Hi CV, Edmondwolfman, and all,
Interesting discussion, and I think a few comments are appropriate. Until recently, there has been very little controlled double-blind testing of trained versus untrained listeners in loudspeaker listening tests. As many regulars on the forums know, I participated for years doing double-blind listening tests of various speakers (for magazines I was editor of) at the National Research Council in Ottawa, Canada. Indeed there was a training period for all regular members of the listening panel so we could quickly zero in on such speaker colorations as "fat" bass and other anomalies.
In recent years, however, acoustical scientist Sean Olive (who was finishing his graduate work at the NRC under Floyd Toole when I was on various listening panels) conducted an elaborate study using 300 untrained and trained listeners at the Harman acoustical facility, which is a more elaborate setup than the NRC's listening room. What's fascinating about the results is that essentially the untrained listeners' ratings of good and bad (accurate and non-linear or inaccurate speakers) nicely tracked the ratings of the trained listeners. The major difference was that the trained listeners gave lower scores overall than the untrained listeners.
You can read the study in greater detail at Sean Olive's blogspot and download the study for free here:
http://www.aes.org/e-lib/download.cfm?ID=12206&name=harman I'd also point out that tube amplifiers do not necessarily sound different from solid-state amplifiers (except when pushed beyond their output limits into distortion). Tube amplifiers can be well designed for linear smooth frequency response and low distortion. In one double-blind test of amplifiers at the NRC in which I participated, there was a large tube MacIntosh amplifier that measured flat and had low distortion. No-one on the panel was able to distinguish it from a number of other solid-state amplifiers during the tests.
Regards,
Alan
