nyob123 wrote




Guess what Sean Olive had to say about Ludovic's remarks?

The inescapable conclusion from Sean Olive's loudspeaker testing is
that human beings perform best when asked: "Which one do you like
better?" rather than "Is A different from B".

Based on my work, this conclusion is completely false since for loudspeaker
tests, I rarely ask listeners the question "Is A different than B"?" In
most cases, the differences between the loudspeakers under test are
measurable (both objective and subjective)and therefore the more interesting
question for me is "Which speaker do they prefer, by how much, and why?"

If listeners can demonstrate clear preferences between A and B in
double-blind tests, then by inference, they can clearly differentiate
audible differences between A and B.


The preson making the above statement also concludes that your tests on
speakers revealed that people were not able to distinguish better quality
speakers from lesser ones when doing so blind

This is not true. In fact, quite the contrary,. Listeners both trained and
untrained can discriminate between good and bad speakers in blind tests so
long as they have normal hearing. Furthermore, there is remarkable
agreement in their loudspeaker preferences. The trained listeners are
simply more discriminating and reliable in their judgments.

My research is being misrepresented by this person to serve their own biased
opinions. This would be readily apparent to anyone who has read our research
papers.

We have done ABX tests where the results have been positive; for example, I
have found positive results between different amplifiers where one had a
high output impedance that caused a measurable frequency response change at
the loudspeaker terminals. Floyd Toole at the NRC used to frequently find
positive test results when the amplifiers were over-driven and had different
clipping behavior, or more audible noise. Today these amplifier would be
considered poorly designed, so it is not very interesting to publish the
results, when there is an easy scientific explanation of the results.

By the same token, if the ABX results are negative there is not much
motivation to publish that either. Who wants to admit that their product is
indistinguishable from a competitors', unless their product costs
significantly less money than the highly regarded competitor's product??

Cheers,
Sean Olive, Manager Subjective Evaluation
R&D Group, Harman International
8500 Balboa Blvd.
Northridge, CA, 91329



I do not believe Mr. Olive had said the above. How can one be sure ?