While the 'subjectivists' will groan, even though I tend to choose audio equipment based on what I think of the sound of it, limitations like expectation bias are a genuine issue. I also agree that comparing a modified amp to a fixed reference in a system context may not be useful. Using the example given, what if the recording of the cello really is lacking in 'body'? If an amp adds more 'body', this does not necessarily make it better- it may really be slightly worse but sound more like a real cello (or the listeners memories of real cellos) in that room, with that system, than it did before. But there is another side to the coin; experiential evidence has helped push developments in audio design and measurement techniques which have resulted in what many would regard as better sound and better engineering. Also, I have found that expectation bias is not quite as powerful and pervasive as is sometimes suggested. Just anecdotal I know, but I have listened to comparisons between pieces of hifi gear that have not gone at all as I expected. I have been surprised. Also, I have heard equipment that sounded so-so in one context sound excellent in another. So, while I respect those who are attempting to push the boundaries like Ric, I do have concerns about repeatability and consistancy. 'Context' is a really big deal in this context! I do worry that, even with the best intentions, it is easy to become involved with modifications and experiments to the extent that one is emotionally entangled with outcomes, to the potential detriment of accurate reporting. I'm not saying this is necessarily so, but it is a risk. There is a danger that if one is too sceptical, genuine and useful improvements might be ignored. Then again, 'subjectivists' need to be cautious about making pronouncements based on necessarily limited personal experiences as if (ironically?) they were objective.
