American Cinematographer - December 1996 - 94

Robin, after having previously

the fact that the

teamed with

cinematographer
Stephen Goldblatt, ASC on Batman
Forever. "If you're matting some¬
one in an
all-green suit, using a
greenscreen is obviously going to

used in the

make it a lot more difficult to sepa¬
rate that person from the screen it¬
self. A number of factors contribute
to the success of the process, but
the actual idea behind the process
of color difference traveling-matte

light sources used to illuminate
the backings often contain large

photography in front of a blue,
green or red screen is to capture the
difference between the subject and
the color of the

screen

behind the

subject. The greater the difference,
the better the matte. Human fleshtones

are

mostly composed of red

fair amount of green infor¬
mation. Of the three color records
and
on

a

film, the color that least contrib¬

utes to the

rendering of human

skin is blue. You don't get too
much of the shape and texture of
human beings in the blue record.

However, you
situation

in

might also have
which

a
Arnold

a blue
shot. The closer that
costume is in tone and hue to the
color of the bluescreen, the more

Schwarzenegger is wearing
costume in

a

pigments

screen are

or

densities and balances

dyes

not theoreti¬

cally perfect; that there is
'cross talk' between the

are

grossly

exceeded, the software and algo¬
rithms

creating the digital compos¬
greatly compromised. In
practice, I find that shooting a film
test a day or so ahead of the actual
bluescreen or greenscreen shoot is
very important in order to calibrate
my light meters, lenses, and film
stock. After the test negative is pro¬

some

ite

red, green

and blue records within the emul¬
sions of all films; and that the

amounts of red and green energy.

It is

important to achieve the great¬
separation in the film negative
between the foreground object and
backing through the proper choice
of materials, equipment, and pho¬
tographic technique. I have found
that being able to light the fore¬
ground object separately from the
green or blue backing helps im¬
mensely in controlling the quality
and accuracy of the finished el¬
ement. Properly exposing the
foreground object and balanc¬
ing it to the level of the screen is
also very important.
"For example, an actor
wearing a white shirt against a
bluescreen should be exposed in
such a way that the 'blue' compo¬
nent and density of the white shirt
is less than or equal to, but not in
excess of, the 'blue'
component and
density of the bluescreen. If these

can

be

cessed, it is read on a color densi¬
tometer. As a standard, I expose a
Kodak gray card to obtain negative
densities equal to LAD (Laboratory

est

Aim

Density), which Kodak speci¬

fies for each of its film stocks. For

instance, in the

case

of 5247, Kodak

recommends Status M RGB (red,
green, blue) densitometer readings
of .80ND, 1.20ND and 1.60ND for

properly exposed gray card un¬
tungsten light. I would use this
density information to expose my
foreground subject. As for the

a

der

bluescreen

or

greenscreen expo¬

I tend to look for

a negative
density that is between .30ND and
.40ND above the LAD density for
the appropriate colored screen
(blue or green). In other words, for
sure,

bluescreen, I would look for

a

screen

density

on

a

the negative be-

likely it is that the costume will
become indistinguishable from the
bluescreen itself. Therefore, the
notion of doing the shot via greenscreen becomes more valuable."
"When you talk about the

Another of
Schuman's

paint. For The

backing, whether it's greenscreen
or bluescreen, what
you're trying
to do, photographically, is to sepa¬
rate the foreground objects from
the background," relates Hoyt

Ghost and the

Yeatman, founder of Dream Quest

Darkness, the

Images. "Through the use of expo¬

charts, this time
for

Tempo Blue

fabric and
Rosco's Blue

compositors
requested a
blue layer
density of

and the color difference be¬
tween the foreground objects and

is also included
for both the
blue and green

certain amount of red and green
information found in the film nega¬

material tests.

tive of the bluescreen. This is due to

92

American

2.00ND, which
is indicated

by

Schuman's LAD

figures of .20,
.80 and 2.00 for

the two blue
materials. For

a

comparison
reference,

Test No: 17

Meter Used: Pentax V w/ Zone VI Conversion

Stock:5293-273

ASA: 200

Lens:

an

exposure

wedge of an
18% Gray Card

Cinematographer

N

Anamorphic Zeiss 100mm

Surface

Kino Flo

Light
L. A. D.

Super Blue

=

f/ 5.6 Vi @ Vbo

Filter: 0

E.V. 9.1

Date: 11/15/95

sec.

.18

Base D-Min:

Rosco Blue Paint 5722

Tempo Blue Fabric

sure

bluescreen/greenscreen backing,
we are able to extract the
objects
digitally or photographically and
place them into a new environ¬
ment, creating a traveling-matte
composite. For instance, if you
were
shooting a bluescreen, the
blue layer of the film would ideally
be exposed only by the pure blue
backing and would contain little or
no
exposure or information in the
red and green layers of the film. In
practice though, there is always a

here

Prod. Title: The Ghost and the Darkness

°

Tungsten @ 3100 K

Kodak 18%
E.V. 9.8

.20

.80

2.00

.20

.80

2.00

I

I

I

I

\

I

R

G

B

R

G

B

1

I

.88

± N

+3

272

+3

272

+3

+272

2.8

+272

2.8

+272

±N

m
2.8

1

2.8V6 +2

2.872 +2

.24

1.11

2.42

4

+VA

.23

1.05

2.37

4

4/2

+1

.22

.97

2.29

5.6

+1/2

.20

.86

N

.20

8.0

-72

8'/2

-1

11

f/

±N

.91

Gray Card

Tungsten @ 3180

M

V

.57

°

K

1.27

m<

CD

1.61

I T

I

R

G

B

.85

1.46

2.36

1.14

1.71

2.05

+ 172

.79

1.39

2.32

4

+ 172

1.08

1.65

1.97

472

+1

.73

1.30

2.25

472

+1

1.02

1.56

1.88

2.17

5.6

+72

.65

1.19

2.15

5.6

+72

.93

1.45

1.77

.77

2.06

5.672

N

.59

1.11

2.07

5.672

N

oo cn

1.35 1.68

.19

.71

1.96

8.0

-72

.53

1.05

1.98

8.0

-72

.79

1.27

1.59

.19

.66

1.87

872

-1

.48

.98

1.89

872

-1

.73

1.19

1.51

-172

.18

.64

1.77

11

-172

.40

.91

1.78

11

-172

.65

1.09 1.41

11 72

-2

.18

.62

1.69

1172

-2

.33

.84

1.69

1172

-2

.60

1.02

16

-2 72

16

-272

16

-272

1 6I/2

-3

1672

-3

1672

-3

5.6U

December 1996

2.872 +2

1.32
American Cinematographer - December 1996

Table of Contents for the Digital Edition of American Cinematographer - December 1996
