Re: One for the bottleheads

My impression is that tests have been done which take into account the
points you make. >>

Hi Jim - Do you have references? Sounds interesting.


<Thus it is reasonable for you to say you have doubts about a given
set of results for some specific and definable reason(s). But only
if you can then propose a fresh hypothesis which can be tested wrt
to the proposed reason(s) for the doubt. And you can only regard
the doubts as 'valid' and undermining the initial experimental
evidence when you have the 'new' evidence and it undermines the
original evidence. >

Yes, that's exactly what I'm saying for the reasons given - the test
situation should mirror how users typically use audio equipment - they
are very familiar with the software and hardware and have heard many
variations of each in a sighted way at their pleasure. I think it's
reasonable to require that a test situation should replicate the
BEHAVIOUR we are seeking to explain. (i.e. why people report such
things) Hence I don't think Occams razor applies. I would add that I
see no reasons why one should not use such preconditions. In the Quad
test the Yamaha Ls100 was imposed on the test subjects (apparently)
which I would regard as unsatisfactory since the subjects were not able
to chose ancilliary equipment which satisfied them.

<<This does rather imply that you view is that the audible differences
are so
subtle/slight as to be easily obscured unless great care is taken when
comparing.>>

This may well be the case, indeed. But I think we both know this.

<<Yet my understanding is that many people claim things like that these

differences are 'obvious' or that 'amps tend to sound distinctly
different from one another' - even when they took no such precautions
as you demand when coming to this conclusion.>>

That's what they "say". I have as many doubts as you do over how well
this could be replicated in a DBT. In a DBT I would anticipate
considerably less 'obvious' differences, and in at least some cases a
failure to differentiate at all reliably.

<I presume, therefore, that you do not really agree with them, and tend

to regard most of such claims that people make as likely to be simple
mistakes on their part. Is this the case?>

Not simple mistakes as such, but mistakes for some of the reasons we
have mentioned - anticipated results, bias towards preferring a certain
outcome. We may have to include "degree of malt consumed"
here........... I'm quite prepared to believe that users exaggerate
differences on a habitual basis, but what interests me is if
differences can be reliably identified by listeners trained to
discriminate on the best possible ancilliary equipment in a DBT. I
would expect differences to be subtle - the question is if they can be
identified to any significant degree. This may not be 100% but it might
well be much more than random.

.

Relevant Pages

  • Re: An Open Letter To Rob Savoy
    ... I say that for 2 main reasons: ... people acknowledges that people make mistakes. ... choice rob made is his own, he's certainly done enough for the family ... That you're better than bodhi? ...
    (alt.gathering.rainbow)
  • Re: Logic in Schools
    ... >> opposing the notion that logic should be taught universally, ... "aptitude" results in mistakes and that correction those mistakes ... >> The same reasons that I already gave you: ... >> 'mathematical minds', or however you put it); ...
    (sci.logic)
  • Re: High end, small frame mixer for under $1000?
    ... kinds of mistakes that there is much to learn from, ... that it is a waste of resources. ... One of the reasons for performing uncontrolled ...
    (rec.audio.pro)
  • Re: Giving an application a window icon in a sensible way
    ... mention the possibility of actual adverse consequences from trying ... accept that when mistakes are made, there are always reasons for them. ... impression that you have wholly benign intentions either. ...
    (comp.lang.java.programmer)