A Comparison of staggered position one angle process color printing with four angle and one angle process color printing

(1)

Theses

Thesis/Dissertation Collections

10-1-1988

A Comparison of staggered position one angle

process color printing with four angle and one

angle process color printing

Blair Richards Jr

Follow this and additional works at:

http://scholarworks.rit.edu/theses

This Thesis is brought to you for free and open access by the Thesis/Dissertation Collections at RIT Scholar Works. It has been accepted for inclusion

in Theses by an authorized administrator of RIT Scholar Works. For more information, please contact

[email protected].

Recommended Citation

(2)

Rochester, New York

CERTIFICATE OF APPROVAL

MASTER'S THESIS

This is to certify that the Master's Thesis of

BLAIR RICHARDS, JR.

with a major in Printing Technology has been approved by the Thesis Committee

as satisfactory for the thesis requirement for the Master of Science degree at the convocation of

OCTOBER 1988

Thesis Committee;

Miles SOllthworth

(3)

FOUR ANGLE AND ONE ANGLE PROCESS COLOR PRINTING

I, Blair Richards,Jr., hereby grant permission to the Wallace Memorial Library of the

Rochester Institute of Technology, to reproduce my thesis in whole or in part.

Any

reproduction will not be for commercial use or profit.

/1

!Jec

I

ft

(4)

FOUR ANGLE AND ONE ANGLE PROCESS COLOR PRINTING

by

BLAIR

RICHARDS,

JR.

A thesissubmitted in fulfillment oftherequirementsfor the degree of Masterof Science in the School of

Printing

Management and Sciences in the College of Graphic Arts and

Photography

ofthe RochesterInstituteof Technology.

OCTOBER 1988

(5)

A

Master's Thesis

requires the singlededication of its author in allfacets of its development. It

demands

a

immense

amount of

energy

and thededicationof

many

lonely

hours. A thesisalso

provides opportunities for

experiencing

new ideas and for interaction with a wide _variety of

people. Anumberofsuch people are responsiblefortheworkand completion ofthis thesis.

My

advisors Miles Southworth andJames Rich have been constant sources of ideas

and feedback. I amthankfulfortheirsupportwhich has beenunrelenting.

My

appreciation also

extends to the

following

_people; Jennifer _{Wills for the photography} of the original

transparency;

William Birkett Jr. of Precision Color forthe use ofhis Hell 299L

Scanner;

Bekir

Arpag,

Steve

Terrillion,

and Mark Hoffman for their advice and

help

with the image

assembly; Bob Hacker for the newsprint; Jack Jenkins and his press crew of Ron

Stirpe,

Bob

Mancini,

Brian

Ambroy,

Tim

Meegan,

and Todd Socia for the press run; Paul

Wilson and Dan Smialek fortheir

consulting

inthe experimentaldesignand statistical _analysis;

Franc Grumm forthe useof theMunsell

Lab;

Stan Rosen for hisviews on dot-on-dotprinting;

Frank Cost for his computer_wisdom;

Sherry

Addington and Glenn

Vogelsung

for a place

to

sleep

in

Detroit;

Melissa

Berry

for her constant encouragement; Joe Noga for

being

an

understanding supervisor; Michael Linsner of the

University

of Rochesterfor his

long

hours

analyzing and_computingthe

data;

Joe Pasquarelli ofthe

University

of Rochesterfor hisword

processing advice and expertise; Ed Granger for his ideas forthe

Conclusion;

the tenjudges:

Joe

Noga,

Rob

Aronson,

Jill

Houghton,

Phil

Mackowiak,

Xenon

Elyjiw,

Scott

Peace,

Sven

Ahrenkilde,

Dave

Cohn,

Dave

Tontarski,

and Bob

Chung

for ajob well

done;

and a_veryspecialthanks to DavePeschfor hisproof _readingandthemaintainenceof _my

sense of humourthroughoutthis whole experience. There is a part of all ofyou inthe

following

(6)

Page

LIST OF TABLES vi

LIST OF FIGURES vii

ABSTRACT ix

CHAPTER

1. INTRODUCTION 1

The Purpose 1

The Background 3

Primary

and

Secondary

Moires 3

Dot-on-Dot

Printing

and Color Shift 5

Footnotes for Chapter 1 9

2. THE TECHNIQUE 10

The Staggered Position One Angle Method 10

3. LITERATURE REVIEW 14

4. HYPOTHESES 17

(7)

Image

Assembly

21

Press Run 22

Sampling

22

Subjective

Judgements 22

Objective Measurements 23

Statistical Analysis 23

6. EQUIPMENT AND FACILITIES 27

7. RESULTS AND ANALYSIS 28

Rosette Patterns 28

Moire Patterns 31

Tone Reproduction 33

Sharpness 37

Color Shift 39

Spectrophotometer Measurements 41

Overall Acceptance 42

8. CONCLUSIONS 44

9 DISCUSSION 46

(8)

BIBLIOGRAPHY 53

APPENDIX A 55

Testing

Instructions

Scoring

Sheet

Spectrophotometer

Measurements-Flesh Patch

Spectrophotometer

Measurements-Gray

Patch

DensitiesofOriginal

Gray

Scale

MeansofPrinted

Density

Measurements-65 Lpi.

Meansof Printed

Density

Means of Printed

Density

APPENDIX B 56

Staggered Position

Printing

Samples

Four Angle

Printing

Samples

(9)

Table

No. _Page

1 _{ROSETTE PATTERN: Analysis} of Variance 29

2 MOIRE PATTERN: Analysisof Variance 32

3 TONE

REPRODUCTION:

AnalysisofVariance 36

4 SHARPNESS: Analysis of Variance 38

5 COLOR SHIFT: Analysis of Variance 40

6 DELTA E ANDJUDGES'

SCORES CORRELATION 41

7 OVERALL ACCEPTANCE: Analysisof Variance 43

(10)

Figure

No Page

1

Primary

Moire Pattern 4

2 Dot

Centered

Rosette 4

3 Clear

Centered

Rosette 4

4 Enlarged45

Halftone Screen with 50% Dots 10

5

Staggered

Position Dot Center Configuration 1 1

6

Stripping

Imposition 21

7 Rosette Pattern: Meansof Judges'Yes Responses 28

8 Moire Patterns: MeansofJudges'Yes Responses 31

9 65 Lpi Tone Reproduction 33

10 85 Lpi Tone Reproduction 34

(11)

11 100 Lpi

-Tone

Reproduction

34

12 All Lpi-Tone Reproduction 35

13

Sharpness:

Means of Judges' Scores 37

14 Color Shift: Means of Judges' Scores 39

15 Overall Acceptance: Meansof Judges' Scores .42

(12)

The

Staggered Position One

Angle method of

preparing

colorseparationsdoes not

adjustthe dot angle foreach color. The dot pattern isstaggered

horizontally

or_vertically

by

halfarow ofdots foreach separation

maintaining

thesameangle.

The investigation compared the

Staggered

Position method to the four angle

method andto the

dot-on-dot

method. Newsprint was_{the substrate,} and

65,

85,

and 100

lines per inchwerethe rulings. Experienced judgesviewed the representative prints and

scored each_{for sharpness,}colorvariability, andoverall acceptance.

They

werealso asked

to

indicate

the evidence of rosette and moire patterns. Densitometric and

spectrophotometer readings wereusedforobjectiveanalysis.

The judges found the

Staggered

Position to beequal orbetterthan both fourangle

and_{dot-on-dot printing in} mostinstances. Anon existant rosettepattern and aminimal risk

of

moire'

patterns also resulted. The _{Staggered Position printing} excelled in the

ability

to

produce a

sharp

reproduction, and itstone reproductionwas foundto be equivalentto the

fourangle anddot-on-dotprinting. Theuse ofDelta E forthevisual rejection oracceptance

of animagewas foundnottobe a goodindicatorin thisparticularstudy.

The

study

opens_up anumber ofareas for further research intothis newseparation

method, but becauseof itsgood performance on newsprint, itshouldbe considered as an

alternativeto the fourangle methodfornewspapers.

(13)

THE PURPOSE

The purposeoftheresearchwasto evaluate anew colorseparation _screening method

called the

Staggered

Position One Angle technique. To accomplish

this,

color printing

producedfrom separationhalftone negatives prepared

by

this newtechniquewerecompared

to color _printing produced from separation halftone negatives prepared

by

the four angle

method and separation halftone negatives prepared

by

the singleangle method.

The Staggered Position One Angle

technique,

from here on referred to as the

Staggered

technique,

is a _screening process developed to overcome the deficiencies that

existwith boththe fourangle andthe single angle method. The Staggeredtechnique uses

thesame angleforallfourseparations. Rather than_anglingthescreento repositionthe

dots,

aswiththefourangle

technique,

thescreenismoved

laterally

so astostaggerthedotpattern

foreachcolor. Thespacial

integrity

_usuallyaccomplished

by

_anglingthedots is achievedwith

the Staggered technique

by

_movingthe same angle screen

horizontally

or_vertically 1/2dot

diameter distance.

This

new conceptwasdeveloped asa result of researchdone

by

James Rich into one

angledot-on-dotprinting1

. One angledot-on-dot printing uses colorseparations ofthesame angle sothat the dotsof each color print on

top

of each other. Itwas foundthat dot-on-dot

printing eliminatedthe rosette patterns and produced asharper

image,

but italso confirmed

the old problem of color shifts due to misregister. The Staggered

technique,

which will be

described in detail

later,

modifiestheone angle method sothatits advantages are maintained

(14)

of rosettes and moires.

They

were also askedtoscorethe acceptabilityof images basedon

(15)

Fourcolor

lithographic printing

requiresthat_{four printing} plates be producedfrom four

halftone

_negatives, each one a photographicrecord of _{the yellow,} magenta, cyan and black

contained in anoriginal. Afterallcolorshave beenprinted onthe substrate, thecollection of

printed halftone values producethe illusion of continuous tone color. The different shades

are a result of_{the varying dot}sizes printedonthewhite paper.

The dots forthesefourseparationscan be produced

by

halftone screens or generated

by

adot

generating

color scanner. Whichever methodisused, thedotsoccurinparallel and

perpendicular rows and at a specific

frequency

forallfour halftones.

The four

halftones,

withthis

frequency

of _equally spaced rows of _{dots running in}the

cross-line_pattern,create asituation where avisible moire mightoccur. Avisiblemoireoccurs

whenever two regular patterns which differ in

frequency

or angle, or

both,

are

superimposed.2

As a_{result, the}_printing

industry

hasused angled screensof

90,

75,

45,

and

15degreestoavoid a noticeable moirewhen_{printing four}colors.

PRIMARY AND SECONDARYMOIRES

John

Yule,

in his book PrinciplesofColorReproduction.3_{classifies moires} _{into primary}

and _secondary types. _{A primary} moire isan objectionable lightanddarkareathatappears

whentwo screens ofthesame angle are not exactly in registerto each otheras in Figure 1.

The most objectionable pattern appears as the screens approachtheir perfect registration

(16)

A

secondary

_moire,

commonly

calledtherosette, occurswhentwo tofour screens are

at30

degrees

toeach other. It is the leastobjectionable atthis30degreemeasurebut isstill

noticeable,

especially

inneutral areas where_yellow, magenta andcyanprint40 to50percent

each. The rosette is most objectionable at 45

degrees,

which _{unfortunately is} the same

degree measurementthat the_primarymoireistheleastoffensive. Thetermrosettedescribes

a_varietyof rosette configurations. A rosettecanbe dotcentered (Figure

2)

or clearcentered

(Figure

3)

or all the variations in between.

They

are the result of the screen angles and

frequency,

thevariabledot_sizes, and misregisteroftheprinting.5

o

<

2

O

n

Vo

.o

O

o

.

<J

o

O

oO'Q

o

Figure 2

Dot Centered Rosette6

^

CO*3

.

O

Figure3

Clear

Centered

Rosette7

c?

o

(17)

fourscreens are neededfor the fourcolor process.

Fortunately,

yellowisoneof ourfourinks

and will show _{very little} _moire when placed

halfway

between two ofthe anglesor 15from

each. Yellow is almost an ideal

ink,

meaning

that _{it is very}closeto a perfectabsorber. The

significance ofthiswillbeexplained next.

Whena moire occursin multi color_printing,what_{is actually occurring is}acontrast oftwo

overlapping

inks,

having

a common _{absorption, occurring in} theregular pattern of a halftone

screen. The inks used in

printing

are not perfect absorbers. This meansthat each ink does

not absorb all of the light in that region of the spectrum it is

intended,

nor_{does it perfectly}

transmit the_remaining regionsofthespectrum. Alloftheinks _{have overlapping absorbency}

regions andwhen printed over each other

they

produce light and darkpatterned areasofa

particularfrequency.8 Becauseofitscloseto ideal_absorbencyand

transmittance,

yellowink

has the least occurrenceof _{overlapping abosrbency}regions and canbe printed at an angle

lessthan the 30degreesrequiredfor theotherinks.

Even after allthescreening has been carefully

done,

thereare otherfactorsthatcould

promote a moirecondition-poor

trapping,

non-absorbentstock, regularpatternsinthe stock,

and regular patterns in the imaged Conditions of poor

trapping,

aswith coated paper and

glossinkswith wet onwetprinting, encouragedotspreadwhichwillviolatetheangle

integrity

and produce moire.10 Commercial

items,

suchasfabricand shade materialthathavepatterns

themselves,

willcause interferencewiththescreenanglesand_mayresult inmoires.

DOT-ON-DOTPRINTING AND COLOR SHIFT

The alternativeto _printingfourcolorsatfour differentangles has beento print themat

the same angle. This method has many names: one angle, single _angle,

dot-on-dot,

one

(18)

to avoid patterns would be to print all the halftones at

exactly

the

same angle.

This

would not be

satisfactory

because the color produced

would

depend

on whetherthe dots fell on

top

of each other or side

by

side. With dots on

top

of each _other, a much lighter tone would be

produced, because therewouldbe amuch largerareaof unprinted white

paper.11

However lightthe color, itwould_{be very}sensitivetovariations in register. Misregisteroccurs

in all four color

printing

and causes even a slight color shift with four angles. The color

variation

occurring

withdot-on-dot is foundwithinthe imageaswellasfromsheettosheet and

is often quite noticeable. The shift is most noticeable with dot-on-dot because ofthe dot

placement andthelargeareaof unprinted white paperthatisexposed. Themisregistration of

one _color, and the _{resulting dots printing} off of their dot-on-dot position and onto the

unprintedwhite paper, createsan exaggereatedcolor change. The dot-on-dotarea has lost

someof itsabsorption, andtheunprinted paper whichbeforereflected all bandsnow absorbs

some bandsand reflects fewerthan it did before. Thecolor shiftthat existswithfourangle

printing is not noticeable becausetheamountof unprinted white paper_{is considerably less}as

a resultofits dotconfiguration.

James

Rich,

in his1982 "Comparison of FourColor

Printing

atOne Angle and at Four

Angles"12,

confirmedtheexistence ofthecolorshift, but foundthevariation negligible inthe

realmof acceptable commercial printing. His studycomparedthe two techniquesprintedwet

on weton 50 lb.coated stock_usingsixdifferentscreen rulingsfrom 100to200linesperinch.

Allofthe one angle printing, exceptforthe 120 lpiwas rejected forpoortone reproduction.

Richblamedregistration andthe_resultingcolorshiftfor this rejection.13

Idealtone reproduction isthe_{ability to} reproducethetones of an original as

closely

as

possible, given the limitations of the printing process. _{It is usually determined}

by

the

reproductionof a _{grey step}tablet included inthe colorseparations and its approximation to

(19)

Rich's14

sharpness analysiscomparedone angle_printing ofa specificscreen ruling, to

four angle

printing

of the next finest screen ruling. For example; one angle 100 lpi was

comparedtofourangle 120 lpi.

One

anglewas_unanimously consideredsharperin the 100

and 120 lpi rulings.

One

angle screen rulingsof

133, 150,

and 175 were considered sharper

80, 60,

and 70 percentofthe time respectively. Sharpnessor resolution refers to the ability

to reproducefine detailandclearcut images.

There was no noticeable rosette pattern in the one angle printing. A rosette pattern

was noticed in thefourangle _printing in screen rulingsof 175 lpiand coarser.15 _The eye is

unabletoresolve rosette patterns inthe 200 lpiscreen ruling.

Uncoated paper, especially newsprint, seems to work better with dot-on-dot color

printing

when_usingthecoarser screen rulings. TestsoftheChemco Company's

"Easy

Color

System"16 _show _that _dot-on-dot

color on newsprint produces superior color, gives the

impression of afiner line screen, eliminates the rosette and moire patterns, and produces

consistentfleshtones. Even deliberatemisregistrationdid not produce a moire pattern of

any

kind. _{When printing}one angledot-on-doton newsprint with acoarse screen ruling, thecolor

variation dueto misregisterappearsto be minimal. Thiscan be explained

by

the newsprint

allowing the dotsto spread,whichcausesthemto be less sharp.17

As stated in the

Introduction,

the Staggered technique evolved from research done

with dot-on-dot printing. This researcher, who was very familiarwith the conventional four

angle

technique,

was _seeking an easiermethodto screen separations. Those experiments

confirmed the problems that have existed with dot-on-dot for a

long

time. The

Staggered

technique is an attempt to overcome the problem of the color shift and maintain the

improvements over the four angle method - the _removal _of rosettes and

increased

(20)

printing is

conventionaly

screened color_{printing, it}must becomparable in aspects oftone reproduction

and _sharpness, withthe added advantagesof rosetteelimination and an unnoticeable color

shift. If the purported advantages can be met, and the technique can be performed with

reasonable ease and _accuracy,the Staggeredtechnique _{may be} considered a replacement

(21)

2F.L.

Wurzburg,

"Understanding

Moire Screen

Patterns,

Gravure Magazine. September

1961,

p. 17.

3J.A.C.

Yule,

Principles ofColor Reproduction. (New York: John

Wiley

and

Son, Inc., 1967)

p.330

-344.

4lbid.,

p.331.

5lbid.,

p.339.

6lbid.,

p. 340.

7lbid.

8F.

Pollack,

"The Dependence of the Contrast of Moire Patterns on the Colours of the

Printing

Inks",

TAGA Proceedings. (Los

Angeles, 1958)

p.44.

^Wursburg,

p. 20.

10P.

McKinney,

"Screen Patterns in

Printing",

Photoplatemakers Bulletin. No. 11

(1958)

p. 57.

11

Yule,

p. 335.

12James

S.

Rich,

"A Comparison of Four Color

Printing

atOne Angle and at Four

Angles",

TAGA Proceedings.

(Toronto,

1982). p. 361 - 377.

13lbid.,

p. 368.

14lbid.,

p. 367.

15lbid.

16"Color

Separation Uses One Screen/One

Angle",

Editor and Publisher. June

7,

1980,

p. 46.

(22)

CHAPTER

2

THE TECHNIQUE

The Staggered Position One Angle Method

The

Staggered

technique screens color separation _{halftones using} the same angle.

The screen is repositioned

horizontally

or _vertically

(staggered)

to a new position for each

separation. The dotpositionischanged, the image remainsstationary. The measure ofthe

stagger is one half the side of the 50 percent dot forthat screen ruling.1

Forthis thesis

experiment square dots were used. The screen _ruling of a screen refers to the numberof

linesperinchmeasured _alongthe screenangle(dot diagonal). Withthisin mindandreferring

to Figure

4,

it can be stated that the printed dot diagonal and the distance between the

screen dot centers are equal. This distance

(d)

is equal to one inchdivided

by

the screen

ruling, ord=1 "/linesperinch.

50

%

dof

Strain

Screen

Figure 4.

(23)

Again

referring

to

Figure

4,

andwiththeassumptionthat thefoursidesof a50% dotare

equal, the

Pythagorean

Theorem can be used to determine the measure of the sides(s) of

the

50%

dot:

S2 +

s2-

2s2-

S2-s

-s -d2

d2

d2/2

V(1"

/linespar

inch)

(1/2)

When

screening

_{the separations, 1/2} s is the distancethe screens are moved in the

direction parallel to the sides of the 50 percent dot.

The

screen position for the first

separation is the

starting

point. Thescreenforthe second separation ismoved 1/2 sto the

right parallelto the horizontalsidesofthe50 percentdot. The screenfor the thirdseparation

is moved 1/2 s

down,

parallel to thevertical sides ofthe50 percent

dot,

or_{from the starting}

point 1/2s

horizontal,

1/2svertical. Thescreenforthe fourthseparation ismoved 1/2sto the

left from the third separation parallel to the horizontal sides or 1/2 s down fromthe

starting

point paralleltothevertical sides.

1/2s

YELLOW

O

060000

^

000000

MAGENTA

0

000000

CYAN

0

0 0

000000

>

0*00000

000000

Figure 5.

(24)

Staggering,

instead of _angling,

distributes

the dots in a uniform pattern of consistent

distances

which will reproduce

images

thataresharper withoutarosette. Like

dot-on-dot,

the

eliminationofthe rosette patternwillgive a sharperimage.2 _Unlike

dot-on-dot,

thedots are

not printed on

top

of each other but side

by

side. There will be less unprintedwhite paper

exposed,

making

a color shiftdue to misregister less exaggerated.

Ideally,

the color shift

shouldbeas unnoticeable asit iswithfourangleprinting.3

The above procedure has described the Staggered position technique in terms of

manually adjusting

thescreenforeachseparationhalftone. Thisauthor realizesthatbecause

ofthe predominance ofcolor scannersinthe

industry,

anewtechnique suchasthis must be

adaptable to the scanner. To add _credibilityto both the technique and this research, the

separationsthatwereprinted and evaluatedweredoneonacolorscanner. The detailsofthis

(25)

FOOTNOTES

FOR CHAPTER2

1 1nterviewwithJames S.

Rich, January,

1984.

2James

S.

Rich,

"A Comparison of FourColor

Printing

at One Angle and at Four

Angles",

TAGA Proceedings.

(Toronto,

1982). p. 368.

(26)

CHAPTER

3

LITERATURE REVIEW

The

majority

of the pertinentwork

done

in this area has been mentioned in previous

chapters of this thesis. The selection of alternative methodsto screening separations with four

differently

angled screens was limited. Brent Archer's 1963patent, "Halftone Screen for

Color Separation"1

, simplified the angled screen process

by

using the

"tri-screen"

This

screendid not needto be rotatedto anewangle foreach color. Itcontainedthree individual dye screens, each of whichwas angled

differently

and each was sensitiveto eithergreen,

red, orbluelight.

Only

thefilterandthefilmneededtobechanged.

The firstreferencetodot-on-dot printingwasin 1950

by

W.P. Greenwood2_{in his} report

to TAGA. He examined color saturationforthe_{different printing}configurations ofdotbeside

dot,

dot-on-dot,

and dots at 30 degrees to each other. He reported that dot beside dot

exhibited _{very strong} saturated color, dot-on-dot exhibited brighter but desaturated color,

and dotsat30 degrees exhibited color

falling

between the two.

Further investigation into one angle color provided only references with no real

scientific investigation. John Yule3 acknowledged itsattributes butrealized the

impossibility

ofperfectregistrationand _avoidingthecolor shift. F. Pollack4_felt_that_{one angle color}

printing

waspossible only iftransparentinkswereused,andthepossibilityofamoire eliminated.

Only

Rich's

investigation,

"A Comparisonof Four Color

Printing

atOneangle and at Four

Angles"5,

investigatedan actualoneanglecolorpress run withtheresults analyzed andpublished.

In 1968 Kenneth Rarey6_patented _a

screening technique for Xerographicprinting. His

technique shifted hexagonal dot screens forthree colors to positions not printed on

by

the

previous two screen positions.

(27)

screen

specifically

designed forone anglecolorseparation. Because of itsunique _array of

concentric parallelogramswhichcomposetheindividual

dots,

moire patterns are not

formed,

rosettes are _eliminated, andthe manipulation of the single one angle screen simplifies the

color reproduction process.

The

1982 master'sthesis topicofJang-Fun

Chen

studied_{the predictability}ofthecolor

shift found with dot-on-dot printing.8

Investigating

the dot area levels of

25,

50,

and 75

percent and

misregistering

one color at atime

by

1/4,

1/2and3/4of a

dot,

he determinedthe

extent thata color could bemisregistered beforethecolor shift becameobjectionable. Chen

also developed a mathematical model which was employedto indicatethe direction of the

color shift in termsof

hue,

saturation and

lightness,

aswellas_predictingtheshift. Twoyears

later,

Christopher Vasko in his thesis "A

Study

of

Chromaticity

Changewith Respectto

Dot-on-Dot Multicolor Halftone. Printing"9 _took _Chen's

work a _{step forward} and analyzed

chromaticity changes due to misregister in the three additive primaries of actual printed

sheets.

Ted Chen's 1987 master's thesis compared the image _quality of four angle

reproductions anddot-on-dotreproductionswith differentvariables of input image qualityand

screen

frequency,

withaquantificationoftheirresults.10

The Staggered technique is thefirst attempt to purposely changethe dot position of

same anglecolor separationstoeliminatethecolorvariation. If thecolorvariationcanbemade

unobjectionablewiththeadded advantages ofasharperimage andtheremoval of rosettes, a

techniquewill existto significantly improve newspaper color. The results ofthe experiment

andtheevaluations ofthe results, will indicatetherealityof suchatechnique. The Staggered

technique _may perhapsbe an alternativeto the fourangle method in commercial multi-color

(28)

FOOTNOTES FOR CHAPTER 3

1H.B.

Archer "Halftone Screen for

Color

Separation",

United

States

Patent No.

3, 085,878,

April

16,

1963.

2W.P.

Greenwood,

"The Colorof Halftone Dot

Patterns",

TAGA Proceedings.

(Rochester,

1950). p. 62.

3J.A.C.

Yule,

Principles of Color Reproduction (New York: John

Wiley

and

Son,

Inc.,

1967).

p. 335.

4F

Pollack,

"The Dependence of the

Contrast

of Moire Patterns on the Colours of the

Printing

Inks",

TAGA Proceedings. (Los

Angeles,

1958)

p.37 -44.

5James

S.

Rich,

"AComparison of Four Color

Printing

Angles",

TAGA Proceedings.

(Toronto,

1982). p.361-377.

6K.W.

Rarey,

"Methods and Apparatus for

Minimizing

Screen Patterns in

Xerography,

Electrostatic Screen Process and Other Forms of

Printing",

United States Patent No.

3,363,552,

January

16,

1968.

7H.K.

Liu,

"HalftoneScreenwith Cell

Matrix",

United States PatentNo.

4, 188,225,

February

12,

1980.

8J.F.

Chen,

"An Investigation of Color Variation as a Function of Register in Dot-On-Dot Multicolor Halftone

Printing",

(Master's

Thesis,

RochesterInstitute of

Technology,

1982).

9C.M.

Vasko,

"A

Study

of

Chromaticity

Change with Respect to Dot-on-Dot Multicolor

Halftone

Printing",

(Master's

Thesis,

Rochester Instituteof

Technology,

1984).

10T.C.

Chen,

"A

Study

ofColor Image

Quality

withRespecttoScreen Angle Arrangements in

(29)

CHAPTER

4

HYPOTHESES

It is the objective of this research to compare

Staggered

_{Position printing} with four

angle and dot-on-dot printing.

Thirty

prints of each type were viewed

by

ten experienced

judges. Each judge answered questions about_{the visibility}ofrosettes and _moires, and was

askedto indicate the

acceptability

of the prints in areas of _sharpness, colorvariability, and

overall color reproduction quality.

Tone reproduction and spectrophotometer readings were made in addition to the

previously

mentioned subjective evaluations

The results of the above evaluations and measurements helped determine the

statistical significance ofthe three methods forqualities of overall acceptability, sharpness,

tone reproduction, color shifts, rosettes and moires. It is hypothesized that there is no

significant difference between Staggered position _printing and eitherfour angle _printing or

dot-on-dot printing inthe aboveareas.

Specifically

the hypotheses read:

There is no significant difference intheexhibition of a colorshift between Staggered

PositionOne Angle printingand Four Angleprinting.

There is no significant difference intheexhibition of a color shift between Staggered

PositionOne Angle printing and Single Angle Dot-on-Dotprinting.

There is no significant visual difference in the overall acceptance between

(30)

There

is no significant visual difference in the overall acceptance between

Staggered

Position

One

Angle

printing

and

One

Angle

Dot-on-Dot

printing.

There is no significant difference in tone reproduction between Staggered Position

One Angle

printing

and Four Angle printing.

There is no significant difference in tone reproduction between Staggered Position

One Angle

printing

and One Angle Dot-on-Dot printing.

There is no significant difference in sharpness between Staggered Position One

Angle printing and Four Angleprinting.

There is no significant difference in sharpness between Staggered Position One

Angle printing and One Angle Dot-on-Dot printing.

Staggered Position One Angle printing doesnot exhibit a moire pattern.

Staggered Position One Angle printing does not exhibit a rosette pattern

There isacorrelation between theacceptance ofa print's pressrun color shift andthe

(31)

CHAPTER

5

METHODOLOGY

The

prevalence offourangleprinting, in mostif not all multi-color_printing applications,

confirms its

eligibility

forcomparison _{in any}

printing

situation. The previousworkdone inthe

area of dot-on-dot

printing

has

determined

when _printing at the same angle does well and

when it performs poorly. Thespecifications_{for my} experimental comparisonwillbe basedon

these previous results.

Chemco's success on newsprint, and Rich's rejection rate on coated stock made a

newsprint substrate a logical choice. Screen rulingsof

65, 85,

and 100 lines perinch were

chosen because oftheir use in the newspaper

industry

and the dot-on dot's success in the

Chemconewspaper system.

The

following

steps weretaken to produce the printing, to evaluatethe results, and to

test the hypotheses inthis research.

Test Image: A

transparency

was photographed and used asthe original _copyfor all

separations made in this experiment.

Along

with a _variety of colored objects there were

objects contained in the

transparency

to aid the evaluators of the printing.

Memory

color

itemswere included to aidtheviewer in

detecting

acolor shift. These included itemsthat a

viewerwould recognize and have some idea ofwhat color

they

should

be,

such as

logos,

fleshtones,

fruitsandvegetables. Alsotoaidthe evaluators, aselection of neutral_{grey items}

and a neutral_greybackgroundwere included. Colorshiftsandrosettes occur_{readily in}these

areas.

Objects,

which themselves contain repeating patterns, were included to increase the

probabilityof a noticeable moire and

help

determine how each separation method handled

(32)

The

transparency

was photographed in a studio with strobe

lighting

and abalance of

lowandhigh

key

objects on4x5 Kodak

Ektachrome Daylight

film.

Color

Separations:

The

transparency

was mounted ontothecolor_{scanning drum}of

the Hell 299L

direct

screen scanner at Precision Color

Plate

of

Plymouth,

Michigan. Register

marks were

included

on either side ofthe

transparency

asthe

top

and bottomwastaped. A

gray

scale was also included. For each screen _ruling a patch of50 percent dots produced

fromthatscreen wastaped

securely

toone sideatthe

top

ofthe transparency. Thiswas used

for

positioning

the

Staggered

separations.

The four angle separations were made _using a setoffourscreensof

15, 45,

75 and 0

degrees for

black,

magenta, cyan and yellow respectively. The dot-on-dot and Staggered

separations were madewiththe 45degree screen. _{When making}thedot-on-dotseparations

the scanner operator_carefully returned the _{scanning head}and _{exposing head}to the same

positionfor thestartofeach color. Thisassuredthedot-on-dotconfiguration.

The Staggered positionseparations required as much if not more attention to detail as

thedot-on-dotsets. _{The starting}pointforacrossthe _{scanning drum}wasthesame asforthe

dot-on-dot separations. The circumferential _starting point, orthe around the drum _starting

position was set atthe

top

ofa_specificallymarked50percentdotthatwas _previouslytaped to

the

top

of the transparency. The cross hairs of the viewer and the magnifier aided the

operatorin this procedure. _{The exposing head} stayed atthe same _starting pointthat was

used forthe one angle separations. Theyellow separation was scanned forthese points.

The magentaseparationwas scannedasthe

first,

exceptthecircumferential_startingpointwas

moved to the middle of the designated 50 percent dot. This output the image on the

exposingdrum halfofadot fromtheyellow separation. Thecyan separation wasscanned as

the second, except the exposing head was moved half a dot to the right. This output the

cyan separationhalfadot fromthe magentaseparation. Thissmallmovementofthe

exposing

(33)

head

when itwasbrought backtoits

starting

point. Thisspacer was a pieceof craft orpress

blanket packing

paperwhosethicknesswas equivalentto one halfof adot forthatparticular

screen ruling. The

starting

block was then moved _{up to} contact _{the exposing head} and secured.

The

blackseparation was scanned as_{the cyan,} exceptthecircumferential_starting

point on the

scanning

drumwas moved back to the

top

of the designated 50 percent dot. Thisoutputtheblackseparation halfadot from thecyan separation.

This

procedure was performed for each different screen ruling. Separations were

negative, right

reading

on the base. All separations were scanned to newsprint specifications:

C M Y K

Highlight 4 2 2 0

Midtone 50 38 38 15

Shadow 97 90 90 80

The H-M range was .90. A UCR value of 300was usedalongwith a reduction of unwanted

colors and an increase of wanted colors in the overall color correction settings.3 _{All detail}

enhancement and_{grey balance knobs}wereat normal settings. Matchprintproofswere made

foreachset of separations.

Image Assembly: The author assembledthe films sothat thedifferent screen rulings were acrossthe web,andfourangle,staggeredand dot-on-dotwere parallelto theweb:

65.

85 100

/

V

4angle 4angle 4angle

stag. stag. stag. WEB

d/d d/d d/d DIRECTION

(34)

The

authorassembledtheseparations

according

to theaboveimposition. Theregistermarks

and the images themselves were usedto strip-in-registerthe four angle separations. The

assembly

ofthedot-on-dot images required not_onlythe register marksin alignmentbut also

thedots. Theregister marks were used

initially

toalignthe

Staggered

position stepsbutthe

established "stagger"configurationof each

step

determinedtheirfinal position. A common

setofregister marks wereincludedoneachofthefour flats.

Press Run: Thepressusedforthis _studywastheGoss

Community

NewspaperPress

in the Newspaper

Laboratory

intheSchoolof

Printing

at RIT. Thepress crew was instructed

to treat theexperiment asan _ordinaryfourcolorjob. Registrationwasdonewiththecommon

register marks and color was matchedto the Matchprint proofs as _closely as possible after

reaching

the designated solid ink densities. After registration and color were OK'd the

counter wasactivated. Therun was_{approximately 3,000} impressions.

Sampling: From the run of

3,000,

one sample was pulled _every 300 impressions.

This provided 10 sets ofimagesper screen ruling. Within each screen ruling, the individual

imageswere separatedfromeach otherandmarkedforidentificationonthe back. Thisgave

thejudges 30 images per screenrulingto evaluate.

Any

morethan this would haveresulted

in the judges

becoming

fatigued.

Subjective Judgements: For the subjective evaluations, ten experienced judges

wereselectedfromRIT's SchoolofPrinting's

Faculty

and

Printing

Technology

Graduateclass

as well as from RIT's

Technology

and Education Center. Their

familiarity

with color_shifts,

rosettes, moires, andcolorprintingwas essential. The imageswerejudged undera standard

light source of 5,000 degrees Kelvin. This requirement standardizedtheviewing conditions

forallthe judges.

The 10 judges viewed and scored the imagesone set attime. Aset consisted of the

fourangle, the

Staggered,

andthe dot-on-dot print ofthesame screen

ruling

fromthe same

(35)

An

introduction

to each set of

judgements

was readto eachjudge with adefinition of

the

quality

heorshe was

judging

displayed

during

thesession. Asample ofthis is found in

the

Appendix.

Theirscoresfor

acceptability

or non

acceptability

oftheprintsforcolorshift,sharpness,

and overall acceptancecould rangefrom 1 to 10. Thescore of 1

being

unacceptable andthe

scoreof

10-being

veryacceptable. Their judgementsformoire and rosette patterns required

either a yes or no response.

Objective Measurements:

The objective evaluations consisted of densitometric

readingsofthe_{26 step grey}scalesincludedwitheach print. Theseassistedinthe evaluation

oftone reproduction capabilities.

Spectrophotometer

readings were done in the grey and flesh patches found in the

Macbeth

Chart

printed in each image. These assisted in the objective evaluation of color

consistency

of each method.

Statistical Analysis: Because the press run produced 3,000 impressions with nine

images per

impression,

theanalysisof alltheimageswouldbe impractical. The bestapproach

toa population of this sizeistosystematicallyselect samplesfromthe run and usetheseas

the representatives of the entire run. Even with the 90 representative _{samples, the}

enormous _quantityofdata generatedfromthe ten

judges,

the 26 step grey scales and the

two color patches make an analysis of an average for each method the most practical

approach.

By

incorporating

theAnalysis ofVariance4 itcan be saidthat iftwo populations have

identical averages the averages of their selected representatives will also be identical.

Therefore,

any significant difference in the entire population will be mirrored

by

these

representativesamples.

It isquite normalforpopulations to have variabilitywithinthemselves. Itwould bethe

(36)

the analysis of variance this random error and the

differences

that might be a result of an

introduced

variable are computed together and expressed in a f-ratio. _{The similarity} of

populations are questioned when this

f-ratio

exceeds an established "critical value" The

critical valuetakesintoconsiderationthenumber ofcomparisons, the

quantity

of

data,

andthe

level of the measure.

This

level of measure is expressed as alpha and most scientific

investigation as with this

study

is done with an alpha =

.05. Therefore the f-ratio and its

comparisonto the critical valuewilleither acceptorrejectthehypothesesthat twopopulations

are similar at aprescribedlevelof alpha.

Twoanalyses of variance were performedonthejudgesscoresforoverall_acceptance,

color shift and sharpness. The

Staggered

Position prints scoreswere compared to thefour

angle prints scores and

they

werecomparedagaintothescoresofthedot-on-dotprints. The

analysis of variance expressed the significant

difference,

and the means determined which

population scored higherorlower. Thiswasdone forscores at

65, 85,

and100 linesperinch.

Three Analysis of Variance were performedto the scores for rosettes and moires. Since

theseobservations requireda yesorno _{response, their}resultswere comparedto a normal

set ofdata

-aset of responses which would occur naturally. This set consisted of50% yes

responses and 50% no responses. The analysis consisted of this normal set compared

separatelyto theStaggered

Position,

fourangle, anddot-on-dotprint scores. Theanalysisof

variance determined the significant difference between the two and the mean score

determinedwhetherthesignificantdifferencemeant yes or no. Thiswasdone for

65, 85,

and

100 lines perinch.

The

density

readings for tone reproduction were also treated with an analysis of

variance. Staggered Position readings compared tothe fourangle readings and Staggered

Position readings compared to the dot-on-dot

density

readings. This determined the

similarity of tone reproduction and was done at

65, 85,

and 100 lines per inch.

(37)

image

weretaken. Foreach screen_ruling

65, 85,

and

100,

average

L,

a,andbwerederived.

This average becamethe standardfromwhichthe Delta Evalues were based. The Delta E

value is not an indicator of good or _{bad color, but} an indication of its difference from an

established standard. Previous studies of Delta E and subjective _{viewing have found} a

threshold

Delta

E value of six where color becomes objectionable. With this in mind a

Pearson's Product Moment Correlation5

was run between the Delta E values and the

correspondingaveragejudges' scoresforcolor shift. Thecloserthiscorrelationcomesto one

or negative one the strongerthecorrelation. The Pearson wasdone foreach methodfour

(38)

FOOTNOTES FOR CHAPTER 5

1_{James S.}

Rich,

"A Comparison of FourColor

Printing

Angles",

TAGA Proceedings.

(Toronto, 1982),

p.361-377.

2"Color

Separation Uses One Screen

One

Angle",

Editorand Publishes. 7June

1980,

p. 46.

3Miles

Southworth,

Advanced Color Separation Class

Notes,

Rochester Institute of

Technology,

Spring

Quarter,

1984.

4Richard

M.

Jaeger,

Statistics: A Spectator Sport

(Beverly

Hills: Sage

Publications, Inc.,

1984). p. 257-283.

5Gouri

K. Bhattacharyya and Richard A.

Johnson,

Statistical Concepts and Methods (New

(39)

CHAPTER 6

EQUIPMENT AND FACILITIES

HELL 299L Color Scanner-Precision Color Plate

Plymouth,

Michigan

Three ScannerScreen Sets:

65,

85,

and 100 linesperinch

Kodak Ultratec

Processing

Kodak Ultratec UHF Film

3M Matchprint

Proofing

System

3M Color

Key Proofing

Systems

Polychrome WGAN Offset Plates

Polychrome Plate Processor

Goss

Community

NewspaperWeb Press

FlintNewspaperInks

AbitibiPrice Newsprint

Hunter Spectrophotometer

(40)

CHAPTER 7

RESULTS

AND ANALYSIS

ROSETTE PATTERN

HYPOTHESIS: Staggered

_{Position One Angle printing does}notexhibit arosette pattern.

Accepted

for all screen rulings at the .05 alpha level.

ROSETTE PATTERN

co UJ CO z o Q. CO LU CC

CO LU

>-CC 111 GO s 3

Z < LU 5

B

FOUR ANGLE

1

STAGGERED

?

DOT ON DOT STANDARD

65 LPI 85 LPI 100 LPI

Figure 7

(41)

TABLE 1

ROSETTE

PATTERN

Analysis

of

Variance

Four Angleand

Standard

Staggered

Position

and

Standard

Dot-on-Dot

andStandard

65 lpi

Source

Method

Error

DF

SS

MS

1 12.50 12.50

198 25.00 .13

F

96.2*

DF SS MS

112.5012.50

19825.00 .13

F

96.2*

DF SS MS

1 7.61 7.61

198 34.79 .18

F

42.3

85 lpi

Source

Method

Error

DF

SS

MS

1 6.13 6.13

198 37.75 .19

F

32.3*

DF SS MS

112.50 12.50

19825.00 .13

F

96.2*

DF SS MS

1 12.01 12.01

198 25.99 .13

F

92.4

100 lpi

Source

Method

Error

DF SS MS

1 2.65 2.65

198 44.71 0.23 F

11.5*

DF SS MS

112.5012.50

198 25.00 .13

F

96.2*

DF SS MS

1 12.50 12.5C

198 25.0 .13

F

> 96.1

*

Significance

atthe .05alphalevel

The judges'

scoresforthe evidence of rosette patternswere compared to a standard

score set. The standard consisted of scoresthatwould occur without an

introduced

variable.

Asignificant statisticaldifferenceandthe meansdeterminedthestatus ofthehypothesis.

The appearance of rosettes is still adirect result of _angling the dots foreach process

color. As expected, rosetteswere not _{significantly}visible in either ofthe methods_using one

angle, although therewas a singlejudgewho saw rosettesinsomeofthe dot-on-dot images.

Therosette pattern inthefourangleprints_{is significantly}visibleandbecomes lessnoticeable

as the screen

frequency

increases. The previous studies on coated stock indicated a

disappearanceofthe rosette pattern atthe 175 lpiruling. Rosettes would

probably

become

lessnoticeable at acoarser_rulingwhenprintedonnewsprintthanon coated substrates. The

(42)

graynesswouldbethecause ofthis. Amoredirect andcontrolled_studyof rosette patterns on

(43)

MOIRE

PATTERNS

HYPOTHESIS:

Staggered

_{Position One Angle printing does} notexhibit amoire pattern.

Accepted

for all screen rulings at the .05 alpha level.

MOIRE PATTERNS

co LU CO z

o

0. CO LU CC

CO LU >

CO Ul

o

Q

O

CO z

<

LU

0

FOUR ANGLE

i

STAGGERED

?

DOTON DOT STANDARD

65 LPI 85 LPI 100 LPI

Figure 8

(44)

TABLE 2

PATTERNS

Analysis

ofVariance

Four

Angle

and

Standard

Staggered

Position

and

Standard

65 lpi

Source

Method

Error DF 1 198 SS MS 6.48 6.48 37.04 .19 F 34.1 * DF 1 198

SS

MS 8 8 34 .17 F 47.1 *

85 lpi

Source

Method Error DF 1 198

SS

MS 5.45 5.45 39.11 .20 F 27.3* DF 1 198

SS

MS 4.81 4.81 40.39 .20 F 24.1 *

100 lpi

Source

Method Error DF 1 198

SS

MS 7.22 7.22 35.56 .18 F 40.1 * DF 1 198

SS

MS 8.41 8.41 33.19 .17 F 49.5* Dot-on-Dot andStandard

DF SS MS F

1 5.78 5.78 30.4*

198 38.44 .19

DF SS MS F

1 8.82 8.82 55.1*

198 32.36 .16

DF SS MS F

1 7.22 7.22 40.1*

198 35.56 .18

*

Significance

at.05alphalevel.

Thejudges'

scoresfortheappearanceofa

moire'

pattern werecomparedtoa standard

score set. Thestandard consistedofscoresthatwould occur without an introducedvariable.

A significantstatisticaldifferenceandthemeansdeterminedthestatusofthehypothesis.

Many

of thejudges feltthat the appearance ofa rosette pattern was equivalentto the

appearance of a patternand scoredtheprintsaccordingly. Thiscould explainthehigh

scoresfor patterns inthefourangle printing.

Nevertheless,

theappearance of

moire'

patterns _{is significantly} minimal for all three methods in all three rulings. The chance of

encounteringa with eitherStaggered Positionordot-on-dotseparationsappearstobe

(45)

TONE REPRODUCTION

HYPOTHESIS:

There

is

no significant

difference in

tone reproduction between

Staggered

Position One Angle printing

and Four

Angle

printing.

There

is

no significant

difference

in tone reproduction between

Staggered

Position One Angle

_printing and

Single

Angle Dot-on-Dot

printing.

Both

hypotheses

accepted for all screen frequencies at the

.05 alpha

level.

1.40-

1.20-

1.00-

0.80-

0.60-

0.40-

0.20-

0.00-65

LPI TONE

REPRODUCTION

>

CO z Ul Q

Q Ul

FOUR ANGLE

STAGGERED

DOT ON DOT

H- Er

Z

0. r^fl M

1 ₁ 1 1 ₁ ' ₁ ' ₁ '

0.00 0.50 1.00 1.50 2.00

ORIGINAL GRAY SCALE DENSITY

2.50 3.00

Figure 9

(46)

85 LPI TONE

REPRODUCTION

CO z

Ul

Q

Ul

Z CC Q.

1.40

1.20

1.00

0.80

0.60-0.40

0.20

0.00 T I > ₁ ' _r

0.00 0.50 1 .00 1 .50 2.00

ORIGINAL GRAY SCALE DENSITY

Figure 10

85 LPI- Tone Reproduction

2.50 3.00

100 LPI TONE

REPRODUCTION

0.00 0.50 1.00 1.50 2.00

ORIGINALGRAY SCALE DENSITY

Figure 11

100LPI - Tone Reproduction

(47)

ALL LPI TONE

REPRODUCTION

1.40

0.00 0.50 1.00 1.50 2.00 2.50

ORIGINAL GRAY SCALE DENSITY

3.00

Figure 12

(48)

TABLE 3

TONE REPRODUCTION

Analysis

of

Variance

Four

Angle

and

Staggered

Position <

Dot-on-Dotand Staggered Position 65

Source

Method

Error DF

SS

1 .15 518

74.46

MS .15 .14 F 1.07

Source

Method Error DF 1 518 SS .11 74.81 MS .11 .14 F .79 85

Source

Method Error DF

SS

1 .10 518

73.40

MS .10 .14 F .71

Source

Method Error DF 1 518 SS .03 74.97 MS .03 .15 F .20 100

Source

Method Error DF

SS

1 .63 518 89.33 MS .63 .17 F 3.70

Source

Method Error DF 1 518 SS .37 97.39 MS .37 .19 F 1.95 All

Source

Method Error DF SS 1 .25 1558 239.59 MS .25 .15 F 1.62 Source Method Error DF 1 1558 SS .07 252.76 MS .07 .16 F .44 *

Significant

differenceat .05alphalevel

Tone reproductionis affected_{very little}

by

the_anglingorthe_positioningofthehalftone

dots. The

lightening

and colorvariation,whichoccured in other studiesof_{dot-on-dot printing}

and caused their rejection for tone reproduction, did not appear in either the

Staggered

Positionorthe dot-on-dotprinting. The 100 lpiprints did show a measurable

difference,

but

the difference _{is statistically} insignificant. A

darkening

ofthe tones occured intheStaggered

Position and _{dot-on-dot printing} atthis screen

frequency,

instead oftheir

being

a

lightening

(49)

SHARPNESS

HYPOTHESIS:

There is no significant

difference

in visual sharpness between Staggered Position

One Angle printing

Accepted

for the 85 lpi at the .05 alpha level.

There is no significant difference in visual sharpness between Staggered Position One Angle printing andSingle Angle Dot-on-Dot printing.

Accepted for 85 and 100 lpi at the .05 alpha level.

SHARPNESS

co LU CC

o o

CO

CO LU

O

Q 3

3

FOUR

1

STAGGERED

?

DOT-ON-DOT

65 LPI 85 LPI 100 LPI ALL

Figure 13

(50)

TABLE

4

SHARPNESS

Analysis

ofVariance

Four Angleand

Staggered

Position

65

Source

DF

SS

MS

F

Method

1 21.78 21.78 4.94* Error 188 828.60 4.41

85

Source

DF

SS

Method

1 2.64

Error 188 1028.23

100

Source

DF SS

Method 1 359.12

Error 188 640.86 3.41

MS F

2.64 .48

5.47

MS F

9.12 105.31*

ALL

Source

DF

Method 1

Error 588

SS

MS F

161.2 161.2 33.80*

2804.2 4.77

Dot-on-Dotand

Staggered

Position

SS

MS F

521.64 521.64 161.50*

607.23 3.23

SS MS F

15.68 15.68 2.36

250.14 6.65

SS MS F

.72 .72 .17

818.28 4.35

Source

DF

Method 1

Error 188

Source

DF

Method 1

Error 188 1

Source

DF

Method 1

Error 188

Source DF

SS

MS F

Method 1 224.47 224.47 37.47*

Error 588 3519.91 5.99

*

Significant

difference at.05 alphalevel.

When

Staggered

Position

printing

was

developed,

it was assumed that some

sharpness would be lostas a resultofthelateral movementofthedot. Theresultwas not as

anticipated. _{The Staggered Position printing} performed betterthanboth the fourangle and

the dot-on-dot printing.

The diminished rosettepatternwas notenoughto

help

thefourangle prints perform as

wellastheStaggered Position prints, andthedot-on dotprinting,which _usuallyprints_{sharp in}

(51)

COLOR

SHIFT

HYPOTHESIS: There

is no significant

difference

in the exhibition of acolor shift between

Staggered

Position

One Angle printing

Accepted for the 85 and 100 lpi rulings at the .05 alpha level.

There

is no significant difference in the exhibition of a colorshift between

Staggered

Position

One

_{Angle printing} and Single Angle Dot-on-Dot

printing.

Accepted

for the 85 and 100 lpi rulings at the .05 alpha level.

COLOR

SHIFT

0

FOUR ANGLE

I

STAGGERED

?

DOT-ON-DOT

65 LPI

Figure 14

(52)

TABLE

5

COLOR

SHIFT

Analysis

of

Variance

Four

Angle

and

Staggered

Position

65

Source

DF

SS

MS F

Method

1 _{222.60 222.60} _49.25 *

Error 188 850.27 4.52

85

Source

DF

SS

MS

Method 1 .02 .02

Error 188 919.98 4.89

100

Source

DF

SS

MS

Method 1 5.44 5.44

Error 188 910.95 4.85

ALL

Source

DF SS MS

Method 1 51.62 51.62

Error 588 2967.52 5.05

*

Significant

differenceatthe.05alphalevel

Dot-on-Dotand

Staggered

Position

Source

DF SS MS F

Method 1 38.72 38.72 8.42*

Error 188 864.86 4.60

F

Source

DF

SS

MS F

.00 Method 1 .02 .02 .14

Error 188 950.78 5.06

F

Source

DF SS MS F

1.12 Method 1 2.2 2.2 .44

Error 188 930.67 4.95

F

Source

DF SS SS F

10.22*

Method 1 15.68 15.68 3.27

Error 588 2813.84 4.79

Staggered Position printing was developed to minimize the color _variability which

existed in dot-on-dot printing. The results do not show a dramatic improvement

by

the

Staggered

Position over the dot-on-dot method, but

they

do show

Staggered

Position

printing's comparability to four angle printing. The exception to this is the 65 lpi screen

frequency. The Staggered Positionprints scored significantly lowerthan thefourangle and

dot-on-dot prints inthis ruling. Possible causesforthe_{significantly}poorperformanceof both

the single angle methods in this screen

frequency

could be a combination ofthe coarse

(53)

SPECTROPHOTOMETER

MEASUREMENTS

HYPOTHESIS:

There

is acorrelation betweenthe acceptanceof an image's colorshift and

the

Delta

Evaluefor that image.

Hypothesis

is rejected.

TABLE 6

DELTA E ANDJUDGES'

SCORES

CORRELATION

Grey

Patch Flesh Patch

Correlation

65 _{Four Angle} -.40 .60

Staggered

-.13 -.03

Dot-on-Dot

-.34 -.56

85 Four Angle -.41 -.34

Staggered

-.10 .03

Dot-on-Dot

.35 .76

100 Four Angle .10 -.50

Staggered

-.63 .42

Dot-on-Dot

.51 -.02

There is no correlation between Delta E and thejudges scores foreither ofthe color

patches. It was speculatedthat the derived Delta E values of the _gray and flesh patches

would provide an objective indication of the judges' perceptions of color variability. A

correlation between the Delta E values and the judges scores was anticipated but did not

materialize. _{Further study along} these lines might investigate other patches of color or

perhapsthe standard deviation of the Delta E measurements as bettermeasures of color

(54)

OVERALL

ACCEPTANCE

HYPOTHESIS: There is no significance

difference

in theoverallvisualacceptance between

Staggered

Position

One

Angle

printing

Accepted for the 85 lpi at the .05 alpha level.

There is no significant

difference

in the overall visual acceptance between

Staggered

Position One Angle

printing

and Single Angle Dot-on-Dot

printing.

Accepted for the 85 and 100 lpi rulings at the .05 alpha level.

OVERALL

ACCEPTANCE

co Ul

cc

o CJ

CO

CO Ul

o

a =>

0

FOUR

H

STAGGERED

?

DOT-ON-DOT

65 LPI 85 LPI 100 LPI ALL

Figure 15

(55)

TABLE

7

OVERALL ACCEPTANCE

Analysis

of

Variance

Four Angle and

Staggered

Position

65

Source

DF SS MS F

Method 1 30.42 30.42 6.10*

Error 188 937.50 4.99

Dot-on-Dot

and

Staggered

Position

Source

DF

SS

MS F

Method 1 176.72 176.72 37.20*

Error 188 892.06 4.75

85

Source

DF

SS

MS F

Source

DF SS MS F

Method

1 3.12 3.12 .64 Method 1 .84 .84 .15

Error 188 912.47 4.85 Error 188 1045.95 5.56

100

Source

DF

SS

MS F

Source

DF

SS

MS F

Method 1 36.98 36.98 8.78*

Method 1 9.24 9.24 1.87

Error 188 791.02 4.21 Error 188 930.15 4.95

ALL

Source

DF SS MS F Source DF SS MS F

Method 1 .47 .47 .093 Method 1 41.6 41.6

7.61*

*

Significant

differenceat.05alphalevel

The _{Staggered Position printing}was scored_{significantly}better than fourangle_printing

at100 lpiand itwas_{significantly}unacceptableat65 lpi.

The

Staggered

Position printingwas scored_{significantly better}thandot-on-dotat65 lpi

and was_{significantly}unacceptableat100 lpi.

If thejudges' scoresfor Sharpness and ColorShift were to be averagedtogether the

results would parallel the judges' scores for Overall Acceptability. _{Color variability} and

sharpness togetherare good indicatorsof how the judges evaluatedthe images foroverall

(56)

CHAPTER

8

CONCLUSIONS

Anumberofconclusionscanbe made about

Staggered

Position

printing

basedonthe

analysisof boththe subjective

judgements

and objective measurements. Theseconclusions

are applicable to the parameters used for this

investigation-65,

85,

and 100 lpi screen

frequencies on newsprint.

1.

Staggered

Position

printing

does not exhibit a moire pattern. Moire

patterns were not _{significantly} observed in either four angle or dot-on-dot

printing. The possibilities of _encountering moire patterns with one angle

printing andfourangle_printingareequal.

2.

Staggered

Position and dot-on-dot printing do not exhibit a rosette

pattern. Fourangle _printing exhibits a_{clearly identifiable}rosette.

3. Staggered Position printing's tone reproduction capabilities are

equivalent to both four angle printing's and dot-on-dot printing's tone

reproduction capabilities.

4. Staggered Position _printing is equivalent in sharpness to four angle

printing at 85 lpi and significantly sharper than four angle _printing at 65 and

100 lpi.

5. Staggered Position printing is equivalent in sharpness to dot-on-dot

printingat85and 100 lpiandsignificantlysharperat65 lpi.

6. The color variability (color shift) exhibited

by

Staggered Position

printing

is equivalent to the color variability in four angle and dot-on-dot

(57)

Staggered

_{Position printing} at65lpi is

significantly

morevisible.

7. Staggered Position

printing

is equivalent to four angle _printing for

overall visual appearance at 85

lpi,

more acceptable at 100 lpi and less

acceptable at65 lpi.

8. Staggered position _printing is equivalent to dot-on-dot _printing for

overall visual acceptance at both 85 and 100 lpi and more acceptable at

65 lpi.

9. There is no correlation between Delta E values and the observers'

perception of a colorshift.

10. Staggered Position _printing is equivalent to four angle and dot-on-dot

(58)

CHAPTER

9

DISCUSSION

The rosette pattern is considered

by

many

to be

distracting

and objectionable, often

times blamed for the loss of _{sharpness, but}

becoming

less noticeable as the screen

frequency

becomes finer. The results of this investigationsupportthese observations. This

thesis investigation also confirmedtheeliminationoftherosette patternasan option withfew

disadvantages. Therosettepattern isavoided

by

_generatingseparationswiththesame angle

forall four colors. The result is a sharper reproduction at even the coarse screen rulings.

Staggered Positionordot-on-dotseparations are a safe_option,withtherisk ofmoire patterns

being

minimal. Further investigation inthisarea could_studythescreen_rulingon newsprintat

which rosettes are unnoticed, orthe correlation between the perception of sharpness and

thevisibilityof rosettepatterns.

The obvious strength of the Staggered Position process _{is its ability} to produce an

image with superior sharpness. Dot-on dot printing had produced such sharp images in

previous studies that

any

alterationto its dotconfiguration wasbelieved tocompromisethis

ability,

however,

Staggered Position _printing enhanced this ability. This enchancement is

more than just the elimination of rosettes. The dot configuration in Staggered position

separations providesmatricesofdotsthatdescribethesamedetail inan original withfour dots

in four different positions. Thedot-on-dot separations confinethedotsto the same

dot-on-dot position forthefour colors andtherefore describe the detail of the original at a single

position.The four angle separations hold the dot positions to the four angle configuration

describing

detailinamanner subjecttotheanglesofthescreens.

A diagonal line passingthroughanimage willmorethan

likely

resultin "stairstepping"

(59)

diagonal line reproduced

by

the

Staggered

position method will be smooth because more

dots have been usedto reproduceits form.

The

effectiveness of

Staggered

_Position

printing

to produce sharper images on

newsprint is a significant result.

Generally,

a reproduction on newsprintwill look soft and

unsharp. To compensatefor

this, increased

edge enhancement or_{unsharp masking is built}

into the separations, but this often times results in objectionable lines in and around the

subject of the image. The

Staggered

Position method provides a means to increase

sharpness without this excessive detail enhancement. _{Further study} could compare

Staggered Position

printing

with normaldetailenhancement andfourangle_{printing done}with

different degreesofdetail enhancement.

Colorvariation or colorshift does not appearto be as significant a problemforeither

Staggered Positionor_{dot-on-dot printing}on newsprint as inpastinvestigations. The images

in the previous studies of _{dot-on-dot printing}were rejectedfortone reproduction becauseof

the color shift. These studies were done on coated stock. Inthis study on newsprint, the

tone reproductionwas measuredto be equivalent, and thejudgesfoundthe methods equal

forcolor shift in twoof thethree scre