THE SECOND ALARMING TREND
Updated 9/27/01
Although CD and digital
technology has fostered many wonderful improvements in our recordings, it has also spawned
a trend that is absolutely trashing the sound quality of many potentially great sounding
albums. And that is the current trend of LOUD, LOUD and LOUDER.
In many ways this is an
offshoot of the "first alarming trend" as described above; mastering being
performed by unqualified people. The human ear can be very easily fooled, and differences
in "perceived loudness" is one thing that can most easily confuse our ear/brain
mechanism.
Here are some
interesting facts about what we hear versus what our brain actually interprets. (For more
information do a web search on "psychoacoustics".) First of all, if given a
choice between two altered versions of the same sound, we will almost always choose the
louder one as the better sound. It's just a fact in the complex way that our ears work.
So, if a mastering
engineer is compressing and limiting your music so that it is eventually twice as loud as
the original, and then he does an A/B comparison for you, you are going to say the louder
version sounds better. It will appear to be clearer, punchier and brighter.
The reason for this is
quite surprising. Our ears (or our brain, rather) adjusts it's own internal equalizer
depending on the volume of what we are hearing! (For more information, search out info on
the "Fletcher-Munson Curve".) At low volumes, we lose our sensitivity in the
lowest and highest frequencies, so the sound appears midrange heavy. As you increase the
playback volume, the music appears to also increase in bass and treble content. At
approximately 80 to 85 dB spl, we hear the signal as it truly is. (That is why 85 dB spl
is considered the ideal monitoring volume for mixing and mastering.) As the volume
increase above this, our hearing mechanism actually boosts the bottom and top end! So at
loud volumes, we hear more bottom and top end than there actually is.
So, if you lower the
output of the processed signal to match the "perceived loudness" of the original
signal, then you will be able to hear the true difference. If the music was properly
mastered, it will be noticably improved in clarity and punch. And it will most often be 2
or 3dB louder than the original. If the music has been over or improperly procesed, what
you will hear in the processed version is a loss in clarity of the upper midrange, a
brittle harshness in the high frequencies and a myriad of trashy artifacts in the quietest
musical passages being caused by the compression, especially if it is digital.
Even with the very best
mastering compressors and limiters, there is no free ride in this regard. A good mastering
engineer using true master-quality gear can minimize the degradation of the signal, but
once you go beyond a certain point in loudness, the signal will suffer in some ways. It's
up to you to decide if the trade off is worth it for the extra few dB of level.
To compound the
problem, so much mastering is being done now through sub-par equipment, whether it is the
newest cheap digital box, tube gizmo or workstation plug-in. And it is often being
performed by individuals who don't have the experience necessary to operate the gear
properly or to be able to hear the problems they are creating.
So, the trend toward
louder and louder CD's is pretty much an offshoot from this; the unqualified mastering
engineer has learned that all he has to do to impress the client is to make the signal
louder. And by comparing it to the original signal without compensating for playback
volume, it appears that a significant improvement has been made, when in fact the sound
quality has been degraded. And the client is even more impressed when this mastering
genius can actually make it LOUDER than the loudest CD you can bring in for comparison.
The result of an
over-compressed/limited signal is what is called the 'loud tiny sound'. It has really hot
average level, but listen to it at medium to low volumes, and it sounds small and wimpy.
In fact, this is precisely why Muzak sounds the way it does. The Muzak signal is extremely
compressed, but surprisingly not as compressed as a lot of current release music.
Now I will get on the
soap box. We have control of this ourselves. Do not be drawn in by the seeming allure of
loudness. It is up to you as a musician/producer to say "enough". Of course you
want your CD to be competitive in volume when it comes up on the changer or juke box, but
resist the idea of 'louder is better'.
Why do I consider these
two trends to be so alarming? Simply this; we are right now ruining the sound quality of
this current era of popular music. When all of this wonderful technology has made it
possible for us to improve things, instead we have abused it and are rapidly going
backwards in sound quality.
When I first got into
mastering, I used to think that the saddest thing was that so much music outside of major
releases went unmastered. If it had just sounded right, it may have been accepted. But
these current trends are actually worse. The reason is simple: If the original masters
could be found on the old, unmastered music, it could be properly mastered now. But if a
signal is over compressed, limited or eq'd, the resultant loss in sound quality cannot be
recovered! And the problem is, this is often now being done during the mixdown, so there
is no unprocessed original master.
.................
|