svart-hvitt
Major Contributor
- Joined
- Aug 31, 2017
- Messages
- 2,375
- Likes
- 1,253
It is not just subjectivity but also objectivity. Often to bring out a point, I may have to change the settings for a measurement. Recent testing of headphone amps shows this where there are myriads of configurations in a headphone amp from gain to input and output voltages. Picking one out of that infinity as a "standard" is impossible and may create an unfair condition for one of the two products under test.
I see your point, absolutely.
On the other hand, we have people discussing artificial intelligence and how AI will revolutionize everything, making cars autonomous etc. I think the AI talk is mostly BS, because things are often a bit more complicated. But I think there’s a middle road somewhere, and I think audio measurements could be more standardized, i.e. subject to some scoring algorithm. In other words: I don’t think algos will drive cars in normal traffic shortly, but I think it may be possible to make an algo to score audio products after measurements. This algo may not be perfect, but I think objectivity is so important I can live with an algo that is not perfect. In absence of an algo (i.e. an objective formula made in advance and not during or after measurements), there is risk that one changes methods during the research process, which may be influenced by biases.
I guess I’m a «standardization» man, probably more so than most.