the addressability, the size and shape of the spot, misconvergence, misregistry,
and phosphor pitch.
References: ARP1782, Photometric and Colorimetric Measurement Procedures for
Airborne Direct View CRT Displays, SAE, January 1989.

Austin, R.L., and Otto, HJ., A Technique for Precision Measurement of Display
Spatial Profiles, SID '87 Digest, pp. 206 - 210.

Beaton and Farley, Display Measurement Issues in the ANSI/HFS100-1988

Standard, SID'91 Digest, p. 648.
Briggs, S. J., Soft Copy Display of Electro-Optical Imagery, SPIE Vol. 762
Electro-Optical Imaging Systems Integration (1987), pp 153-170.
ISO 9241 Part 3—Visual Displays, "Visual Display Terminals (VDTs) Used for
Office Tasks—Ergonomie Requirements; and Part 3—Visual Displays, "Final
Text" as of June 1992.
TEB25, A Survey of Data Display CRT Resolution Measurement Techniques,
EIA, 1985.
TEB27, Relating Display Resolution and Addressability, EIA, 1988.
TEP105-9, Line Profile Measurements in Shadowmask and Other Structured
Screen CRTs, EIA, 1987.
Miller, A. and Murray, W.J., Fourier-Transform Methods for Determination of
Contrast-Modulation Indices from Luminance-Variation Data SID 94 Digest, pp
531-534.

O'Callaghan, J.P., and Veron, H., A New Approach for Analyzing the Visual
Resolution Characteristics of Shadow-Mask CRT Monitor Configurations,
SID'89 Digest, pp. 208-211.

Veron, H., The Measurement of Resolution of Shadow-Mask CRTs, SID'85

Digest, pp. 298-301.

VESA Standard: Display Specifications and Test Procedures, Version 1.0, Rev.
1.0, 3 October 1994.

21
 Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

Test Pattern: Use grille patterns of alternating pixel groupings forming either vertical lines or
horizontal lines on the CRT display as shown in Figure 5.2-1.

V-grille H-grille

Full screen level-p/level-v grille test patterns

Figure 5.2-1

20 mm

20 mm

Nine screen test locations.

Figure 5.2-2

Grille patterns must be displayed using input counts for Level-p and Level-v pixels
which are previously determined in Section 3.0 Initial Monitor Setup. Each
successive pattern of horizontal and vertical lines exhibits increasingly higher spatial
frequency. Use video patterns of lines 3-pixels at Level-p, 3-pixels Level-v then 2-
pixels at Level-p, 2-pixels at Level-v, and 1-pixel at Level-p, 1-pixel at Level-v.
Also, display a flat field pattern at Level-p and measure the contrast modulation
perpendicular to the scan lines.

Procedure: Follow the recommendations outlined in Section 5.2 Contrast Modulation in

NIDL Publication No. 171795-036 Display Monitor Measurement Methods
under discussion by EIA (Electronic Industries Association) Committee JT-20
Part 1: Monochrome CRT Monitor Performance, Version 2.0.

Shadow mask
From the specified dimensions of the monitor raster, the number of pixels in each
direction, and the shadow mask pitch, calculate, for both the horizontal and
vertical direction, the spacing between adjacent rows or columns of sampling
apertures in the shadow mask, and the spacing between adjacent pixels. Calculate,
in each direction, the ratio of the pixel spacing to the sampling aperture spacing.

22
National Information Display Laboratory Resolution Characterization

Example: The data cited here are for a commercial 19-inch diagonal monitor
with a 1280 x 1024 pixel landscape-format display, a raster with dimensions
13.48" x 10.79" (342.4 x 274.1 mm), and a delta shadow mask with 12.20 mils
(0.31 mm) pitch.

Pixel spacines:
13.48" x 1000/1280 = 10.53 mils (0.2675 mm) horizontal
10.79" x 1000/1024 = 10.54 mils (0.2677 mm) vertical
Sampling aperture spacings:
12.20sin(60°) = 10.57 mils (0.2684 mm) horizontal
12.20cos(60°) = 6.10 mils (0.155 mm) vertical

Spacing ratios:
10.53/10.57 = 0.996 horizontal
10.54/6.10 =1.727 vertical

The occurrence of ratios close to simple rational numbers, and particularly, of

ratios close to unity as in the above example, indicate that the monitor may exhibit
severe moire or other aliasing effects. Figure 5.2-3 shows a luminance
distribution with one-pixel on and one-pixel off along the horizontal direction
measured for this monitor. The alternating sequence of high and low peaks
indicates little if any shift in the phase between the pixel pattern and the aperture-
mask sampling over the entire range of the measured pattern. The pixel and
sampling spacings are equal to within 1 percent. The ratio between the
amplitudes of alternating peaks, and consequently the effective contrast
modulation at any screen position, is a sensitive function of, e.g., the centration of
the raster or the ambient magnetic field. A ratio of the pixel spacing to the
sampling aperture spacing of close to 3/2 can also lead to especially undesirable
behavior. In this case components at the pixel spatial frequency (e.g., the scan
line pattern) and the sampling spatial frequency can interact to produce a response
at the spatial frequency corresponding to one-pixel on and one-pixel off, even
though no such component is present in the input.

Use, as appropriate, either the Lpeak/Lvalley, or Fourier-transform procedure

specified in Section 5.2 Contrast Modulation in NIDL Publication No. 171795-
036 Display Monitor Measurement Methods under discussion by EIA
(Electronic Industries Association) Committee JT-20 Part 1: Monochrome CRT
Monitor Performance, Version 2.0 to measure the contrast modulation with
white (tricolor) patterns. A photopic filter should be employed to match the
spectral response of the photodetector to that of a standard observer. Figure 5.2-4
shows a measured luminance distribution for which the Lpeak/Lvalley procedure
is difficult or inappropriate to apply, but for which the Fourier-transform
procedure may be used in a straightforward way. Figure 5.2-3 shows a
distribution for which either procedure is appropriate.

23
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

Measured luminance distribution for a one-pixel

on, one-pixel offpattern on a monitor with nearly
equal pixel and aperture sampling spacings.
Figure 5.2-3

If it is possible to implement on the particular monitor being tested, a moving-

beam method, such as that described in TEPAC Publication TEP105-9, Line
Profile Measurements in Shadowmask and Other Structured Screen CRTs, EIA
1987 may be used to eliminate shadow mask sampling effects. Figure 5.2-5
illustrates data that can be obtained by this method. The results of such measure-
ments have to be applied with caution when the ratio between pixel and sampling
aperture spacings is close to a simple rational number.

Measured luminance distribution for which

Fourier-transform procedure can be applied in
a straightforward way.
Figure 5.2-4

24
National Information Display Laboratory Resolution Characterization

Optionally perform contrast modulation measurements for individual red, green,

and blue beams by the procedure selected above. A photopic filter is not required
for these measurements.

If it is desired to use a single procedure for all contrast modulation methods, the
Fourier-transform method is recommended because it is applicable in the greatest
variety of cases, and because the results of the procedure, which are analogous to
the optical Modulation Transfer Function, provide a quantitative interface with
application-specific evaluations of monitor performance.

Lpeak Lpeak

Lvalley Lvalley

Sample display of contrast ratio Sample display of contrast ratio

using moving beam method to measurements using diode array optics
eliminate shadowmask sampling on shadowmask color screen

Figure 5.2-5 Figure 5.2-6

The sample contrast modulations shown in Figure 5.2-7 are not fully realized
because of the presence of moire caused by aliasing between the image and the
shadowmask. Because contrast modulation values are, calculated for the
maximum peak and minimum valley luminance levels as indicated in the sample
data shown, thev do not include the degrading effects of aliasing.

25
 Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

120 T 200

90 -- 150

100 --

50 --'

500 500

Monitor A Monitor B
0.28mm dot pitch 0.26mm dot pitch
Contrast modulation for sample luminance profiles
(1-pixel at level 50,1-pixel at level 0)
for monitors exhibiting moire due to aliasing.
Figure 5.2-7

Moire occurs when the phosphor pitch is too large in comparison to the pixel
size. Studies have shown that a phosphor pitch of about 0.6 pixels or less is
required for adequate visibility of image information without interference from
the phosphor structure. In Figure 5.2-7, the Monitor A phosphor pitch is 0.90
pixels as compared with 0.84 pixels in Monitor B. Moire is more visible in
Monitor A, appearing as long stripes where contrast modulation has been
degraded. In Monitor B, moire is less visible, appearing as "fish-scales" where
contrast modulation has been reduced. Even though Monitor A exhibits a greater
loss of contrast modulation from the presence of moire on 1 -on/1 -off vertical
grille patterns, there is little or no visual impact when aerial photographic images
are displayed. NBDL experts in human vision and psychophysics were unable to
discern presence of moire on either monitor when grayscale imagery was
displayed.

Accuracy: Mask structures in color CRTs may prohibit sufficient beam sampling to
accurately determine Lpeak and Lvaiiey intensity levels. Improved accuracy can be
achieved by using either the Fourier-transform procedure, or the moving-beam
method of measurement to eliminate shadow mask sampling effects. The
Fourier-transform procedure provides an estimate of contrast modulation that
includes the possible reduction of modulation depth at the measurement site on
the screen due to the effects of shadowmask aliasing. The moving-beam method,
on the other hand, does not capture these effects. Contrast modulation results
obtained by different laboratories agree to within 5%. Uncertainty of the contrast
modulation measurement is 5%.

26
National Information Display Laboratory Resolution Characterization

5.3 Convergence

Objective: Characterize three-beam convergence of the color CRT display. Misconvergence

effects the true appearance of colored features in an image and can contribute to
the loss of resolution of the display.

References: Wojtowicz, Utilization of Symmetry in CRT/Yoke Manufacture and Analysis,

SID'91 Digest, p.886.

ISO/TC159/SC4 WG2/N219, Final Text for ISO 9241, Part 3—Visual

Displays, December 1990.

Kawakami, Y., and Palmer, W., High-Accuracy Convergence Measurements,

SID Seminar Lecture Notes, Vol. I: May 6,1991, pp. M5/1 - M-/26.

VESA Standard: Display Specifications and Test Procedures Version 1.0, Rev.
1.0, 3 October 1994.

Equipment: • Video generator;

• Spatially calibrated CCD or photodiode array optic module;
• Calibrated X-Y translation stage.

Test pattern: For visual examination, inspect convergence using Crosshatch pattern consisting
of vertical and horizontal lines each 1-pixel wide, spaced 5% of the screen
width/height apart as shown in Figure 5.3-2. For optical measurement at standard
test locations shown in Figure 5.3-1, use V-grille and H-grille video patterns
consisting of vertical and horizontal lines each 3 to 5-pixels wide as illustrated in
Figure 5.3-4. Use of lines greater than 1 or 2-pixels increases luminance profile
sampling and improves measurement repeatability on shadowmask CRTs.

Clock positions shown in Figure 5.3-3 reference test positions which are
symmetrically spaced to encompass the entire viewable display area. The D-
points are located 20 mm in from the top/bottom and side edges of the viewable
phosphor screen. The center point is located at the mechanical center of the
screen. All other points including the A points are equally spaced in-between
along imaginary lines forming a 5x5 grid not shown.

The locations of corner screen test points are arbitrarily defined to be severe
enough to adequately evaluate the resolution capabilities of CRTs used to display
high pixel-density imagery.

27
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

20 mm

20 mm

Nine screen test locations.

Figure 5.3-1

o O o1 o
10D ?1 12 2D

o n o o o
2:30
9:30 10A 12A 2A
o9 o 3°A
o3
9°A Center

o 0 O
6A 4°A
o
3:30
8:30 8A
0 o 0 o
80 „„7 R -4QJ

Crosshatch test pattern Twenty-five Screen test locations

Figure 5.3-2 Figure 5.3-3

V-grille H-grille
5-pixel wide lines
Grille test patterns
Figure 5.3-4

Procedure: Use optic module and x-y translation stage to measure x and y separations
between centroids of 3 to 5-pixel wide vertical and horizontal lines for blue with-
respect-to red, and optionally green with-respect-to red. Measure vertical and
horizontal misconvergence at nine standard screen test points, optionally twenty-
five (25) screen test points shown in Figure 5.3-3 using grille video test patterns
illustrated in Figure 5.3-4. Visually examine overall convergence performance
using a finer Crosshatch pattern as shown in Figure 5.3-2. Record measurements
at any screen location where significant misconvergence is apparent, but not
characterizable at the standard twenty-five screen test locations.

28
National Information Display Laboratory Resolution Characterization

BBROK: atU<l-G>Mi «i]s<B-G>iii »ils(R-J)ini

LATEST -4.81 -.122 -1.41 -.B35 -3.48 -.B8B

Sample display of convergence measurements

using diode optics on a color dot screen
Figure 5.3-5

Data: Misconvergence measurements are arranged by screen position in Table 5.3-1,

and are reported in mm and in pixels.

Table 5 3-1. Sample Blue-to-Red Misconvergence. Misconvergence (in mm) is at the

Maximum Luminance for a Three-pixel Line. Values Given are the Mean of Five
Runs.

Position rlonzoniai aeparauuu ^

mm pixels mm pixels
-0.061 -0.3 0.109 0.5
Center
Minor Axis 6 -0.003 -0.01 0.201 0.8
12 0.038 0.2 0.089 0.4

Major Axis 3 -0.130 -0.5 0.213 0.9

-0.089 -0.4 0.201 0.8
9
Corners 2 -0.086 -0.4 0.287 1.2
4 -0.231 -1.0 0.170 0.7

8 -0.371 -1.6 0.229 1.0

10 -0.343 -1.4 0.142 0.6

Average 0.150 0.65 0.182 0.77

magnitude

do 12 2\
Key to clock
9 CENTER 3 positions used
8 4 in the tables
I 6 J
29
 Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

Analysis: Quantify convergence uniformity by reporting vertical and horizontal misconver-

gence separately (H x V). Report convergence data in same format as Table 5.3-II
below.

Table 5.3-II Sample reported convergence data

Convergence (HxV):
center 0.061 x 0.109 mm (2.4 x 4.3 mils)
average periphery 0.163 x 0.168 mm (6.4 x 6.6 mils)
worst location (@ 8:00) 0.371 x 0.229 mm (14.6 x 9.0 mils)

Worst location is defined as the test location on the screen where the maximum
combined horizontal and vertical convergence errors occur. The combined error is
the magnitude calculated using the square-root of the sum of the squares:

(H2 + V2) 1/2

>

where:
H is horizontal separation
V is vertical separation of blue with-respect-to-red lines.

Optionally, symmetrize misconvergence data [Wojtowicz]. Quantify design

misconvergence with symmetric components of convergence data. Quantify
manufacturing and set-up errors with asymmetric components of convergence
data.

Accuracy: Convergence measurement results obtained by different laboratories agree to

within 0.05 mm. Uncertainty of the convergence measurement is 0.05 mm.

Large-size CRTs typically exhibit 1 mm or more misconvergence error as a

function of inter-beam space charge repulsion. Misconvergence effects due to
space charge are not present when red, green and blue beams are displayed
sequentially. Misconvergence effects due to space charge can be measured,
however, by using CRT measurement systems which are capable of determining
misconvergence of a displayed three-beam white line test pattern. Such systems
are commercially available.

30
National Information Display Laboratory Resolution Characterization

5.4 Moire

Objective: For color displays, qualitatively characterize luminance spatial effects resulting
from the mixing of the spatial frequencies of the shadow mask pitch and the
beam spot size by making visual observations of various test patterns on the
display. Moire has the capacity to deteriorate resolution of high frequency
information in images on the display by superimposing shaded wavy patterns on
the image. Moire patterns occur when the beam 5% spot size is less than 2.5 to
2.6 times the mask pitch [TEPAC 105-9]. Other studies have shown that a
phosphor pitch of 0.6 pixels or less is required to maintain the visibility of image
information without interference from the phosphor structure at any spatial
frequency.

References: Wittke, Moire Considerations in Shadow Mask Picture Tubes, SID'87 Digest,
p.347.
Benson, K. Blair., Television Engineering Handbook, McGraw-Hill Book Co.,
1986,p.l27.
TEPAC 105-9, Line Profile Measurements in Shadow Mask and Other
Structured Screen Cathode Ray Tubes, EIA, January 1987.

Procedure: Inspect the display for variations in luminance levels that are visually noticeable
as alternating lighted and shaded stripes of Moire patterns of spatial frequencies
between < 1 and ~ 25 cycles per inch of display screen area. An example of
Moire patterns on a color CRT is shown in Figure 5.4-1 below.

W'l".!:11
I1
I -.1-

ii

Ii-hil!1'H1',<I "i \v';;'•.■'*/?'''j"J"'f

Moire patterns on a color CRT

Figure 5.4-1.

Inspect for Moire using green beam at 50% Lmax using the input count level
determined in Section 3.0. Record occurrences at each of nine standard screen
locations where noticeable Moire is observed.

31
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

A procedure is described below for making modulation-depth measurements on

moire patterns if desired, using an array detector (e.g., CCD or diode).

Preliminary statements should be made about the following items:

- Illumination from a single beam should be used in making the
measurements.
- The depth of the modulation depends on the beam profile. One might
wish therefore to make measurements:
- at different beam currents;
~ at different parts of the screen;
~ for each of the three beams.
- Provisions should be made to determine the location of the display pattern
that is imaged on the detector array. For example, strips of opaque
adhesive tape can be placed on the face of the display to establish
boundaries, or to serve as fiducials.
The detector should be positioned at a sufficient distance from the display to
obtain an image of appropriate magnification. (For example, the magnification
might be such that the image of the desired region is about one-half the size of the
detector array.) The use of a zoom lens would permit various magnifications to
be chosen without repositioning the detector head.

Luminance profile of moire using focused optics contains high

spatial frequencies ofphosphor pitch.
Figure 5.4-2

Initially, the lens on the detector head should be focused to produce a sharp image
of the display as shown in the sample output in Figure 5.4-2. In this condition
there are aliasing effects due to the beating of spatial frequencies in the screen

32
National Information Display Laboratory Resolution Characterization

pattern (electron beam and shadowmask) with those due to the spacing of the
elements in the detector array. The detector output at this stage is unsuitable for
measurement of the modulation depth of the moire pattern.

The fine structure can be significantly reduced by integrating over a suitable

number of neighboring elements in the detector array. This is conveniently
accomplished by slightly defocusing the lens. As the lens is gradually defocused,
the aliasing pattern quickly disappears, and, as shown in the sample output in
Figure 5.4-3, the detector output closely represents the intensity variation of the
moire pattern. (The moire patterns of interest have much smaller spatial frequen-
cies than those of the electron beam pattern, the shadowmask, or the detector
array elements, and are little affected by the defocusing. The required degree of
defocusing is not critical; it is observed that as the focus is changed, the modula-
tion depth of the detector output remains on a "plateau".)

valley

Profile of moire using slightly defocused optics to filter out

high spatial frequency effects ofphosphor pitch.
Figure 5.4-3

As the amount of defocusing is further increased, the modulation depth in the

detector output begins to decrease. The lens should then be brought back towards
the sharp focus setting until the "plateau" condition is restored It should be con-
firmed that the number of peaks in the detector output corresponds to what is
perceived by eye on the bounded region of the display. The modulation depth of
the detector output is then measured and recorded.

33
 Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

Data: Record occurrences of visually noticeable Moire for grille test patterns displayed
at 50% Lmax (using the input count level determined in Section 3.0) over a range
of spatial frequencies.

Table 5.4-1. Sample sheet for Recording Occurrences of Moire

Noticeable Moire
indicate by V

Grille: 3 on x 3 off 2 on x 2 off 1 on x 1 off

Location
Center
2 V
3 V
4
6
8 V
9
10 V V
12

Analysis: TBD

34
National Information Display Laboratory Geometrie Characterization

6.0 Geometrie Characterization

6.1 Waviness

Objective: Measure center (green) beam position on the CRT display to quantify effects of
waviness which causes nonlinearities within small areas of the display distorting
nominally straight features in images, characters, and symbols. The presence of
waviness also causes large area raster distortions including pincushion, trapezoid
(keystone), rotation and orthogonality.

Procedure: For color CRT displays, measure waviness of the green beam according to the
procedures outlined in Section 6.1 Waviness in NIDL Publication No. 171795-
036 Display Monitor Measurement Methods under discussion by EIA
(Electronic Industries Association) Committee JT-20 Part 1: Monochrome CRT
Monitor Performance, Version 2.0.

6.2 Linearity

Objective: Measure the relation between the actual position of a pixel on the screen and the
addressed position to quantify effects of raster nonlinearity. Nonlinearity can be
expressed as a variable pixel density Nonlinearity of scan degrades the
preservation of scale in images across the display.

Procedure: For color CRT displays, measure linearity of the green beam according to the
procedures outlined in Section 6.2 Linearity in NIDL Publication No. 171795-
036 Display Monitor Measurement Methods under discussion by EIA
(Electronic Industries Association) Committee JT-20 Part 1: Monochrome CRT
Monitor Performance, Version 2.0.

6.3 Raster Size Stability

Objective: Assess the stability of the high voltage supply by measuring the change of raster
size as a function of the average luminance of the display. Since more current is
required at higher luminance, the accelerating voltage will decrease, and thus the
size of the raster will increase, if the power supply has less than perfect
regulation.

Procedure: For color CRT displays, measure raster size variation of the three-beam white flat
field test pattern according to the procedures outlined in Section 6.3 Raster Size
Stability in NIDL Publication No. 171795-036 Display Monitor Measurement
Methods under discussion by EIA (Electronic Industries Association) Committee
JT-20 Part 1: Monochrome CRT Monitor Performance, Version 2.0.

35
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

6.4 Scan Variability With Time: Jitter, Swim, Drift

Objective: Measure amplitude and frequency of variations in beam spot position of the CRT
display. Quantify the effects of perceptible time varying raster distortions: jitter,
swim, and drift. The perceptibility of changes in the position of an image depend
upon the amplitude and frequency of the motions which can be caused by
imprecise control electronics or external magnetic fields.

Procedure: For color CRT displays, measure raster jitter of the green beam test pattern
according to the procedures outlined in Section 6.4 Scan Variability With Time:
Jitter, Swim, Drift in NIDL Publication No. 171795-036 Display Monitor
Measurement Methods under discussion by EIA (Electronic Industries
Association) Committee JT-20 Part 1: Monochrome CRT Monitor Performance,
Version 2.0.

36
National Information Display Laboratory Reporting

7.0 REPORTING

Objective: A standardized performance certification report for each monitor type that enables users
to:
- rapidly ascertain the monitor's performance capabilities;
- easily compare the capabilities of different monitors;
-- judge the capability of the monitor for meeting their image evaluation needs.

SAMPLE EVALUATION DATASHEET

L MANUFACTURER'S DATA

Manufacturer Name Company ABC

Model # 1A
Monochrome or Color Color
Screen Diagonal 21 inches
Horizontal Scan Rate 89.71 kHz
Vertical Scan Rate 72.00 Hz
Image Size (H x V) 380.0 mm x 284.5 mm (14.96 x 11.20 inches)
Addressable Pixel Number 1600 x 1200
Pixel Size 0.237 x 0.237 mm (9.35 x 9.33 mils)
Dot or Stripe Pitch 0.28mm (11.0 mils)

n. MEASURED PERFORMANCE

A. Performance Related to Luminance

Warmup Time 20 minutes to ±1%
Full Screen Luminance 103 cd/m2 (30 £L)
Luminance Uniformity 76.67 - 96.13 cd/m2
Color Coordinates x = 0.282, y = 0.295
Color Uniformity 2.9% in x, 4.1% in y
System Gamma W=2.45 R=2.35 G=2.53 B=2.40
Luminance Stability <12%

B. Performance Related to Geometry

Waviness < 0.4%
Linearity < 2.6%
Raster Size Stability < 0.1%
Jitter < 0.13 mm (< 5 mils)

37
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

c. Performance Related to Resolution

50% Linewidth (HxV):
center 0.328 x 0.287 mm (12.9 x 11.3 mils)
average periphery 0.340 x 0.284 mm (13.4 x 11.2 mils)
worst location (@ 10:00) 0.399 x 0.290 mm (15.7 x 11.4 mils)
Convergence (HxV):
center 0.061 x 0.109 mm (2.4 x 4.3 mils)
average periphery 0.163 x 0.168 mm (6.4 x 6.6 mils)
worst location (@ 8:00) 0.371 x 0.229 mm (14.6 x 9.0 mils)
Faceplate Reflectivity specular 20%
diffuse 3%
Contrast Ratio 75:1
Halation < 5.6%
1-on/l-off Contrast Modulation (HxV):
center 43 x 31%
average periphery 16 x 38%
worst location (@ 8:00) 6 x 46%
Resolvable Pixels (HxV) (screen average)
@ Cm = 25% 1412x1174
@Cm-50% |1047x970

liability and Life Performance

MTBF 10,000 h
Cathode life at 100 cd/m2 luminance 10,000 h

E. Evaluator —.
Organization Name Testing Lab XYZ
Address Tucson, AZ
Phone ( )
Evaluation Dates 3/1/93 to 4/1/93
Equipment Used Photo Research PR-704, Microvision SS 100

F. Additional Performance Measurements Available: (Y_X_ / N_ )

38
National Information Display Laboratory References

REFERENCES

ARP1782, Photometric and Colorimetric Measurement Procedures for Airborne Direct View CRT
Displays, SAE, January 1989.

ASTM E1336 - 91, Standard Test Method for Obtaining Colorimetric Data from a Video Display Unit by
Spectroradiometry.

ASTM E1341, Standard Practice for Obtaining Spectroradiometric Data from Radiant Sources for
Colorimetry.

Austin, R.L., and Otto, H.J., A Technique for Precision Measurement of Display Spatial Profiles, SID
'•87 Digest, pp. 206-210.

Beaton, R.J., and DeVilbiss, CA., Assessment Method for the ANSI/HFES 100-1988 Guideline on
Display Jitter, SID 89 DIGEST, pp. 216-219.

Benson, K. Blair., Television Engineering Handbook. McGraw-Hill Book Co., 1986, p. 12.7.

CIE 38 Radiometrie and Photometric Characteristics of Materials and Their Measurement, 1977.

CIE 63, The Spectroradiometric Measurement of Light Sources, 1984.

CIE No. 69, Methods of Characterizing Illuminance Meters and Luminance Meters, Publication, 1987.

ISO 9241, Part 3—Visual Displays: Visual Display Terminals (VDTs) Used for Office Tasks -
Ergonomie Requirements; Part 3—Visual Displays: Final Text as of June 1992.

ISO/TC159/SC4 N201, Ergonomics Requirements for Office Work with Visual Display Terminals
(VDTs), Part 8—Requirements for Displayed Colours, December 1990.

Kawakami, Y., and Palmer, W., High-Accuracy Convergence Measurements, SID Seminar Lecture
Notes, Vol. I: May 6,1991, pp. M5/1—M5/26.

MPR 1990:8, Test Methods for Visual Display Units, Swedish National Board for Measurement and
Testing, December 1990.

Myodo, O., et. al, An Analytical Evaluation of Moire Pattern for Color CRT Displays, Japan Display
'86,'pp. 52-55.

OCallaghan, J.P., and Veron, H., A New Approach for Analyzing the Visual Resolution Characteristics
of Shadow-Mask CRT Monitor Configurations, SID'89 Digest, pp. 208 - 211.

Tchen, H., et al, Photocolorimetric Measurements of TV and HDTV Display Devices, Application Notes,
' SID'92, pp. 75-78.

39
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

TEP105-9, Line Profile Measurements in Shadowmask and Other Structured Screen CRTs, EIA, 1987.

TEP105-11-A, Measurement of the Color of CRT Screens, EIA, December, 1988.

TEP116-B, Optical Characteristics of CRTs, EIA, 1989.

Veron, H., The Measurement of Resolution of Shadow-Mask CRTs, SID'85 Digest, pp. 298 - 301.

VESA Standard: Display Specifications and Test Procedures Version 1.0, Rev. 1.0, 3 October 1994.

Wittke, Moire Considerations in Shadow Mask Picture Tubes, SID'87 Digest, p.347.

Wojtowicz, Utilization of Symmetry in CRT/Yoke Manufacture and Analysis, SID'91 Digest, p.886

40
National Information Display Laboratory Appendix A: Definitions of Terms and Acronyms

APPENDIX A
DEFINITIONS OF MEASUREMENT TERMS AND ACRONYMS
Addressability—the inter-pixel distance (TEP192), "addressability" defines how precisely one can
position the electron beam spot on the screen.

ARP Aerospace Recommended Practice associated with SAE.

ASTM American Society for Testing and Materials

Candelas per square meter (cd/m2)—metric unit of measurement of luminance. 1 cd/m2 = 0.2919
Footlambert (fL).

CD color difference

CIE Commission Internationale De l'Eclairage (International Commission on Illumination)

Contrast Modulation, (Cm)—measure of the luminance ratio between the "lit" and "unlit" portions of
grille patterns. A grille pattern of given frequency is considered to be resolved when the
contrast modulation is greater than 20%.

CRT Cathode Ray Tube

Contrast Transfer Function, (CTF)—curve of contrast modulation values plotted as a function of

spatial frequency.

Convergence—measure of the separation in landing positions of separate beams directed toward the same
point on the screen. The main misconvergence errors involve blue-to-red separations and
green-to-red-blue-average separations (coma), and are measured in both horizontal and
vertical directions. Misconvergence errors greater than the pixel size degrade resolution and
cause spurious color fringes at edges in images.

EIA Electronic Industries Association

Fill Factor the ratio of "lit" to total active area of a display screen.

Hish Voltage Regulation—measure of variation of overall raster size with changes in luminance (as
caused by changes in electron beam current). High voltage output from a well regulated
supply will not change with beam current.

ISO International Standards Organization

Linearity measure of the preservation of the scale of image contents across the screen.

Luminance Stability as a Function of Fill Factor—measure of variation in luminance as a function of

the fraction of screen area that is being "lit" (i.e., the fraction of the frame time in which the
electron beam is actually turned on).

A-l
 Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

MPR [Swedish] National Board for Measurement and Testing

Moire a spurious repetitive luminance pattern observed on displays as a result of the beating of the
phosphor structure pattern against the pattern of information imposed on the screen. Moire
is highly visible when the frequencies of the two patterns are comparable. The presence of
Moire destroys contrast modulation, and hence, resolution.

NIDL National Information Display Laboratory

NIST National Institute of Standards and Technology (formerly NB S)

NBS National Bureau of Standards

Pixel picture element

Resolution measure of the ability to discriminate picture detail; i.e., ability to distinguish two adjacent
spots on the screen.

Scan Jitter rapid motions of raster on the screen face caused by instabilities in the monitor circuitry.

SID Society for Information Display

Spatial Uniformity of Luminance-measure of how luminance varies across the screen. Luminance
should be as uniform as possible.

SWEDAC Swedish Board for Technical Accreditation (formerly MPR)

System Gamma—the slope of the curve in a log-log plot of output luminance vs. input drive.

TEB TEPAC Engineering Bulletin (published by EIA)

TEPAC Tube Engineering Panel Advisory Council (associated with EIA)

TBD To be determined

Warm-up Characteristic-the time required for the luminance to stabilize at some predetermined value;
e.g. ±1%.

Waviness measure of the degree of curvature of (or departure from) nominally straight lines on the
display screen. Principle components of waviness are:
- Pincushion—a quadratic distortion
- Gullwing—a quartic distortion.

A-2
National Information Display Laboratory Appendix B: Screen Test Points

APPENDIX B
SCREEN TEST POINTS: COLOR CRT DISPLAYS

The number and location of screen test points display. Preferably include four additional
depends on the performance parameter being measurement points located at both ends of
evaluated, the size of the display screen, and the major and minor axes of the screen. Typical
intended use of the display. This appendix is screen-points are located at mechanical center
intended to provide guidance for the use of and 20 mm inward from the viewable edges of
alternative test points when appropriate. the screen as depicted in Figure B-l. When
using alternate screen locations, consider that for
20 mm -*-. p*-
the typical CRT, luminance usually degrades
JJ-J monotonically between screen center and the
periphery.
20 mm To exemplify the effect of test-point location on
luminance uniformity, luminance was measured
at screen center and at points along the 2 o'clock
corner diagonal of a 19V color CRT monitor
and the results plotted in Figure B-3. This
example illustrates the extent to which color
Nine screen-test-points for routine CRT luminance uni-formity can vary depending
upon exactly where the corner screen
CRT display evaluation. measurement location is defined.
Figure B-l.
o o o o o
10D 11 12 1 2D
Color CRT Display Considerations o o o o o
9:30 10A 12A 2A 2:30
CRT luminance and color performance typically o o3
is best at screen center and degrades with screen o9 o
9A
o
Center 3A

position towards the periphery. Further, CRT o o o o

4A
o
3:30
performance in the corner screen locations is 8:30 8A 6A

generally worse than the performance along the o8D o7 o6 o5 o4D

major and minor axes of the screen. CRTs
exhibit significant asymmetries in performance Twenty-five screen-test-points to assess
errors due to manufacturing and setup; convergence and raster distortions during
therefore, a complete evaluation of a given
performance parameter requires measurements CRT design and factory setup.
be taken in each of the four quadrants of the Figure B-2.
CRT screen so that the contributions of each of
the asymmetrical parts are taken into account.
Increased numbers of test locations are war-
Luminance, color uniformity, and color ranted for thorough evaluation of very high
convergence measurements resolution or large screen size color CRT
displays. It is customary to use an arrange-ment
Use five screen-points located at screen center of twenty-five or more measurement points as
and four corners, minimum. On the typical depicted in Figure B-2, making measurements
CRT, center and corner locations exhibit the best for evaluating raster distortions such as inner
and worst performance, respectively.
pincushion or gullwing, and misconvergence
Measurements at all four corners are required to
distinguish asymmetrical properties of the during the CRT design and factory setup stages.

B-l
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

-0.87-inch dia.
measurement field
on full flat field
target
u -0.87-inchdia.
a measurement field
on character-size
3
target

Diagonal distance from screen

center (inches)

Figure B-2. Luminance measured on character and full screen flat field targets using a character-sized
photometer measurement field plotted as a function of screen position. Center-to-corner luminance ratio
depends on exact screen locations. Luminance values for full screen target average 7% higher than for
character-size target.

Even when the CRT is properly designed,

variations in color purity can result when
shielding and degaussing fail to compensate for
the effects of external magnetic fields.
Misregistry variations across the tube face may
occur without obvious patterns, and so, it is not
uncommon for manufacturers to assess color
purity by measuring misregis-try errors at
ninety-nine test-points or more as depicted in
Ninety-nine screen test-points arranged in an
Figure B-4.
11x9 grid for measuring misregistry to
assess color purity on a 27V4 x 3 aspect
ratio CRT.
Figure B-4.

B-2
National Information Display Laboratory Appendix C: Colorimetry and the CLIELuv Diagram

APPENDIX C
COLORIMETRY AND THE CIELuv DIAGRAM

The following discussion is meant to pro-vide the video concepts of hue and saturation
the reader with an introduction to the concepts (Sproson, p. 19).
and terms of colorimetry sufficient to
understand their use in this document. The
reader is encouraged to refer to the texts listed
below for a more thorough coverage of this
topic. 1.0 -r

The measurement values for the display colors 0.8 --

are typically presented in terms of 1931 CIE
x, y colorimetry coordinates. These G
coordinates refer to the horseshoe-shaped 0.6 -
diagram in Fig. C-l.
NTSC ^*
0.4 -
This method of describing colorimetry, while Gamut ^
widely used, has two drawbacks. First of all,
the space is not visually uniform, i.e. colors 0.2 -
equally spaced on the CIE diagram are not
perceived as being equally different. Secondly, B\
00 - K— 1 r-
this presentation does not deal with differences
0.0 0.2 0.4 0.6 0.8
in luminance.
x
Several approaches to perceptually more
uniform spaces, that also include luminance,
have been developed, including the CIELab, CIE 1931 Chromaticity Diagram showing
CIELuv, and Munsell spaces (Judd and the gamut of the NTSC phosphor set. (Typical
Wyszecki, p. 281). Of these the CIELuv space monitor gamuts are much smaller; the green point, in
is particularly appropriate for discus-sing particular is much farther down and to the right).
monitor performance, because it has a Fig. C-l
reasonably straightforward relationship with

Table C-l
Chromaticity Coordinates and Correlated Color Temperatures
for Some Common White Points

Correlated CIE 1931 x,y CIE 1976 u',v'

CIE Color Temp Chromaticity Coordinates Chromaticity Coordinates
niuminant (K) X y u' v'
A 2856 0.448 0.407 0.256 0.524
B 4874 0.348 0.352 0.214 0.485
c 6774 0.310 0.316 0.201 0.461
D55 5503 0.332 0.348 0.204 0.481
6504 0.313 0.329 0.198 0.468
D65
7504 0.299 0.315 0.193 0.459
D75

C-l
Part 2: Color CRT Display Measurement Methods National Information Display Laboratory

The CIELuv space is built on the perceptually

more uniform CIE 1976 (u\ v') colorimetry
huv= Hue Angle
diagram obtained from x and y by the
View Direction following transformation:
u' = 2x/(6y-x + 1.5)
v' = 4.5y/(6y-x + 1.5)

The CIELuv space translates the (u', v')

diagram to be centered on the white point (u'0,
v'o), the new coordinates are designated («*, v*K
The white point depends on the display used;
C*uv=Chroma typical white points are listed in Table B-l
along with their correlated color temperature.
Illuminant C is the NTSC standard white; Ü65
is the PAL standard.
Defining the maximum white luminance as Yo
(following the traditional use of the tri-stimulus
value Y for luminance), an approximately
The (u*,v)*plane centered on the D65 white uniform lightness parameter, L*, is defined
point showing the NTSC gamut. using a cube-root relation. The complete
Fig. C-2a. transform set is:
L* = 116 (Y/Y0)l/3 -16 Y/Yo>0.008856

L* = 903.29 (Y/Y0) Y/Yo<0.008856

L = Lightness

White
w* = 13L* (u'-u'o)
v* = 13L*(v'-v'0)
Limit of
Limit of
Green Gun
The CIE 1976 chroma C*uv, psychometric
Red + Blue
Guns saturation suv and hue-angle huv for the
CIELuv space are:

C\v = (a*2 + v* V/2

Available Suv -C uv i*-1
Gamut
Saturated
Colors huv = tan -Hv*/u*)

— Magenta Green —>* Figure C-2 shows the (u*,v*) plane and a
1 1 1 1 r-s—l
Black C uv schematic cut along the L* axis for a typical
color CRT. Note that for low L* the width of
A cut along the lightness axis showing how the CIELuv cone increases linearly with L*.
luminance affects the available gamut. Fig. C- The maximum value of Y, corresponding to
2a shows the viewpoint.

Fig C-2b.

C-?
National Information Display Laboratory Appendix C: Colorimetry and the CLIELuv Diagram

where the Ä's in the brackets are the differences differences. (MacAdam has shown that a truly
uniform space must be non-Euclidean; CIELuv
between the u*, and v* coordinates of the two
is a Euclidean approximation). Psychophysical
colors being compared. studies indicate that one AE unit corresponds
Because the space is relatively uniform per- rough-ly to a just noticeable difference; two
ceptually, distances in the space between L*, u, points in the space separated by more than 5
*v* points are a reasonable measure of color AC units are clearly distinguishable.

References:
D.B. Judd and G. Wyszecki, Color in Business, Science and Industry, Wiley, New York, 1975.
D. L. MacAdam, Color Measurement, Springer-Verlag, Berlin, 1985.
W. N. Sproson, Colour Science in Television and Display Systems, Adam Hilger Ltd., Bristol, 1983

C-3
Part 2: Color CRT Display Measurement Methods National Information Disvlav Laboratory

This page intentionally left blank

C-4