An interesting thing about this thread is the tension between objectivism and subjetivism. Sometimes we end up in our little camps and it becomes a trench war. I have to say that I don't like it that way because many lessons are lost in warfare! Valid points aren't acknowledged and the brain power better used in learning is turned into proving one's point, being right and fortifying whatever view you came to the battle with, at all cost!
Personally I describe myself as a moderate objectivist. In other words, I believe in the scientific method as the best path to reveal the truth, that differences can be revealed with measurements and that as Paul Klipsch succinctly pointed out, if two things sound the same and measure differently, you're measuring the wrong thing!
However, we also live in the real world where it's probably not worth trying to quantify and measure everything. Where if we find that one cap sounds better despite measuring "close enough," we might not be on a huge mission to find that measurement that tells us why. We probably want to test it under controlled conditions if we think it's really important to be sure, but it's probably not worth it just to prove ourselves right in an internet debate where the stakes are so low!
When it comes to a driver that rings like mad, this is where we want to be measuring to see what we are getting. You can sit back with some music and say "I don't notice a problem!" Trouble is, it might be a reflection on the music you are using, or that you are simply not recognising a problem that is there. Perhaps if it were fixed and you could switch instantly from fixed to unfixed, you would then say "WOW I had no idea it was that bad!" Then you can't go back to unfixed - your brain gets a lock on the problem and now you hear it all the time - this is what I call the man in the moon effect. One seen, you can't "unsee." Try as you might to tell youself, there is no man in the moon, you will always see it. There was one really big blind test in the industry, I believe with thousands who could not hear a problem with a particular codec. One engineer thought it was nonsense, that he could hear it plainly. He pointed it out to those in the test, suddenly everyone could hear it.
I think most of the time there is a problem with subjective comparisons that are done. I was involved in an informal blind test of sorts, with DACs. We were comparing 3 of them. Lampizator, Lenehan PDX (an Aussie local DAC that you might call "mid priced") and a Playback designs unit with its own inbuilt DAC. I believe they were worth about 7k, 3k and 12k or something like that. We had a room full of people listening to a very resolving system, instantly switching from one to the other. Host would say which DAC was being played, then we'd listen. Then we had some runs of trying to pick them blind. I noticed two obvious differences - they each had a different gain level, and a diffferent perceived tonal balance. Obvious enough that I could tell them apart blind without much difficulty. One thing I noticed was that the group, being more towards subjectivist, were all listening subjectively for the overall "feel" of the DACs. When I had the differences figured out, listening in a more deliberate and focused way, I announced "right, A was PDX, B was Playback designs, C was Lampizator." The others who had been listening in a more general way piped up "oh no no, that's not right. I think ...."
Guess who was right? Yup, you guessed it. Not suprising since I had focused carefully on the more obvious cues - a very artificial way of listening, but far more reliable if you need to find something. To be a more serious test, we really needed to have them at least matched in level. Perhaps even corrected in the digital domain for any frequency response shifts. Otherwise we might be comparing tone controls. This is why you have to treat internet opinions on these things with some caution, if you want to know the truth. The degree of rigour to really weed out anything misleading, is usually well beyond what people are willing to put in.
Now if we are going to compare crossover parts then I think what we really want is a setup where you can sit in the listening position and press a switch that instantly changes over from one to the other with no break in the music. This probably means using a relay in place of a manual switch. You power the relay with a little plugpack and switch the power into the plugpack with a remote GPO switch. This would also allow comparisons of cables and perhaps just about anything except a source.
