The term "complementary colorations" is one
that I have coined to describe the circumstance
where in a hi-fi system a series of components and cables are assembled to result in a good
overall "sound." On the surface this sounds like not only a reasonable way to do things, but
even the correct and normal way to put together a system. In many cases this is the way systems
are assembled, but doing things this way often has certain unanticipated consequences.
Having said this, the best way to assemble
a system is without any "complementary colorations"
at all! But, before going into how to do that, let me explain more about "complementary colorations"
and what it is.
An Example of Complementary Coloration
Here's are simple examples of a series of components and cables that exhibit complementary coloration...
The idea behind "complementary colorations"
is to correct a deficit, excess, or tendency by adding in a
component or components that does precisely the opposite, thus creating a perfect balance.
The Problem of Complementary Coloration
The problem of "complementary coloration"
is that it is actually impossible to find or create an exact
opposite of one "coloration" by employing another. There may be a way to do this in a laboratory test
where very specific changes are first made, and then a "reverse" filter applied, but in practice, this is
an impossible task to perform - certainly it is wildly improbable that such a thing could be arrived at
by casual selection of components!
Let's consider what happens in the real world
when one tries to use complementary colorations to create
a balanced "sound."
The following is a generalized conceptual
model to make the idea of "complementary coloration"
clear - it is not intended as a scientific paradigm or theory.
A Conceptual Model of Complementary Coloration For Audio
To conceptually explain the idea, think of
a large transparent sphere with a single point at its exact
center point. The center point is the point of "ideal" and "perfect" sound.
Any given component that one can select is
defined as being at some distance and angle (vector) from
that ideal center point, depending upon its own specific "coloration." Different directions from
the center point are different types of colorations, while opposite directions from the center point
(like "north" and "south") are opposing colorations (like "smooth" and "hard"). Different amounts
of coloration determine the distance from the center point. Precisely equal and opposite colorations
can be said to cancel. Opposite colorations that do not cancel end up at some distance and angle (vector)
from the center point.
For the purposes of this example, I'll simplify
for clarity and say that each component can
be assigned a single unique point in space at some distance from the "ideal" center point.
Now, consider the most simple case - a single
component - if it is non "ideal" it has some vector
(angle and distance) from that center point. Add in a second component, then the result is the
sum and of these two component's vectors - a new point. This "sum point" is then a distance
and angle from the center point.
Now say that a new "sum point" can be determined
from the sum of all system vectors, each vector
being a single system component. If there were only one component, then the new sum point is
that component alone. If we add a second, and third, and fourth component, the sum of all these
vectors is the new sum point.
The distance and angle that the new sum point
is from the ideal center point can be said to be the
characteristic sound of the system. The farther from the center point the greater the overall
coloration of the system.
Examples of Complementary Coloration Using The Model
Consider our sphere model and the system vectors that sum to a non perfect coloration.
IF I define the system components as all
having vectors that happen to fall within a small
sub sphere surrounding the center point, no single component's vector exceeding the diameter
of the sub sphere, then it is clear that the simple addition of a component whose vector lies
outside of the sphere would have a marked effect. In other words, it would be easy to hear
the effect of such a component when added into this model system.
Similarly, if you had 9 components tightly
grouped in that little sphere, around the center
point, and only ONE was added with some vector outside - the entire system would be dramatically
colored by that ONE component!! (indeed, this is a pitfall to be aware of...)
On the other hand, if the components all
have large vectors spread about the larger sphere,
the sum point could end up in precisely the same position as the above system, yet the addition
of a component whose vector will completely alter the above system might have a very minor
effect by comparison - ie. be very hard to hear!
The second case is more or less idealized,
since in most cases the sum point is not likely to be
so nicely placed. More often than not, it is pulled away from the ideal in some discernible
direction (you'd call it things like "bright, dark, recessed, forward, hard, thin, fat, etc...)
In which case small changes brought by a new component entering the system (even pulling toward
the ideal center point) are likely to be swamped by the larger system vectors - or in other words,
be audible, but difficult to discern exactly what or why is going on!
What It Means To You
This conceptual model serves to illustrate
(incompletely) why complementary colorations are
a problem, and rarely succeed in creating a system that is completely satisfactory. Furthermore,
it also illustrates why in many cases changes to a system are perceived as changes, but they are
unclear or nebulous to the listener.
So, the goal when assembling your system
might best be to aim for the minimum coloration in
all aspects and respects - trying to get everything tightly grouped around that single center
The BEAR Labs design philosophy is to aim
for minimum coloration, and to offer products that
are clearly grouped tightly around that conceptual "center point."
I expect to add another
section on how we hear stereo, and why minimum coloration helps us
to not only hear "coloration" but also makes it actually easier to hear and to listen...