Why HiFi Gear Measurements Are Misleading (yes ASR talking to you…)

“Why bother measuring things if it doesnt equate with sound quality?”

@kenjit , out of interest I guess? It’s interesting to consider how a component performs, physically, in the real world, independently from human perception which is more important, IMO.

Not sure if I may have overstated my position earlier but I may have. I can admit I believe measured performance has an impact on how I perceive sound quantity. But my main point is I don’t believe it’s a guarantee of how I will perceive sound quality, because I don’t believe we have measurements that comprehensively predict this.  Amir takes the position that yes, we DO have all of those measurements.  And my question is, how can we know that when we don’t know what we don’t know?  My question is ridiculous on its own, as some have pointed out.  I wouldn’t go around questioning everything we know on that basis.  But my own subjective experiences in HiFi have given me enough of a glimpse to firmly believe that there is more that we don’t know.  That we don’t know how to accurately predict how we will each, individually respond to a component with a particular set of measured performance metrics.

To use the car analogy from above, horsepower and torque are valuable measurements but don’t guarantee one will enjoy driving the car.

I’m in favour of JA’s approach to measurements in Stereophile, where subjective listening is the focus. But it’s fantastic that the measurements are there just to see if SOME correlation with the subjective experience can be gleaned. Why? I find it interesting.

People are perfectly free to disagree with me!  It does not bother me one bit that ASR exists and I’m happy if they keep banging the objectivity drum.  Some say it unfairly harms certain brands but the subjective review sites can balance this.  Makes our hobby all the more vibrant, provided we can keep the personal attacks out of the equation.

But my own subjective experiences in HiFi have given me enough of a glimpse to firmly believe that there is more that we don’t know.

Arent you just contradicting yourself? You just said we dont know what we dont know. So perhaps we have all possible measurements or perhaps we dont. You cant then claim that your subjective experience gives you a firm belief that theres more that we are not able to measure. Your subjective experience might also just be imagined. Perhaps that $5000 cable is useless after all?

The other thing is that jitter is not one number as is often talked about. Above you see multiple jitter sources at different frequencies and levels. This makes FFT analysis far superior to any time domain jitter measurement that spites out either a number or even a distribution.

That is true. Jitter could be random, periodic, ISI, DCD and I am sure there are more. The tone in your example measure periodic jitter which can be done in time domain as well. For a pure sinewave, you could use frequency domain, but with non-sinusoidal waveform (which is more akin to music), it’s better to use time domain.

There are transient events that can only be capture in time domain such as overshoot or undershoot or ringing or more .,.. It is fine if you only use frequency domain but it won’t be a complete test.

There are transient events that can only be capture in time domain such as overshoot or undershoot or ringing or more .,.. It is fine if you only use frequency domain but it won’t be a complete test.

Can you give examples of these transient non periodic events that would occur during playback?

but with non-sinusoidal waveform (which is more akin to music), it’s better to use time domain.

We use non sinusoidal waveforms all the time, though they may be made up of a multitude of sine waves. I already talked about transient testing.

@kenjit , fair point. But I don’t think we can treat all subjective experiences equally. We can be more certain of some subjective experiences than others. If I were to see a person in front of me I wouldn’t question if that person was there (an extreme example). In my example above, there is little chance I was imagining the longer USB cable sounding better than the shorter one when I was biased to thinking the opposite. 25 years ago, I also knew that when I heard the meteors sizzling, I knew that I really heard them. Even though when I immediately checked the internet, all researchers had said this was in my head. I’ve learned when to trust my senses. And also, just as importantly, when not to!