Good points Unsound. I also tend to lean that way. It is not that tubes sound "better" due to being "more realistic under adverse conditions", i think that SS typically just shows us how bad MOST of our recording process really is or can be. Since most recording gear is SS though, one can run into a compounded problem if care is not taken in regards to the selection of both the recording and playback components. Kind of like a "double whammy" of reality which might not be pleasant at all. One can either use tubes to "gloss over" or "fill in" the mistakes ( i.e. errors of omission ) or listen to those mistakes with bleeding ears from most "sterile" yet "accurate" sounding SS gear.
In order to get SS to both sound good and measure good, you have to have gear that is VERY fast, wide bandwidth and uses very high quality parts. At least that is my take on the situation. Obviously, others may have different opinions and experiences.
As to using test tones, one can create tones that are as complex as musical patterns but are easily repeatable. Obviously, this is far more convenient in terms of product R & D ( research and development ) or independent testing than trying to rely on other means to obtain a calibrated point of reference.
Having said that, I do agree that while the standardized test tones used ( sine and square waves ) do have their values, they are over-rated and relied upon too heavily by many in this field. Moncrieff demonstrated that components that show similar results with standardized test tones can still sound different and test differently with complex ( i.e. musical ) waveforms.
In this respect, he was the first that i know of to be able to explain ( and literally show via test equipment ) why one digital playback system had better transient response, improved inter-transient silence ( blacker background ), less ringing, etc... over another piece of similar cost and design. Too bad we don't have more "reviewers / audio scientists" around like him. If we did, we might have a few more answers and a few less questions. Sean
>