Audiophiles are dedicated to the improvement and refinement
in the quality of the sound they hear from their sound systems. They change
interconnects and listen for improvements. They change components and listen
for improvements. They usually use familiar recordings as a reference to help
them determine if there are improvements.
A good part of how a system sounds depends on its setup and
speaker position. An equally important aspect of the set-up is how close the
speakers are to the front and side walls of the room, as well as how close the primary
listening position is to the back wall. Eventually most people discover there
is more to how the system sounds than merely its geometry of the listening
When an upgrade is made in a system a perceptible improvement
in sound quality is usually delivered to the listening position, which
justifies the investment. As the upgrade process progresses, eventually a
better piece of electronics that “should” have made a noticeable improvement,
doesn’t. Here’s where the evolution of the audio system often grinds to a halt.
Product reviewers may have given a new piece of gear raves, but
what happens if it doesn’t make any difference at home? Is there a problem with the equipment,
manufacturer, dealer or reviewers? Probably not, but the audiophile may have
reached a performance plateau. From this point on any further electronic
attempts to improve a system will remain inaudible.
Lacking the ability to detect any further improvements in their
system, audiophiles frequently resort to entertaining upgrades that may not
necessarily “improve” the system’s performance, but do create detectable
changes in the performance. However these artifact styled “improvements” become
boring after a while, because they add the same accent to all music, making it
For most audiophiles hitting the upgrade wall is a big
disappointment. They hear better performing systems in other rooms, but since
they can’t seem to get their system to perform better. Faced with this dilemma
some audiophiles resign themselves to their lackluster systems as defining the
end of the road. They give up on the idea of trying to get better sound.
Let’s take a look at
what is really going on.
If program material is buried too deep in the noise floor, no
one can hear it, even though it is there. The noise of an LP record creates a
noise floor that is about 40 dB below the main signal level. Tape machines are
better, with a tape hiss noise floor that was about 60 dB below the main signal
level. And with digital, the signal to noise ratio can be 100 dB or more.
The problem is never about hearing or not hearing the main
signal. The problem is hearing the subtle detail within the signal. The main
signal will always have a strong tonal presence that combines with dozens of
lower level partials or overtones. Some overtones are very low in level. When audiophiles
listen over quality headphones even these quiet musical details are readily
apparent. But when the same selection is played back through a room-based sound
system, these quieter musical details disappear, rendering the playback less
Through reading, web forums and discussions with friends some
audiophiles begin to imagine that the problem might not be with the electronics
of the system, but with the acoustics of their room. The room acoustic is
creating a higher noise floor that is masking the fine, low level musical
A sound meter shows that music is typically played at a level
of about 75 dB, A-weighted, It also shows that the background noise floor in
the room, with all the music off, is about 25 dB A-weighted. That means there
is a 50 dB signal to noise ratio in the room acoustics. This is the steady
state signal to noise ratio.
The problem with listening to sound in a room is that once we
hear the sound, it continues to be reflected around the room until the sound
dies. And music isn’t merely one sonic event followed by a reverberant decay. There
are, on average, about eight separate dynamic sonic events per second, each of
which is followed by their own reverberant decay.
In small room it takes, on average, at least one second for a
particular sound to die down enough so it becomes inaudible. Playing music in
small rooms increases the self-noise floor, which is made up of the various
reverb levels from the music that’s been played during the previous second. The
strength of this self-induced noise floor can equal or exceed the strength of
the direct signal.
When audiophiles listen to a recording over good headphones
they experience a signal that has a very quiet background noise floor. But the
same program material played in a listening room produces a 0 dB to -5 dB
signal to self-noise ratio. Low-level musical detail typically exists in the
range of -20 dB. Over headphones, low-level details remain audible, but in a
room these details are buried in the room’s acoustic self-noise floor.
Audiophiles work on room acoustics because beyond a certain
point in the evolution in the performance of their audio system reducing the
room acoustic self-noise floor in the room becomes the only possible
improvement they can make.
But the ironic thing is that it’s not until after room acoustic
upgrades are finished, that further improvements in the electronics become
audible. But audible progress occurs only to that point where the room’s self-noise
floor again has to be addressed. And back and forth it goes, alternating
between electronic upgrades and room acoustic upgrades, continually stepping
forwards towards perfection in high-end audio
Art Noxon is a fully accredited Acoustical Engineer with
Master of Science degrees in Mechanical Engineering/Acoustics and Physics. A
professional engineer since 1982, Mr. Noxon is licensed to practice engineering
in the public domain with a specialty area of acoustics. He is the inventor of
the TubeTrap, the original corner-loaded bass trap with built-in treble
diffusion. He is the president of Acoustic Sciences Corporation.