Test Equipment vs The Ear

The problem is that human hearing is not very good - your dog is much better at it. The good news is that for most parts of the audio chain it is now possible to design and manufacture units that are better than what humans can reliably discern. The differences that are still there are inaudible, but sometimes still measurable (and different is not necessarily better). The second problem with listening is that conditions have to be carefully controled. Levels have to be matched within 0.2 dB, because the brain interprets louder as better. This can only be achieved with a proper volt meter. The third problem is that comparison has to be near instantaneous because the brain cannot remember sound for very long. Finally, there is that old devil expectation bias (requiring double blind testing).

Show me the research in peer reviewed journals that show that what I said is an old wives’ tale. Here is a famous one from me: http://www.keith-snook.info/wireless-world-magazine/Wireless-World-1978/Valves%20versus%20Transistor...

Sure you can learn to listen. Research has shown that with humans the threshold for hearing level differences (and deviations from the flat frequency response) is about 0.2 dB. If it less than that, you cannot hear it so it does not matter (but you can measure it). On the electronics side, it is relatively easy to stay below this threshold of 0.2 dB, and if e.g an amplifier fails this requirement it is not what Peter Walker called a straight wire with gain (and hence it is not good enough). But with speakers this 0.2 dB is unfortunately an impossible target for now. So speakers will sound quite different from each other, unfortunately. Hence also, it is worth spending relatively a lot on speakers.