The charges in an interline
CCDimager, which is what
most CCDs are, begin an
orderly procession toward
their future existence as digi-
tal numbers like well-
behaved schoolchildren in a
fire drill. At one end of the
imager, the charges move
down and out at the bottom
of the column as if someone
were continually pulling the
bottom can of Coke out of a
dispenser.4
When the last charge has
rolled out of the bottom of the
column, the charges in the
second column shift to fill the
vacancies left by the newly
departed charges. The
charges in the third column
move to the second column,
and the thousands of remain-
ing columns follow their lead
like a panoramic Busby
Berkeley number.5
Unlike the photosites
in CCDs—pretty much
passivecapacitors
that do little but store
an electrical charge
until a control some-
where else tells them
what to do with it—a
CMOS sensor is able
on its own to do some
of the processing nec-
essary to make some-
thing useful out of the
charges the photosites
have obtained from
the light.4
When a column of charges falls out of the imager, it is
detected by the read-out register, which takes them
to an amplifier. Until they are amplified, the charges are
more faint static electricity than electrical current. The
amplifier pumps energy into the charges, giving them a
voltage in proportion to the size of each charge, much
as a flagging video game character is pumped up with
“life force” by jumping on a coin.6
The first thing the
CMOS image sensor
does is use the ampli-
fiers that are part of
each photosite. This
eliminates the need
for the charges to go
through an amplifier
in single file after
they’ve left the sensor.5
More importantly, the
onsite amplifier elimi-
nates the slow class-
room drill CCDs use to
leave their nest. As
soon as the amplifiers
have turned the
charges into actual
voltages, those volt-
ages are read over a
grid of X-Y wires
whose intersections
correspond to loca-
tions of the photosites.
It’s the voltages’ way
of saying simultane-
ously, “Beam us up.”6
What happens next depends on whether the image sensor is a CCD (charged-coupled device) orCMOS (complementary
metal-oxide semiconductor). Don’t bother with the full-fledged names. Everyone uses the acronyms, and they won’t be on the quiz.
You will hear a lot of techno-hype from camera makers citing reasons one technology is better than the other. You can ignore that, too.
The type of imager is just one factor that contributes to the photo that eventually will come out of your printer. It’s the print that’s important.
Whether you’re happy with it or not doesn’t hinge on the type of image sensor. But here, for the sheer joy of knowledge alone, are the
differences in how the two types of chips work.
CCD CMOS
CHAPTER 7 HOW LIGHT BECOMES DATA^109