advertisement


FAO the pro-measurement crowd: ASR to start measuring speaker performance

It seems to me that equipment measurements appear to divide people who don't have experience of making and interpreting measurements (and most people naturally don't). A typical response is to generalize and polarise to either "measurements mean nothing" or "measurements mean everything."

In reality measurements mean something - usually something useful. But you do have to be prepared to work out exactly what that is rather than just generalize.
Well said. It’s surprising how many “audiophiles” are ignorant of the extensive research that has been done on what makes a speaker sound “good” and therefore which parameters are important. That’s not say that everything is known on the subject, but Floyd Toole’s research has advanced things to a significant degree. Anyone doubting that should take a listen to a pair of Revel Salon 2 speakers which were designed to satisfy the criteria that the research identified as being most important.
 
"Lies, Damn Lies and Statistics".

What makes statistics worse than lies and even Damn Lies?

Statistics can be cherry picked, massaged, presented or mis-represented and yet are still given a veneer of absolute objectivity and authority - it's that specific combination that's deadly.

That can be the case with measurements too.

Measurements are of course very important. I don't like reviews without measurements. Good products are developed with engineering and measurements. But I know good engineers can still disagree about what the most useful ones will be.

I have my reservations about the DAC measurements on ASR (see elsewhere). Given speakers are far more complex to measure I don't hold my breath.

The measurements are likely to be useful but will need thoughtful interpretation.

I can see it benefiting salespeople.
 
...A good balance to the sneery subjectivists over on SBAF though.
As a long time member there i see it more as a balance between subjective and objective, some people may sway more to one side or the other. There are extensive measurements there if you are so inclined. Something for everyone, now there is a culture there and you either fit in or not, no problem.
 
Fair enough. I’ve enjoyed reading some of it but they are a bit sneery objective or subjective. And the worship of the god Torq is a bit tedious. Anyway, best not break the AUP here.

Sir Tony of Pinkness has it perfect of course.
 
"Lies, Damn Lies and Statistics".

What makes statistics worse than lies and even Damn Lies?

Statistics can be cherry picked, massaged, presented or mis-represented and yet are still given a veneer of absolute objectivity and authority - it's that specific combination that's deadly.

That can be the case with measurements too.

Measurements are of course very important. I don't like reviews without measurements. Good products are developed with engineering and measurements. But I know good engineers can still disagree about what the most useful ones will be.

I have my reservations about the DAC measurements on ASR (see elsewhere). Given speakers are far more complex to measure I don't hold my breath.

The measurements are likely to be useful but will need thoughtful interpretation.

I can see it benefiting salespeople.

How does more information being made available to hifi buyers benefit salesmen and not us?
 
I wrote it would be useful, with thoughtful interpretation.

Edit: it will be useful to salespeople even without thoughtful interpretation. Hope that explains better what I meant.
 
Last edited:
I take my hat off to anyone willing to put in the work involved in measuring speakers. The sheer logistics and costs are enough to put most of us off.

Amir seems to be capable of developing a repeatable, useful methodology, so the effort may yield worthwhile insight. His investment in test kit and time so far has produced an increasingly credible body of insight into the real performance of audiophile electronics. As other here have said, using that information to make nuanced, realistic intepretations of what the data reveals is the challenge. Some of the more zealous, vocal contributors on ASR unfortunately seem a little less nuanced in their thinking and the SINAD hit parade for DAC's is ultimately a very blunt tool for summarising Amir's work.
 
I hope he also calibrates each human ear’s sensitivity and frequency response, how they subjectively perceive sounds, or it’s half a test in the real world.

The problem with these “one true way” methodologies is that they regard taste as irrelevant, like the Borg...
 
A classic strawman argument. No-one denies that personal taste is a valid reason for buying something, just that it's not a useful way to review equipment because tastes differ, and therefore we need objective standards to compare them against.
 
A classic strawman argument. No-one denies that personal taste is a valid reason for buying something, just that it's not a useful way to review equipment because tastes differ, and therefore we need objective standards to compare them against.

There is a lot of poor measuring gear that gets well reviewed, and well liked by many.

We don’t need a set of objective measurements to say whether they are “right” or “wrong”.

I do think including measurements is a good thing, like the Stereophile sections, but it shouldn’t be the whole review.
 
I hope he also calibrates each human ear’s sensitivity and frequency response, how they subjectively perceive sounds, or it’s half a test in the real world.

The problem with these “one true way” methodologies is that they regard taste as irrelevant, like the Borg...

It's not about taste - that's down to us as listeners.

It's about providing a marketing-free view of what a given bit of kit actually is capable of. Other sites offer either fully subjective reviews with no actual useful information, or mix up subjective with a few measurements which generally tell a far less emotive story. We are all ultimately limited to what our own ears give us, but having a technical basis for a buying decision is useful for some of us. Learning what to really expect from different technologies (R2R Dacs for instance, or switching amplifiers) is very useful when it comes to narrowing down your buying choices.

I would never assume a sophisticated, informed buyer of audio gear would look at a set of impartial measurements and buy on the basis of a few charts, but the information is surely useful in eliminating real stinkers. And cutting through the marketing messages is always helpful.
 
I am very much in the "measurements tell you everything" camp; after all, our ears work by measuring air pressure fluctuations, and they are not all that good at it in absolute terms.
It depends on what you include in "everything". IIUC, the individual ear-brain system is far too adaptive to be addressed by most measurements. So individual personal preference remains well outside the zone (although there are some statistical measurements referred to later on in this thread that do connect population preference to technology).

So underlying the issue here is, for me anyway, that personal preference remains outside the scope of ASR's measurements.
For something like a DAC, measurements are relatively easy to interpret, mainly because the technology has been more or less perfected, for all practical (audio) purposes.
Yes, largely. Measurements mostly evaluate defects against some standard of perfection (e.g. flat frequency response, zero non-linear distortion, etc.). However if someone prefers the characteristic frequency response of zero-order hold reconstruction (e.g. in a NOS DAC) then that deviation from flat is hardly a defect to that person.
For speakers, which still represent much more of a compromise, interpretation/understanding will be a lot more important. One of the best things about the fact that ASR are going to start looking at speakers is the huge opportunity that it will provide for novices like me to learn.
It's a big step in a much more difficult area than DACs and other electronics. It will be interesting to observe.
 
It depends on what you include in "everything". IIUC, the individual ear-brain system is far too adaptive to be addressed by most measurements. So individual personal preference remains well outside the zone (although there are some statistical measurements referred to later on in this thread that do connect population preference to technology).

So underlying the issue here is, for me anyway, that personal preference remains outside the scope of ASR's measurements. Yes, largely. Measurements mostly evaluate defects against some standard of perfection (e.g. flat frequency response, zero non-linear distortion, etc.). However if someone prefers the characteristic frequency response of zero-order hold reconstruction (e.g. in a NOS DAC) then that deviation from flat is hardly a defect to that person. It's a big step in a much more difficult area than DACs and other electronics. It will be interesting to observe.

By "everything" I meant "everything related to the performance of the equipment". Of course, personal preferences are another matter entirely, and far beyond the scope of what ASR claims to address.

But in any case, I agree that it'll be very interesting to watch how this all pans out.
 
Ability to reproduce a square wave always seems a measurement never used these days....probably because few modern speakers can make a decent stab at it. Of the golden oldies Quad ESLs and Ureis could apparently manage the trick....so a definite correlation with quality.
Having just bought a pair of Quad 2805s, with their delay rings, I would be interested to see how this system copes with that.
 
I'm not interested in a reviewers opinion of how they think something sounds with a pair of ears I dont own and in a room I'll never visit. At best they can provide broad strokes.

A set of measurements however allow me to assess if a piece of kit is essentially blameless. For electronics that's all I need, for speakers I need more to figure out if they'll work in my space and thus polar, fr and power plot's become important.
 
The best single source of information on speakers in rooms is, of course, Floyd Toole’s book. Every serious Audiophile ought to have a copy. The book explains which measurements matter and why.

Summaries are available on the web. See:

https://www.edn.com/loudspeakers-ob...ing-the-essential-properties-of-loudspeakers/

https://www.sausalitoaudio.com/wp-content/uploads/2018/07/Interpreting-Spinorama-Charts.pdf


An anonymous blogger, reputedly an Harmon employee, has posted spin-o-rama data here:

https://speakerdata2034.blogspot.com/2019/03/spinorama-data-kef.html
 
I'm not interested in a reviewers opinion of how they think something sounds with a pair of ears I dont own and in a room I'll never visit. At best they can provide broad strokes.

A set of measurements however allow me to assess if a piece of kit is essentially blameless. For electronics that's all I need, for speakers I need more to figure out if they'll work in my space and thus polar, fr and power plot's become important.


Well said!
 
Toole’s research forms the basis for the AES standard for testing and measuring loudspeakers. It’s a shame that more manufacturers haven’t adopted it. But there’s probably a reason for that . . . .
 


advertisement


Back
Top