Re: setting up ABX tests at home

bear <bearlabs@[EMAIL PROTECTED]
> wrote:
> Arny Krueger wrote:
> > "Greg Wormald" <greg.wormald@[EMAIL PROTECTED]
> wrote in message
> > news:folrst02p34@[EMAIL PROTECTED]
> > 
> >> The very set-up of much ABX testing (which is most often
> >> short segments of music quickly switched), cannot test
> >> for what most music lovers are seeking--long term
> >> enjoyment of music listening.
> > 
> > This is a false claim that I've heard on and off for about 30 years,
now.

> Well Arny, where is the correlation between ABX tests and subjective
enjoyment?
> Don't think there is any...?

Preference presumes difference.

You might want to try an ABC/hr, a more common protocol for blind testing
of audio preference.  

> The ABX is intended and designed to merely yield data about audible
differences
> between two variables - nothing more.

> IF it accomplishes that goal is yet not clear and unproven beyond the
specific 
> test that took place. In other words the results of specific ABX tests
can not 
> be generalized to pertain to every situation. While the results *might*
be 
> useful or of general use, it is not an inherent quality of an "ABX"
test.
> The reasons - as have been debated here many times - are multiple.

> > 
> > In fact people have been doing long term ABX tests for about 30 years.
 The 
> > origional ABX Comparator product, had special features to allow tests
to 
> > survive power outages, in sup****t of long term tests.
> > 
> > One of the things that long term ABX tests show is that audiophile
mythology 
> > about long term testing is just that, audiophile mythology.

> No Arny, that is merely your belief and your conclusion.
> It is not fully sup****ted by any factual basis.

> > 
> >> A goodly number of years
> >> ago I ABX blind tested some interconnects, and then chose
> >> the ones I preferred. Three months later I realised I was
> >> not listening as much or enjoying it at the level I was
> >> previously. I found myself turning the music down or off.
> >> When I went back to my old interconnects my listening
> >> increased to it's previous levels of enjoyment.
> > 
> > AFAIK nobody has ever shown that interconnects can be distinguished 
> > sonically using any reliable listening test methodology, unless they
were so 
> > bad that they cause non-trival frequency response variations. And, if 
> > interconnects are so bad that they cause non-trivial frequency
response 
> > variations, they are by definition, defective. 

> This would appear to be a true statement.
> But it is not.
> The reason that it is not true is that the "ABX" tests that have been
touted as 
> providing this information are fraught with unknown variables.
> Let me list a few for the readers:
> - frequency response of the speakers
> - distortion vs. power vs. frequency for the speakers
> - polar response of the speakers vs. frequency
> - reverberation time vs. frequency of the room
> etc. etc. etc....

> EVEN WITH this information, I have yet to read a published ABX test that
would 
> permit someone to precisely duplicate the test. Not enough information
is provided.

> Show me *one* that has the requisite information and controls.

> Dr. Earl Geddes' research - and he is a hardcore 
> academician/scientist/researcher - has shown that absolute values of
distortion 
> do not translate at all to audibility of distortion.

Earl Geddes has also written
http://www.gedlee.com/downloads/Chapter4.pdf

//
If one accepts the fact that subjective evaluations are less reliable than
we would like, then it should
be obvious how objective measurements would play a crucial role in audio.
Whatever one might say
about measurements, for the most part they are stable and yield the same
results over and over.
something that subjective assessments have trouble doing. I used to
believe that objective measurements
alone were all that one needed, and I still believe that objective
measurements can tell us most
of what we need to know about a system or component. I believe that I
could predict the judgement of
a panel of blind listeners with an accuracy of about 90-95% correct from
objective data alone. I know
one company that claims a 99.9% accuracy in correlation between objective
data and panel judgements.
Objective data should always be the place where one starts when looking
for system components.
That is, if one knows how to interpret the measurements that are available
or the data that they
are shown. I find that most people shy away from objective data simply
because they don.t really understand
it.
As far as believing that objective data is all that we need, at the
moment, that probably isn.t the
case. There are things that we are learning day-to-day that have a strong
bearing on the correlation between
measurements and subjective preference. But that.s also not to say that we
can.t ever get to the
point where objective data alone is sufficient. In fact this point may be
closer than you think.

//

Sounds like he's a very strong proponent of objective measurements. I
doubt he's really
as  much on your 'side' of this argument as you think.

> So, the sum total of all this is that ABX is a tool that may or may not
yield 
> information that is useful. Period.

Onbly if you don't define 'useful'

___
-S 
"Hey pip squeak, who's L Ron, some new rapper?" -- Nic