"Robert Morein" wrote in message
...

wrote in message
oups.com...

Robert Morein wrote:
wrote in message
...

[snip]

I tend to agree with you that people tend to imagine differences, and
that
ABX diminshes that.

Yep.

But I also see very strong evidence, as collated by
Ludovic, that ABX diminishes real differences.

Or makes it obvious that the "golden ears" crowd can't always hear
differences.

Don't assume it, Neil. Try to be nonbiased.

In blind test methodology, each sample is
labeled with a bland symbol, such as A, B, X, Y, etc.

The "X" in ABX doesn't represent a particular sample of a component.
The "X" can be any component in the test.

However, the assumption that each of the subconscious processes in the
brain
can work with sample labels of this type is an assumption. Suppose an
AB
test were performed in which the bland labels were replaced by other
kinds
of labels, ie., pictures of fictitious amplifiers, or Picasso nudes?

Replacing the bland labels (A, B, etc.) with other, more interesting
symbols, might be entertaining, and the "golden ears" crowd might
decide that the entertaining labels might make the gear sound better or
those folks could attribute whatever subjective expectations they
prefer to the test. But I think replacing the ABX labels would be silly
and pointless.

You are assuming that, Neil, and investigation might prove you right.
However, it is bad science to rely on such an assumption.

Then it's bad science that has beenadopted in one form or another by
virtually everyone working on audio research..