
contrast ratio, and color accuracy. Measure-
ments were made after adjusting for the
maximum calibrated light output (Contrast
control) with a peak-white PLUGE pattern
(98 IRE stripe against a 100 IRE background),
and the optimum black level (Brightness
control) with a conventional PLUGE pattern.
With the gray scale manually calibrated
to D65, the F5.0 (high contrast)/Low Lamp
mode produced 13.4 fL from my 1.3 gain,
85.3-inch wide, 16:9 Stewart Filmscreen,
which slightly exceeds the 12 foot-Lambert
(fL) SMPTE recommendation for digital cine-
ma. The on/off contrast ratio measured an
exceptional 3005:1, which produced an
absolute black level of just 0.0045 fL when
displaying a 0 IRE (black) field.
The F3.0/Low Lamp mode produced
16.6 fL from my screen, and a still excellent
on/off contrast ratio of 2104:1. That mode
could be used on slightly larger screens if
the F5.0 mode produced too dim of a pic-
ture. The High Lamp mode can also be
used for larger screens, but remember that
a lamp may lose 40 to 50 percent of its light
output over its useful life. So it’s a good idea
to plan your screen size with the Low Lamp
mode, and reserve the High Lamp mode for
use during the latter part of the lamp’s life.
The High Lamp mode increased bright-
ness by about 25 percent. The F5.0 mode
produced 16.9 fL, and the F3.0 mode pro-
duced 20.7 fL from the 1.3 gain screen. The
latter is equivalent to about 454 lumens. If
additional brightness is desired for viewing
sports in a dimly lit room, the High Bright
mode can be enabled. It increased the
brightness about 20 percent, but the color
temperature increased to over 10,000K.
The on/off contrast ratio is extremely
important for DLP and LCD projectors
because it determines the absolute black-
est level in the darkest scenes. A tiny
increase in the black level dramatically
reduces shadow detail discrimination in a
predominantly dark scene. I also find it
objectionable if the “black” background in
an image is noticeably brighter than the
black surrounding the screen.
A distinctly different picture quality factor
is the ability to see contrast differences
when there are bright objects in a scene.
Then the displayable contrast ratio is much
less than the on/off contrast ratio because
light from the bright objects will be scat-
tered over the image obscuring darker
objects. The light scattering occurs within
the lens and the optical system of the pro-
jector, but it may also occur within your the-
atre as light reflects around the room and
back onto the screen.
There is no single measurement that can
precisely describe this intra-field contrast
ratio for all images because it depends on
the size and location of one or more bright
objects within each individual image. Hence
we can only define a figure-of-merit based
on an artificial test pattern. One such figure-
of-merit is the ANSI contrast ratio, which
involves measuring 16 blocks (8 white and
8 black) of a 4 x 4 checkerboard image and
then computing the contrast ratio from the
average luminance of the white and black
blocks. This is laborious and fraught with
practical measurement uncertainties.
I’m using a modified “ANSI” measure-
ment that is the average contrast ratio of the
four central blocks in the 4 x 4 checker-
board. The contrast ratio for each block is
measured separately by fixing the position
and angle of the light meter a few feet in
front of the projector, and then electronically
reversing the black and white blocks of the
checkerboard pattern. This increases the
magnitude of the “black” values to minimize
the influence of room reflections and
improves the computational accuracy. It
also eliminates other issues related to mak-
ing measurements off the screen. Starting
with this issue I will include this modified
“ANSI” contrast ratio measurement in each
projector review. The VP-12S3L measured
604:1. The only other HD2+ DLP projector
that I have reviewed measured 728:1. This
is outstanding performance for both products.
As video levels approach black, dither-
ing noise has been another significant prob-
lem in earlier generation DLP projectors. An
excellent way to look for dithering noise is
to view the 1-10 IRE 10-step luma patterns
from the AccuPel generator. Dithering noise
was visible on the 1 to 3 IRE steps of the
luma pattern when viewed close to the
screen, but it was nearly invisible from a
normal seating distance.
White Field Uniformity
Brightness uniformity on a white-field
test pattern was extraordinary. There was
only a 1 percent variation in brightness from
the top to bottom center of the screen, and
7 percent or less from side to side for both
the F3.0 and F5.0 contrast modes. The
color temperature uniformity was within 70K
over most of the screen, with a variation of
about 230K at one edge in both modes.
Color Accuracy
The CIE x,y diagram shows the 720p
DVI RGB color accuracy compared to the
Rec. 709 (HDTV) colorimetry. The same
measurements are also shown on a CIE
u’,v’ diagram, which provides a more per-
ceptually uniform presentation of the color
space. The projector primaries are slightly
more saturated than the HDTV primaries,
but they are balanced so that the position of
the yellow, cyan and magenta complemen-
tary colors are extremely close to the stan-
dard colors. The hues of flesh tones and the
other colors within the color gamut are quite
accurate and images are brilliantly vivid.
The CIE u’,v’ diagram for the 480i YPbPr
input shows that the complementary color
hues are still extremely accurate, even
though the red primary is a bit farther out-
side the Rec. 601 primary for standard-defi-
nition video. You will likely want to use the
YPbPr (or S-video) input, which provides a
Color control if reds are oversaturated or too
vivid on some video broadcasts or a DVD.
Edge And Detail
Enhancement
The projector includes more than a
dozen adjustments for edge and detail
enhancement that separately affect luma
and chroma portions of the image. The
Luminance submenu in the Fine Menu
includes eight adjustments for the Gain and
Threshold of the horizontal and vertical
Edge and Detail enhancement. The
Chrominance submenu includes additional
adjustments for Horizontal Edge Gain and
Threshold, Vertical Edge Gain, Enhancement,
and Delay. The Enhancement item is only
enabled for composite and S-video signals.
The Threshold adjustments determine the
minimum luma or chroma amplitudes that
the enhancement processing will affect. The
Gain adjustments control how much enhance-
ment is applied to edges or fine details.
The Sharpness control (in the Picture
Adjust menu) acts like a master gain control
for the Luminance enhancement processing. If
the Sharpness control, or all of the Luminance
enhancement gain controls are set to zero,
there is no luma enhancement. As the
Sharpness control is increased, the enhance-
ment increases proportional to the individual
Luminance enhancement gain settings.
The Sharpness control appears to work
more like an on-off switch for the Chrominance
enhancement. If the Sharpness is set to
zero, the Chrominance gain adjustments
seem to have no effect. But if the Sharpness
is set to ‘1’ or higher, the Chrominance gain
adjustments take full effect. When the
Enhancement item is enabled, it appears to
have master control over the Chrominance
gain adjustments.
Even when the Sharpness control and
the other Chrominance gain adjustments
are set to zero, there appears to be some
chroma edge enhancement on the Pb
channel. The leading edge transitions of
blue patches in the Snell & Wilcox pattern
of Video Essentials appear to be shortened,
and the Y/C delay pattern on AVIA: Guide
Equipment
Review
Widescreen Review • Issue 84 • May 2004
42
Page 4/7
#84 Master Pages 26-49 3/22/04 4:31 PM Page 42
Commenti su questo manuale