Maintaining it is one thing - proving it is another. In the absence of objective proof, all that remains is opinion. Opinion doesn't help people much really, because some people have the opinion that X is great and some have the opinion that X is rubbish. If the differences exist objectively then they can be proven to exist. So far nobody has managed to prove that anyone can reliably differentiate between two decent modern SS amps working within their design parameters. OTOH, it has been demonstrated numerous times that people cannot differentiate between such amps. So the burden of proof is on those who claim that differences exist - if they do, and people can hear them, then they can be measured and also shown up in blind ABX tests. So far, this hasn't been the case.
The real point of showing the graph is that these differences can be measured. And the measurements confirm what the ears hear, or vice versa. This is not opinion or whatever - this is objective evidence that Audyssey makes a difference and one would expect anyone listening to be able to hear the difference. If someone can do the same for amplifiers, I have yet to see it.
It's a pity I don't have a graph showing the untreated room with Audyssey on and off, because then the difference would be very dramatic.
Note how little correction XT32 applies above 1,000Hz. This is one of the big differences between XT and XT32 and accounts for a good part of the latter's superiority over the former. Again, one would probably expect to see a bigger difference above 1kHz if my room was untreated, but the relative difference between XT and XT32 would be maintained.
Of arguably more importance than the FR graph is the waterfall. Ringing is more problematic to the human ear than frequency variations of +/-5dB.
Electronic EQ isn't totally brilliant at controlling modal ringing - room treatments (to some extent down low) and speaker and sub placement optimisation are more helpful.
Its all very well for people to say "well music isn't graphs" or "we don't listen to measurements" and so on. Absolutely. But everything that can be heard can be measured. Of course, not everything that can be measured can be heard! Objective measurements are the foundation of good sound. We follow the measurements with listening tests of course - but these will invariably confirm what the measurements also show. Our ears are great listening instruments but very poor measuring instruments, which is why the basic 'tuning' of a room can only be done with measuring gear.
Sound reproduction isn't 'magic'. It's science.
Edited by kbarnes701 - 10/20/13 at 1:06pm