Double blind testing can work if the listener keeps the criteria for comparison simple, such as "is there a differnce." If you start looking for subjective thoughts such as warm, analytical, too many variables are being injected.
I had an interesting experience a few years ago. I was auditioning a CD player that had upgraded parts. (The usual audio jive, eg. Massive power supplies, increased insulation, transparent resistors). After comparing the Players, I thought the new player had better base weight, clarity etc. Since both players were in the same rack, and had the same remote, there were no clues to which player was operating. It was only after I was returning the demo that I realized by clever wife had swithched the cables, and my preferred player was the unmodified one.
I had an interesting experience a few years ago. I was auditioning a CD player that had upgraded parts. (The usual audio jive, eg. Massive power supplies, increased insulation, transparent resistors). After comparing the Players, I thought the new player had better base weight, clarity etc. Since both players were in the same rack, and had the same remote, there were no clues to which player was operating. It was only after I was returning the demo that I realized by clever wife had swithched the cables, and my preferred player was the unmodified one.