Skeptic seems to be the most even-handed here. It is not hard to make a cable a cd player or anything in audio SOUND different. It behooves the manufacturer to try in fact to make their stuff sound different to stand out from the rest. And if it measures differently in the audible spectrum like for instance many Dac chips alone do then if they are not picked up in your standard test then maybe you should look at the standard test rather than drawing false claims.

Anyone can do very reliable tests and get very reliable results - but if it was an invalid test then you have high reliability at getting wrong results and can steer Joe scientist WAY WAY off the wrong track. There was a study done of more than 200,000 people by great scientists that Red Meat is bad for you. Very high reliability indeed. Except the people eating the red meat also ate their meat with baked potatos and white rice. The scientists were out to show that red meat was bad for you gave the people these plate of food and of course the people were getting higher cholesteral, diabetic symptoms etc. But oops they didn't for one lone second think that maybe it was the baked potato and the white rice that was causing the problem and the red meat had zippo to do with the problems - which as it turns out is in fact the case.

You have to look speicifically at what it is you are testing. The DBT is a magnificent tool for a lot of things - it is not however a panacea of "correctness" in human subject testing. Ask any psych proffessor or people doing brain research testing the reaction of human subjects and you will not get this nonsense of support for the null hypothesis with the parameters of this kind of test. You are taking the subject OUT of the normal listening environment PERIOD. You are testing people outside of the normal environment which automatically decreases the validity of said test.

Look at the very juxtoposition of the timing of A/B listening. You need it short to meet the demand imposed of the test of memory retention - but last i checked people do not listen to 12 seconds of a song and switch. Maybe some do which would make the test more valid for them - but if you are the listen to a whole album with a glass of a fine port then putting you in a seat and saying now which is different (even the wording has to be examined as not to lead).

None of this is an apology of sighted listening which of course is full of bias - but that a dbt in testing human subjects has an invalid element that has yet to be corrected.

Hi-Fi Choice oddly which is not a strict "Testing" environment works to a degree better because it's not a Testing environment but an evaluating environment - and that is what every person golden ear or not can do and does do when buying componants. Since this is a normal "Real-life" event then the approach is far more useul to the average shopper like you or I. They remove all sight biases(Price, name brand recognition, and of course levels are matched so no sound bias due to volume). Differences are expected so there is no "test" interation or "stressor" placed on the listener and the right hemisphere is left to do what it does when listening to music in a normal environment.

Then you have several people over several hours fill out cards of what they liked or disliked etc. This isn't perfect but it does do what it is supposed to do - remove the bias...that is why a DBT was used and their method does do that...whether it's acceptable to the AES - well they are not scientists obviously because they have not looked into human testing - so who cares.

Ohh and if you want to tell use a line level headphone amp connect and level match the cd players and have someone else switch(and of course the same cd in both machines playing the sam song). Simple - easy - no room interactions with headphones --- you may or may not be surprised at your results - listen for what sounds better not what sounds different - the way you think about it will change the way you listen - or theoretically could.

There is a reason this is in so much debate - The dbt supporters dislike biased sighted listening which is totally understandable and totally correct - on the other hand the other side looks at the supreme confidence in the method the DBT lover touts and it's got a lot of problems - then you have camps set-up with un-yielding wavering.

Bottom line is the manufacturer can deliberately change the sound of any componant - whether it's better, worse the same cannot be assessed in different rooms, with different gear and different listeners attached. If it measures differently within the audible spectrum then it's different. Whether you can't hear it in these tests says nothing about whether you can hear it in a non testing environment over a longer period of time or just not in the testing environment.

But Mr. PC Tower you will not convince any of these people to look into these external arguments about the brain and the way it works(possibly Skeptic) and why it is not wholly compatible to these tests - instead you will get an gross analogies to ESP, UFOs, and Aliens.