Does measurements really matter?

This video by Thomas interview Harley is one the best way to understanding the topic of measurement  




Measurements matter in correlating what we think we hear with what we do hear with what we desire to hear.

Psycho-acoustics is incomplete as is the science of understanding human hearing is also incomplete. How that all ties to ’not fully understood psychology’ is also an incomplete subject.

I strongly believe in the use of both measurement and listening tests.

Eg, that we can erase and do erase..the noise and constant of interference in our human eyes. We editorialize as a form of discernment. That ’editorialization’... is a subconsciously controlled variable.

We do this In each and every single fraction of a second of using our eyes. (veinous structures in our eyes are easily visible but are removed from our ’view’ through our eyes, by the eye/brain mechanism)

Or in the sensations in our bodies.

Or with our hearing, where we editorialize and remove noise out of the noise floor and remove noise that protrudes above that noise floor. We have subconsciously variable complex filters for our hearing, on an individual capacity level, via learning and inherent individualized capacities - as a pairing.

Hold your eyes completely still and open for about maybe a minute, your eyes will ’grey out’ all that you see. This is because the mechanism of eye and brain is designed to remove/separate the noise floor from the signal. Where your eye says that the things that do not move, are background noise and part of the eye - and not part of what you are trying to see.

There is no ’cart-blanche’ of saying, ’don’t trust your hearing at all" and trust ONLY audio measurement. Such a pronouncement is a fools game, the kind of hard decision made by the illiterate.

People do that, when they reach their cognitive limits, they go all hard and black & white, on/off, yes/no.

It’s a human animal thing, it’s about self protection in the face of dangerous unknowns. The mind, not knowing the direction or truth of a given matter.... is overruled by the body and goes to ground. It’s also a way of finding the edges of people’s cognitive package. You know their limits when you see it as they advertise it loudly - all over the place.

This is my cognitive package’s edge and limits! I will go no further! You are all wrong!

Don’t trust your hearing! charlatans! Fooling yourselves!

(I have my limits as well, we all have them, I’m not somehow exempting myself - I’m not claiming any superiority, here)

The deal is that human hearing is a terrific, highly capable measurement package. Where understanding how it works and what it’s limits are, is key.

Importantly, hearing is as variable in individuals as IQ and cognitive capacities are, and is connected to cognitive capacities in the given individual. Thus, it cannot be nailed down into being some sort of hard number tha is relevant to all scenarios for all people. Not possible.

They matter a lot to some people and matter not a jot to others (like me) - it's all good.... 

Well, if we ever hope to hear recordings as their creators devised them, yes, of course they do matter.

Otherwise we risk forever going around in the audio circle of confusion that Sean Olive and Floyd Toole have spoken about.


Whatever you or I decide to sonically do with that recording afterwards is entirely up to us, our ears, our system and our room.

An excellent video. Harley does a great job discussing the place of measurements in recording, design, and playback. I am a trained scientist and worked as a electro / mechanical design engineer. I quickly realized principles and specs would get you in the ball park… but quickly needed to be abandoned and experiment with materials to get the preferred outcome. Just like I have heard high end audio designers say… these resisters sound like this, these capacitors sound warmer…etc… and they go on to craft a natural realistic sound and leave the specs take care of themselves. I guess, that was one of the first lessons I learned after buying some amp with incredible peak watts per channel and sounded terrible. A real basic principle of high audio… glance at the specs…. Then listen, listen, listen.


If you're going to listen to single driver slim tall boxes they don't matter at all. 

Watch Danny Richie's GR Research You Tube videos to see what and how he measures loudspeakers. Very similar to what John Atkinson does.


. A real basic principle of high audio… glance at the specs…. Then listen, listen, listen.

Excellent summation. 



Teo_Audio- WOW, that was a GREAT post.  One of the best I've ever read.  In such a small word count you explain the issue of hearing vs measurement in a nice nutshell.  I love the simplicity of your description: an "incomplete science".  

I have been demoing high end audio since 1975.  I've heard just about every weird anomaly and made about every dumb error you could make in system set up or design.  First consumer, then live sound reinforcement, now pro level recording studios.  I have heard very smart people say very dumb things, such as the composer who can write a movie score but insists his speakers are buzzing and defective when it's something in his room; like the home audiophile who is convinced his speakers suck when it's 100% room problems; like the producer who thinks he makes better decisions at 120dB SPL or the mixer who cannot hear the obvious power compression in his old passives after working all day.   I've heard great speakers sound absolutely awful in one room, brilliant in another room.  As you say, the mind plays tricks and audio/acoustics is not a simple black and white science.     


I am a trained scientist and worked as a electro / mechanical design engineer.




A real basic principle of high audio… glance at the specs…. Then listen, listen, listen.

As a trained scientist you would be acutely aware of the need to control for confounding variables in coming to your conclusions. When you "listen, listen, listen" how did you control for known variables such as sighted bias, in reaching your conclusions?



Brad--probably off-topic, but I've been meaning to listen to 40A and 50ASLT for a long time. I bet that within domestic reason, they sound similar (and great) from one room to another (as well as similar to each other except for bass and SPL). Without asking you to shill your product, can you comment?

Timely: In his latest GR Research YouTube video (posted today, Wednesday 3-23-22), Danny Richie examines the Magnepan Mini send him by a customer. In the video you see not only how Richie evaluates a loudspeaker---and designs a cure for it's ills, if any---but also a tutorial on comb-filtering, a design challenge inherent in all planar loudspeakers. 

Finding the video is real easy. Go onto YouTube and do a search for GR Research. The latest video will be first in line. A free education in loudspeaker design! 

Measurements and specs mattered when I started in 1982.  In 2022 I'm way less concerned, but they certainly have a place and serve a purpose.  I suspect as a deterrent, they help keep manufacturers honest.  

hedwigs: I sure can answer that question, but here is probably not the place?   Maybe email me at brad at lonemountainaudio dot com?

The short answer, appropriate for this forum, is that the major difference is just the woofer; the mid and tweeter are the same but a lower power version in the 40s (vs 50s on up).  So the major difference will be clarity, definition and roll off point of the low end (below 380Hz), dynamic range of the entire system and maximum SPL level.



I don’t think specs make manufacturers honest in any way shape or form. I think they are viewed as a marketing tool by most. Manufacturers cheat and manipulate specs as a specific effort because they know if a spec is "bad", it affects sales. They know specs are perceived as a value of product performance, not this impartial number that helps you evaluate it. Frequency response for example, a total free for all: some don’t even list the plus or minus so you can say anything you want. Efficiency, you can write this spec in many different ways, this 1W/1M spec can be found with different methods of measurement (what frequency its taken at etc). The more you know about specs, the easier they are to manipulate.

Those with a lot of experience in audio know that a few specs rarely relate to actual real world sonic performance. What good is a wide frequency response and high efficiency if distortion is extremely high? There are so many factors (such as room or source or?) that affect the sound quality that trying to assess a speaker or a amp based on specs is like deciding what car is safe based on its skid pad numbers or its braking distance. It MAY be important but there’s a lot to consider beyond those specs.



The speakers they're talking about are what I've always known as "Girl with a Banjo" speakers. In room might show down to 30hz but at what SPL? SPECs by manufacturers don't tell you much measurements by a third party can.

A speaker needs to sound good that's what home loudspeakers are designed to do but to do so you need basic measurements all transducers and electrical parts used in loudspeakers are carefully designed and measured. So if the designer voices by ear they are still using parts that are all carefully designed and measured even if end results are not. Some designers are so skilled in that aspect that measurements are only used to find defects in assembly. Other companies use a measurement-based design philosophy that takes precedence over voicing. Both are viable after all we are creating devices that are used to play art and art is itself subjective. Loudspeaker design is a lot like cooking some can cook better than others, some can cook well if they carefully follow a recipe and carefully measure all, some can make most any ingredients taste great and can just wing it.

The number of boutique consumer speaker manufacturers using measurement as part of a strict incoming parts QC policy to secure zero variance in all parts is extremely low- perhaps non-existent.

If you do measure, as the engineering driven companies tend to do, OEM parts typically vary quite a bit unit to unit. SO you may do batch testing (test a few from each batch that arrives).  If you are super strict, you are throwing a lot of parts away.  Achieving perfect parts that perform exactly the same all the time and are perfectly consistent over hundreds or thousands is not a realistic target.   

A 1/2 dB variance in sensitivity applied to the entire midrange or the entire tweeter response is very audible to a listener, even if that listener cannot identify a 1/2 boost or cut somewhere along the response curve.  1/2 across the entire band is different which is why parts sensitivity is such a big deal.  Combine that with the idea that a passive crossover cannot be precision adjusted to account for this part variance.   Then one more problem, I don’t think ANYONE is making parts plus or minus 1/2 dB in sensitivity specs over the life of the part. The only way to deliver that is two choices: through anything outside a super tight QC window away or develop precision manufacturing so good you get perfect matching part to part.  I don't think that level of precision in part manufacturing exists.  So variance from part to part and then complete speaker to speaker is a part of the business. That’s why they have a overall plus or minus spec so they can absorb theses variances and still meet an overall claimed spec.

How many in the field, dealers or end users, ever A/B the same speaker?  I would say that almost never happens.  It might surprise you if you did.


Many measurements should matter to anyone putting together an audio system. Many issues arise do to incompatibilities between components.

What comprises a good outcome is mostly subjective.

In a conversation w/Steve Guttenberg Roy Delgado the Klipsch Heresy designer had a conversation w/Paul Klipsch. He told PK that the measurements were fine but it didn't sound correct. Paul then told Roy that there is a 5% Fudge Factor. He told Roy do you want a speaker that a computer likes or do you want a speaker the human ear likes.

@luxmancl38 ,

"In a conversation w/Steve Guttenberg Roy Delgado the Klipsch Heresy designer had a conversation w/Paul Klipsch. He told PK that the measurements were fine but it didn't sound correct. Paul then told Roy that there is a 5% Fudge Factor. He told Roy do you want a speaker that a computer likes or do you want a speaker the human ear likes."


This actually seems to indicate that measurements matter up to 95%, and the rest being the 'fudge factor'.

Of course a true heart over head subjectivist wouldn't care if that ratio was even 99% to 1%.

They'd still only be interested in the 1%.

And 1% of their total music collection is what they eventually might end up listening to.

I guess if they're happy that way... then why not?

Many folks love that 5% fudge the way Klipsch serve it up.

In a word - NO - measurements do not matter (so long as they're not terrible).  

The only thing that matters is how the components sound, which typically / often is not consistent with the measurements.  

The most important consideration is how well the components match / synergize with one another - are the electrical requirements of the speakers provided by the various components, and do the sound qualities of the speakers harmonize well with the components and do they fulfill your preferences in sound quality.  

Which is why... sticking with the components offered by a particular manufacturer and the components the manufacturer suggests - typically produce the best sound quality (e.g.  ARC preamp with ARC amp, McIntosh preamp with McIntosh amp, Magnepan speakers with Bryston electronics, Harbeth speakers with Hegel electronics, etc., etc.).  If you do so - you'll be far more pleased with the sound you hear... and far less frustrated in assembling your system.    

Yes they matter unless they are voiced. If you like that great! I do, I like a strong mid bass and full sound but fast. Hard to find. 

Had some Cerwin Vega DX9s that sounded like Cerwin Vegas. I like a Cerwin Vegas. I had a Lyngdorf 2170 with room correction and hooked them up. The correction amount was huge. Swear they sounded like 10k speakers. I think because they are so dynamic.

I’m looking at active now.  

Measurements are not always electrical information with a tool about a single piece of gear...

Measurements can be precise information about location of device in a room, lenght and diameter of an Helmholtz resonator or a diffuser for example ... It can be the Schroeder frequency which is so important to master in any room etc...It can be the precise ratio between timing of the first wavefronts and reflections relatively to our head position and perimeter......

We all listened to some unconscious or/and conscious "measurements" process , but some are way less informative about sound quality than others...Acoustic measurements are anyway more ESSENTIAL than partial electrical measures of the ASV group ....

This fact escape " gear subjective fetichists" and "tool measuring fetichists" alike...But not to acousticians and psycho acousticians...Here they measure AND listen in a process which is continuous for optimization results...

Then you listen to measurements yourself but you dont know it... 😁😊

I can’t remember the I ever listened to measurements....


Measurements are good especially when it’s comes to individual parts but the overall output are not supposed to depending on the measurements

human with special ears 👂 supposed to listen listen listen listen and tune it because at the end of the day we’re going to listen to it that’s the reason and purpose  it was made for