Re: Record demagnetizers

In article <ksqpj3pijjoqp0mndl7876nha7enbhvn23@xxxxxxx>, Signal
<nobody@xxxxxxxxxxx> wrote:
Jim Lesurf <jcgl@xxxxxxxxxxxxxxxxxx> wrote:

This is why double-blind trials are just as important in assessing
the sonic qualities of HiFi components as it is in testing new drugs.

Well I don't like to draw direct analogies between drug and hifi
blind tests (which are far from equivalent), but yes the underlying
principle is nice in theory. However, pertaining to controversial
areas of hifi, there are practically no DBTs of a grade which could
be called conclusive.

You would need to explain what you mean by "conclusive". Perhaps giving
one or two specific examples to illustrate. :-)

Let me quote Ludovic Mirabel...

[snip details for brevity]


Ernesto is absolutely correct. The methodology of component comparison
using this kind of "DBTs"- as published in the 1980's audio mags etc- is
way below the quality of the worst of any published psychiatric and
psychotherapeutic results- and that is saying a lot, believe me. "

The problem with what you said is that you are taking one specific type of
DBT in one field and then simply assuming that this means that problems of
the same kind must arise with a *variety* of types of test, carried out for
a different purpose, in a quite different field. That is a rather dubious
attempt to argue by analogy.

The design of an experiment depends upon the hypotheses to be compared or
discriminated between. Ditto for the protocol, etc. You also can expect to
employ a variety of designs and protocols in complex situations. And to
then discriminate between what the results can, or cannot, bear evidence
upon. Saying there have been misapplications and unreliable conclusions for
some designs in one field does not make that a general assertion.

The trials I have read about often do seem "conclusive" in a quite
clear sense. In many cases they give results which allow us to conclude
that under the conditions of the test, those involved were unable to
show they could tell one item/arrangement from another used in the
comparisons.

For sure, "under the conditions of the test", which often leaves a lot
to be desired.

Again, you'd need to be specific in the relevant cases - i.e. for such
audio experiments, and to give evidence for your argument. Not just make
assertions and cast doubts as 'possible'. :-)

I worked on a multi-million double blind clinical trial at ICI, and
by comparison the majority of audio DBTs I've heard about are
laughable.

FWIW I've also been involved in medical trials which were very dubious,
despite being large-scale.

Dubious how?

Very poorly controlled, etc. This probably isn't the place to give details,
but they involved a close relative who is seriously disabled, and was
willing to give time (and risk her life and health) to be a test subject in
the hope that the results might benefit others with similar conditions. The
trial itself didn't keep careful track of what medications were taken, etc.
Nor sufficient details of possible side-effects which were left to the
judgement of the consultants whose patients were being used. IIRC I was
also unimpressed by the final data and analysis as it seemed too sparse,
and - as is often the case in academic papers - I had the impression they
simply applied 'standard statistical calculations' with no real thought.

However that was my impression as a physicist judging another field, so may
be unfair. But I was not particular impressed.


Right, well when the funding cheque arrives in the post! You may not be
surprised to learn I don't have the resources to perform a large scale
laboratory grade trial!

You don't necessarily need to. It is quite possible to run a valid
comparison simply using a switchbox or someone to change leads, etc. Some
simple gear like an rms voltmeter or scope (probably using a soundcard and
computer for these) would suffice for many types of test. Depends what
hypothesis you wish to probe. The idea that this is all has to be vastly
complex and costly is wrong. What you do need, is to understand the most
likely confusing factors, how to design and run a comparison, and the time
and patience to do so and analyse the results. The main 'cost' would
probably be the equipment being compared. But this depends what type of
test, for what purpose, you have in mind.

Slainte,

Jim

--
Electronics http://www.st-and.ac.uk/~www_pa/Scots_Guide/intro/electron.htm
Audio Misc http://www.audiomisc.co.uk/index.html
Armstrong Audio http://www.audiomisc.co.uk/Armstrong/armstrong.html
.

Relevant Pages

  • Re: New webpage on loudspeaker cables
    ... devices in audio power amps have gain at frequencies ... any audio amplifier, you can even have software filters with monstrously ... you can design and build amplifiers that are unconditionally stable ... be wowed by amps with a high gain response up to 30 mHz or more. ...
    (uk.rec.audio)
  • Re: 2A3 - Really a beautiful power tube!
    ... I had hoped someone would have addopted my design and wound a quantity to sell back ... But several amateur audio ppl persevered to wind my OPT No1, ... One who winds his OPTs, ...
    (rec.audio.tubes)
  • Re: Remote Power Supply Connection
    ... Tube audio is pretty addictive ... I had 4 ppl just this week with failed Marantz amps. ... They want a simple preamp design ...
    (rec.audio.tubes)
  • Re: [geek] was [Moderate geek. . .]
    ... "10W audio amp. ... For hefty signals a *super*-regenerative design can be quite selective ...
    (uk.misc)
  • Re: Record demagnetizers
    ... sonic qualities of HiFi components as it is in testing new drugs. ... comparison the majority of audio DBTs I've heard about are laughable. ... are turned into evidence for/against a given hypothesis. ... reliability. ...
    (uk.rec.audio)