Why the Obsession With Measurements Is Leading to Worse-Sounding Gear
- Thread starter Mike
- Start date
AJ Soundfield
Well-known member
Yep, we agree.
ASR reviews of things like speakers have subjective commentary. I rarely read DAC etc reviews, but regardless, there are subjective impressions in some reviews.
Ah yes, great subjective listening tests on ASR. I remember one "review" with the usual (useless) SINAD measurements spew, which had this on the listening side: "we had company arrive, so no listening tests were performed". That perfectly encapsulates how listening tests register with measurement hobbyists like this - DEAD LAST. What a joke.
Not long ago, a YouTuber (known by many on this site) who reviews equipment, convinced a couple of his golden-eared audiophile buddies to perform a series of blind tests cable to pick the best cable. One of the cables was a no-name wire cord from a hardware store; the other one was a fancy, very expensive, "audiophile" cable that sells for thousands. The audiophiles picked the cheap cable over the audiophile, expensive cable. After the subjective listening embarrassment, the YouTuber blamed "burn in" as the explanation. What a joke.
Puma Cat
Well-known member
Measurements are an important part of any scientific or engineering-based product design and development, but with respect to audio products, they tell part, but not all, of the story.
With respect to a loudspeaker for example, in-room frequency response is important as well as spectral decay. IMHO, spectral decay is particularly important as it will depict how the drivers behave and/or interact with the cabinet design and materials.
But there's aspects of experiencing music when we we listen to recordings that measurements, IMHO, can't convey. Some of these attributes are spatiality, imaging, soundstage, and most notably, IMHO, qualities and distinctions of tonality and timbre.
Haven't seen a measurement yet that accurately conveys why a Guaneri violin sounds different than a Stradivarius when playing the same exact piece of music by the same musician. This is a distinction of instrumental timbre, and I've yet to see a test instrument capture these qualities and attributes.
Oh, the whole, In my view, it's important to remember that our perception of music is a construct of the brain, and not simply a THD or IMD graph on a spectrum analyzer.
Lastly, I'll just end with a Stoic philosophical perspective that I strive to live by: "The tranquility that comes when you stop caring what they say. Or think. Or do. Only what you do." – Marcus Aurelius
Cheers.
With respect to a loudspeaker for example, in-room frequency response is important as well as spectral decay. IMHO, spectral decay is particularly important as it will depict how the drivers behave and/or interact with the cabinet design and materials.
But there's aspects of experiencing music when we we listen to recordings that measurements, IMHO, can't convey. Some of these attributes are spatiality, imaging, soundstage, and most notably, IMHO, qualities and distinctions of tonality and timbre.
Haven't seen a measurement yet that accurately conveys why a Guaneri violin sounds different than a Stradivarius when playing the same exact piece of music by the same musician. This is a distinction of instrumental timbre, and I've yet to see a test instrument capture these qualities and attributes.
Oh, the whole, In my view, it's important to remember that our perception of music is a construct of the brain, and not simply a THD or IMD graph on a spectrum analyzer.
Lastly, I'll just end with a Stoic philosophical perspective that I strive to live by: "The tranquility that comes when you stop caring what they say. Or think. Or do. Only what you do." – Marcus Aurelius
Cheers.
Last edited:
stereogeek
Member
- Joined
- Mar 15, 2018
- Messages
- 472
If the listeners liked the cheap cable better than the expensive cable in the system they were listening to, so what? I got rid of a very highly rated (and expensive) cable and replaced with a mid-grade Cardas cable and couldn't be happier. The correlation between good sound and high cost is overrated.
That's basically equivalent to having TAS put selected measurements into only some reviews (which it does); in neither case does it contribute anything useful
AJ Soundfield
Well-known member
That's your opinion of course. Fact is there are subjective comments on ASR with at least speakers and headphones. Only audiophile care about the rest.
That audiophiles are unable to understand the importance of measurements is reflected in the hundreds of their systems I've heard. Particularly bass.
There are sometimes listening comments on speakers and headphones, not always. And what about the time that Amir did a listening test on a pair of speakers where one of the drivers was disconnected; he didn't even notice it.
Really, measurements are best utilized as a measure of whether or not the manufacturer accomplished design goals; confirming a manufacturer's specs is useful, especially when those specs are not confirmed by independent measurements.
Really, measurements are best utilized as a measure of whether or not the manufacturer accomplished design goals; confirming a manufacturer's specs is useful, especially when those specs are not confirmed by independent measurements.
AJ Soundfield
Well-known member
That would be on Amir and irrelevant to the importance of measurements.
Exactly. "Specs" and measurements, especially independent ones, are two different things.
And of course as I said, it very easy for me to hear audiophiles lack of measurements understanding/abilities when I hear their speakers in enclosed spaces. Tis the penalty they pay.
But it does illustrate the irrelevance of listening tests at ASR, which was my point
AJ Soundfield
Well-known member
No, it purportedly (no link, no specifics) would show Amirs carelessness or inability to hear something that should be obvious. But it does zero to support the anti-measurements position, since a basic measurement would show a disconnected driver as you claim.
IOW, what you may have inadvertently done, is shown the importance of measurements rather than just pure subjective "listening"
That is your interpretation; mine is simply that ASR pays little attention to listening, only measuring, just as many "audiophile" reviews pay little or no attention to measurements. This is a comment on the Website itself, not on the value of measurements in audio reviews or component evaluations.
Actually measurements can help identify swapped channels, a damaged midrange, a damaged tweeter, bass issues (speaker positioning), wrong placement of acoustic panels. My “I only trust my ears” audiophile friend didn’t even realized he had several of those issues in his system.
and add to that point, Manufacture 'specs' can vary often trigger expectation bias for some.
is it fair to say after 3 pages of this discussion, both subjective and objective rationale are important in this hobby of ours.....
sleepysurf
Member
Years ago, using XTZ Room Analyzer, I discovered one of my ML Summits (purchased used) had a woofer wired out of phase. The original owner had apparently messed with it. Pic below (left speaker response in blue, vs. right speaker green line)...
For over a year I had struggled with speaker positioning and bass controls, trying to improve the upper bass/lower midrange. It wasn't until I bought the XTZ Room Analyzer that I discovered the issue. Fortunately, after sharing this pic with ML Tech Support, they immediately diagnosed the problem, and sent me explicit instructions to fix it!
I've since upgraded to ML Expressions, but still use the XTZ Room Analyzer for setup guidance. However, I now prioritize listening as the final setup determinant.
Puma Cat
Well-known member
Have you ever seen Amir publish a statistically valid MSA (Measurement Systems Analysis aka Gage R&R) using JMP or Minitab to demonstrate with statistical rigor that his measurements are Accurate, Precise, Repeatable, and Reproducible? For example, what is the % Contribution of his Measurement System Variance to his Total System Variance?
Because, I haven''t.
Back when I did DOE using JMP to integrate my REL sub into to my 2-Channel system, the first thing I did was qualify my measurements system by doing an MSA.
Question: If Amir hasn't done an MSA and demonstrated with statistical rigor that his measurement system is statistically valid and fit-for-purpose, how could anyone have any statistical confidence in his measurements?
Answer: You CAN"T.
Last edited:
AJ Soundfield
Well-known member
I don't have to any of that irrelevant red herring nonsense. I'm not Amir. He does seem to trigger certain types.
Regardless, his AP and Klippel NFS measurements are repeatable and verifiable, anyone with either could do so. Anyone with basic electro-acoustic knowledge would already know that also. YMMV.
AJ Soundfield
Well-known member
Excellent example Alan. Measure and listen Always better to be multi dimensional than one dimensional limited.
Puma Cat
Well-known member
AJ, All I was asking is if he's done an MSA or not. If not, then he doesn't know with statistical rigor that his measurement system is fit-for-purpose. This just good statistical best practice. Cheers.