Here is the interesting thing about listening to a system and then commenting on the quality of different tested components. As you all know I am comparing Cast duelunds against hovland music caps. On my big system its easy, because you can hear a fly landing on the speaker box or even a half deaf audiophile coming through the front door, the smallest change via even a piece of wire is audible. But on the second system consisting of a great sounding set of Quad 57s topped with Linauem omnidirectional tweeters, leak stereo 20 Rebuilt with quality hand picked parts and also armed with great valves, stepped resistor volume control, plus a very good sounding expensive professional pioneer DVD player. Now here is the problem when I change parts cables caps resistors it's very much harder to pick exactly what is really going on, there is a equalization of the playing field, differences are somewhat nullified. The differences are there but can we use this information in a real tangible way. So the question needs to be asked, is any of the results even valid, Could anyone build say, incredible sounding amplifiers or preamps if they did not have a great sounding setup, do you need a real sounding system to actually hear the truth. Well all I can say (with this Quad setup) is that i could build a great sounding amplifier to suit the quad 57s. But I could not make with certainty a universal, magical sounding world class amplifier. If we where to do some blind testing of parts, via this system can we really predict anything representative of the truth. Well my answer is not much chance. Anyway I am still listening to these caps, I just need people to decide for themselves how much real world value should be factored into the result. PS, one thing the Quad 57 speakers and leak setup does have, tremendous musical engagement.
I would not have really considered this years ago, yet I have also come to this view myself. A lot of testing (blind or otherwise) might be considered naïve where we run some gear against each other and expect to divide up the winners and losers. If you have done much experimentation with systems over time you may have experience a false result, thought yourself something was a problem (or inferior) when you hadn’t really identified the cause. One simple case was some strange tonal resonances occasionally out of decca Kelly tweeters, no it wasn’t a design fault, it was operator error, they needed to be properly mounted, when they were properly mounted the problem completely disappeared. Some owners might have sold them for that fault. The other common problem is speaker placement, Andy and I have heard first hand how much better for example quads sound pulled out from the walls, and how horribly poor a new pair sounded when a dealer had them right up against walls. You can quite imagine some one concluding quads were terribly poor speakers against others based on faulty implimentation.
Where there a major deficits, and equipment has faults we may not need a system where you can hear a fly landing on the speaker to hear the differences. Yet say we have some speakers that aren’t really tuned well, box issues and room effects, how reliably can we make conclusions about amps feeding those speakers. One amp (amp A) may set off issues with speaker response that the other doesn’t, we falsely conclude that amp A is inferior to AMP B other, when if we listened to both on a speaker that was well tuned and had minimal room interactions we could imagine reaching the opposite conclusions, amp A is superior.