Johnny2Bad
Well-Known Member
With regard to the OP's post, yes, there are some issues with subjective listening tests, but none the less, that is how we expect our equipment to do the task we assign it. So it cannot be said that subjective listening is invalid; it is in essence the only valid assessment.
There are considerable problems with blind (and double-blind) listening tests as well. Ignoring for the moment that I rarely see proper double-blind tests actually performed (making the results invalid from the pure objectivist point of view right from the start, yet conveniently ignored in most cases), there is the fundamental issue that untrained listeners prefer low-fi to mid-fi and mid-fi to hi-fi (as proven as far back as the 1940's)(1) and that those same listeners, if broken into three groups, where one group is exposed to HiFi for a period of time, one group is exposed to Mid-Fi for the same period of time, and the third (control group) is left alone with no listening exposure, then the group exposed to HiFi suddenly prefers the HiFi, the second group now prefers the Mid-Fi and the control group remains preferring the LoFi (experiment done in the 1950's to expand on the conclusions of the first). Conclusion: people prefer what they are used to hearing, but can be trained to listen for quality reproduction.(2)
It amazes me we still are dragging this dead cat around 65 years later and that some insist they can "prove" the cat is still alive.
1: Howard A Chinn and Phillip Eisenberg, CBS
System was identical, studio-grade components (0.3% THD) but frequency-response limited as follows:
40~10,000 Hz
80~7,000 Hz
180~4,000 Hz
2: E. Kirk, Ohio State University
Six weeks duration between first experiment (which duplicated Chinn/Eisenberg and had the same result) and second phase.
There are considerable problems with blind (and double-blind) listening tests as well. Ignoring for the moment that I rarely see proper double-blind tests actually performed (making the results invalid from the pure objectivist point of view right from the start, yet conveniently ignored in most cases), there is the fundamental issue that untrained listeners prefer low-fi to mid-fi and mid-fi to hi-fi (as proven as far back as the 1940's)(1) and that those same listeners, if broken into three groups, where one group is exposed to HiFi for a period of time, one group is exposed to Mid-Fi for the same period of time, and the third (control group) is left alone with no listening exposure, then the group exposed to HiFi suddenly prefers the HiFi, the second group now prefers the Mid-Fi and the control group remains preferring the LoFi (experiment done in the 1950's to expand on the conclusions of the first). Conclusion: people prefer what they are used to hearing, but can be trained to listen for quality reproduction.(2)
It amazes me we still are dragging this dead cat around 65 years later and that some insist they can "prove" the cat is still alive.
1: Howard A Chinn and Phillip Eisenberg, CBS
System was identical, studio-grade components (0.3% THD) but frequency-response limited as follows:
40~10,000 Hz
80~7,000 Hz
180~4,000 Hz
2: E. Kirk, Ohio State University
Six weeks duration between first experiment (which duplicated Chinn/Eisenberg and had the same result) and second phase.
