New vs Vintage
Please note the interpolations. Please forgive the cynicism
expressed.
[ Please attribute your quotations. --dsr ]
William of Occam would have that figured out in a hummingbird
heartbeat.
Either it does or it does not. Which requires a more complex
explanation? That it does or that it does not?
Exactly the same answer. Either it does or it does not. Despite what
one may or may not be told. That one is told a phenomenon exists does
not make it so. Not even a little bit. That one is told something does
not exist does not make that so either.
But it is an excellent way to screen whether something does actually ma=
ke a
difference. No more. But very often that is enough. And very often
fatal to closely held beliefs and other forms of revealed religion.
Sadly.
There's no doubt in my mind that DBT testing removes sighted and
expectational bias from the equation. The test participants simply do not
know which of the two units being compared they are listening to at any g=
iven
point. Beyond that, we seem to be taking the results of those tests on fa=
ith.
Faith is not part of the equation. Using the term "faith" requires
that one does not believe the test ab initio - see "closely held
belief and other forms of revealed religion" above. Either one trusts
ones ears or one does not. And if one trusts ones ears, faith is not
part of the equation. If one does not, then one is as a lamb on the
altar ready for slaughter - no further discussion required.
It seems to be taken for granted that if there is a difference between th=
e
sound of two components, that these differences will be immediately appar=
ent
at the switch point.
I would posit that the 'immediately' needs to be defined. Some fairly
subtle things may take more than a very few seconds of switching back
and forth. Shortly, certainly. And level matching is absolutely
critical for this test to be valid.
I.E., one second you are listening to component A, the
next, component B. Since humans have such a poor aural memory (what we
remember about a sound seems to be our impressions of the sound which we =
take
note of as we listen, not the sound itself), Any difference between the s=
ound
of the two components should be the most noticeable at that point. Now, I
know that this works fine for speakers - they all sound so different that
those differences stick-out like a ham at a Sader, as they say. I suspect
that differences between phono-cartridges would be a similar deal, even
though I've never heard a DBT of phono-cartridges.
Been there - yes, they do.
Perhaps analog tape
recorders would exhibit similar results, I don't know.
Yes, they do.
It's when we get to modern amps, preamps, CD-players and DACs that I star=
t to
get uneasy with the process.
Why? Either you hear something that you can predict and identify with
greater than 50:50 odds, or you do not. If you do not, then there is
no difference - to you. And at that time.
All of these devices exhibit ruler-flat =A0
frequency response and vanishingly low distortion these days, so that
eliminates two very important variables in the human auditory perception
pantheon. The two things that we most readily notice, frequency response
aberrations and high amounts of harmonic and IM distortion are removed fr=
om
the equation. So, what's left? Some say that there are types of distortio=
n
that we can't easily measure, but to which the ear is sensitive.
Either you will hear it or you will not. One of my favorite examples
is an early Dynaco ST-120 using the 2N3055 output transistors. Great
measurements, sounded like glass-in-a-blender. With a few basic mods,
it became a (barely) reasonable unit. Measurements are not necessarily
the final issue. What one hears is the issue. And I betcha that even
you with a severe head-cold wearing ear-muffs could pick an early-
version Dynaco from _any_ other amp in a DBT nine times in ten.
These
include, transient intermodulation or slew-induced distortion (after Otal=
a),
dielectric absorption distortion (after Jung) etc. But these are
controversial. Many audio experts maintain that they don't exist, and in
either case, whether you believe them to be a factor, or not, they are bo=
th
addressed in most modern amp designs. That leaves noise, which again is
vanishingly low in modern analog devices.
You read reviews of amps that allude to textures in the top octaves, such=
as
this device lends a sandpaper-like quality to the reproduction of strings=
,
while this other device is more liquid, smoother sounding in the same reg=
ion.
I read a lot of blather created by writers who have to justify their
paychecks. I do not read information based on an opinion supported by
observable facts. I also note that these same writers are bat-sh*t
terrified of DBT as it would pretty much wipe out their species if
adopted. And so will do all-and-everything-they-can to debunk it.
But what could account for these differences? noise modulation? =A0Some k=
ind of
heretofore undiscovered distortion? This doesn't seem reasonable, and eve=
n if
these differences do exist, would they be obvious at the switch point in =
a
DBT, or would they go unnoticed, giving rise to the result that there is
statistically no difference between the two devices?
Yep. exactly that. And the problem with that would be?
Would this result in a
difference that might only show itself in long-term listening?
Perhaps. Even possibly likely. And for any number of reasons.
These are
questions that I find unsettling in the DBT vs long-term listening debate=
..
And this brings up one last question. If the differences are so subtle th=
at
DBTs cannot uncover them, are they worth obsessing over?
Of course not. Obsession is the realm of closely held beliefs and
revealed religion.
That's a question
that every audio enthusiast is going to have to answer for themselves - o=
nce
the DBT issue is put to rest, once and for all, of course.
Not really. If only audio enthusiasts would stick to what their ears
tell them, enjoy what their ears tell them and behave only based on
what their ears tell them then the concept of DBT or not-DBT is
entirely irrelevant - unless one is attempting to write doctrine and
tenets of faith.
But keep in mind that entire industries are based on the concept that
these differences *necessarily* even when inaudible. exist. Further
that they are necessary issues for discussion. Keep in mind that
Parmedis got to *Ex nihilo, nihil fit* even before Occam.
DBT will ferret out differences. Some more quickly than others. But it
will ferret them out. It WILL NOT tell you which is better, worse or
more indifferent than the other. Only that there is (are) (a)
difference(s). DBT is a means to screen. End of Validity.
What you experience at home over the long term is not relevant to what
DBT can tell you. THAT is based on what your ears tell you. No more.
No less. And DBT was never meant to be anything more than a detailer
of differences-if-any - more so - those differences-if-any audible (or
not) to the target testers.
Peter Wieck
Melrose Park, PA
|