Study Shows No Correlation Between Price and Sound Quality in Headphones

[<eargiant]

Almost all audio columnists claim their memory of previous headphones is adequate for comparative evaluation. Absurd.

If one is talking about the memory between immediate comparisons I don't find that to be absurd at all. I would be willing to submit to a blind listening test between the Sennheiser HD600 & HD800S on the condition that if I can pick the 800S every single time with 100% accuracy, whoever sets up the test will have to buy me a set of 800S .

Accuracy = frequency response. Electrostatic headphones wouldn't have any real advantage over conventional headphone design. Much like loudspeakers.


Inner Fidelity looks like it does good measurements for headphones in their reviews.


If I were headphone shopping today, I would just look for a frequency response that matches Olive, Welti, and McMullin's target curve, the 'neutral' response.

Can you tell me how well Inner Fidelity’s graphs on the HD600 & HD800S match with the O,W & M target curve? Which one would be predicted to be better?

 

[shadyJ]

If one is talking about the memory between immediate comparisons I don't find that to be absurd at all. I would be willing to submit to a blind listening test between the Sennheiser HD600 & HD800S on the condition that if I can pick the 800S every single time with 100% accuracy, whoever sets up the test will have to buy me a set of 800S .





Can you tell me how well Inner Fidelity’s graphs on the HD600 & HD800S match with the O,W & M target curve? Which one would be predicted to be better?

Here is the target curve taken from this article:

Anyone can compare the Inner Fidelity measurements against those curves. Read the article to understand the explanation of the curve.

 

[<eargiant]

Here is the target curve taken from this article:

Anyone can compare the Inner Fidelity measurements against those curves. Read the article to understand the explanation of the curve.

I read and understand the point of the study. I also read when you said "If I were headphone shopping today, I would just look for a frequency response that matches Olive, Welti, and McMullin's target curve, the 'neutral' response." which seems logical to me, but...

The problem is that I am not an expert at reading and comparing graphs. Ideally, I'd like tho overlay the HD600 & HD800S Inner Fidelity graphs (to scale) over the Target graph above but I have no easy way of doing that.

Can someone out there do it for the sake of this discussion? I know which of the two headphones I prefer based on actual listening but I'm trying to better understand how they both compare on paper. The differences were not subtle.

Is my preference for the HD800S closer to the "Target" or further away from it?

 

[shadyJ]

I don't see a frequency response of the HD800S available anywhere, but I do see one for the HD800B. I also see one for the 600. They look fairly close to neutral target response. It looks like the 600 is a bit closer to the neutral response target than the 800B. Both their bass responses look a bit more neutral and a tad lower than listener's preference from the study.

 

[<eargiant]

I don't see a frequency response of the HD800S available anywhere, but I do see one for the HD800B. I also see one for the 600. They look fairly close to neutral target response. It looks like the 600 is a bit closer to the neutral response target than the 800B. Both their bass responses look a bit more neutral and a tad lower than listener's preference from the study.

I think these are the two measurements in question:

https://www.innerfidelity.com/images/Sennheiser_HD800S_PDF_AllPlots.pdf

https://www.innerfidelity.com/images/SennheiserHD600.pdf

Are they really that close on paper (and to the Target) because in actual listening, there is no mistaking which are true high performance transducers. As I said, the differences are not subtle, from the lowest bass through the midrange right up to the highest treble.

 

[shadyJ]

According to the frequency response graph, the HD800S will have more upper treble and deeper bass. It has a wider frequency response band.

One thing that can affect bass response is the seal of the headphones around the ears, so if the HD800S has a better fit on your head, that can also contribute to the bass response.

 

[<eargiant]

According to the frequency response graph, the HD800S will have more upper treble and deeper bass. It has a wider frequency response band.

One thing that can affect bass response is the seal of the headphones around the ears, so if the HD800S has a better fit on your head, that can also contribute to the bass response.

A tight seal can definitely boost the bass especially in a closed design. Believe me I know, I have the Audio Technica ATH-M50. Actually, along with the M50's I also had the Massdrop AKG 7XX on hand which I believe are the AKG Anniversary headphones so I listened to all 4 headphones back to back. But for the sake of this discussion it makes more sense to keep it to the HD600 & HD800S.

The difference between the 600 & 800S is more than just lows or highs but the much more articulate lows, mids and highs that they reproduce. It's the overall transparency and resolution that they provide. Like they say, they remove the glass pane. I hate to use the term "air" because I know many around here don't like it but that's the best way I can describe it. As soon as I listened to them I realized that they provide what I seek out and prefer when looking for speakers.

Interestingly, the HD800S sounded the most similar to how my 805D's sound in my home, go figure (I know some hate how they measure). At first listen of the HD800S I remember nodding my head and thinking "That's what I'm talking about!" To my ears, a stunning top notch headphone.

I worry that inexperienced listeners/enthusiasts will look at thread titles like this and think that if measurements of speakers or headphones are close or within a target band that one transducer will sound similar or just as good as the other. Why bother to listen, you won't hear a much of a difference- it says so right here on paper. I see that kind of thinking all the time. A person looks at a graph and say's "Yup, these will sound just as good as those".

I urge everyone to take the opportunity and go to your nearest Sennheiser retailer and do a side by side listening comparison between the HD600 and the HD800S to better understand what you can gain. It's much more than one would expect from comparing their graphs on a piece of paper.

 

[GrizzledGeezer]

Sean Olive has long been spreading the erroneous belief that frequency response is the only thing that matters -- that (other than distortion) it is the sole determining factor in the accuracy of any product. I guess that means I should toss my STAX electrostatic headphones in the trash.

By the way, STAX made an equalizer (the ED-1) which corrects for the frequency response difference between "free field" and "diffuse field" sound. I have the ED-1, and you can indeed hear the difference (which I judge to be an improvement). But the article says nothing about this.

The following (admittedly hyperbolic) piece discusses the ED-1.

http://rinchoi.blogspot.com/2014/02/stax-ed-1-monitor-legacy-from-past.html

 

[Joe B]

Sean Olive has long been spreading the erroneous belief that frequency response is the only thing that matters -- that (other than distortion) it is the sole determining factor in the accuracy of any product. I guess that means I should toss my STAX electrostatic headphones in the trash.

By the way, STAX made an equalizer (the ED-1) which corrects for the frequency response difference between "free field" and "diffuse field" sound. I have the ED-1, and you can indeed hear the difference (which I judge to be an improvement). But the article says nothing about this.

The following (admittedly hyperbolic) piece discusses the ED-1.

http://rinchoi.blogspot.com/2014/02/stax-ed-1-monitor-legacy-from-past.html

Please, not in the trash! I'd be glad to take them off your hands. I have a spot right next to my Oppo PM-2's where they can live.

 

[GrizzledGeezer]

Please, not in the trash! I'd be glad to take them off your hands. I have a spot right next to my Oppo PM-2's where they can live.

Off my head would be more accurate, I think.

Thank you for the implied agreement as to the superior sound quality of electrostatic and planar-magnetic headphones.

 

[Joe B]

Off my head would be more accurate, I think.

Thank you for the implied agreement as to the superior sound quality of electrostatic and planar-magnetic headphones.

From my subjective perspective, it is not implied, but a reality.

 

[sszorin]

This "study" was an exercise in stupidity. A good way to judge its relevance and value is to make a comparative study and come to the conclusion that prices of automobiles have no relation to the number of wheels of an automobile. So buy a cheapest car because it has 4 wheels like a car that costs 100 000 dollars.

 

[killdozzer]

This is my effort to contribute to this debate. Correct me if my understanding of the article is somehow flawed.


This is what I took with me:

High correlation between FR and SQ and low correlation between FR and HP puts you in a position to claim that high price does not make for good sound quality. Graph also indicates that:

- Red: there are some good sounding headphones at high price that don't have all that good / linear FR

- Blue: there are some expensive, good sounding headphones with good FR

- Green: there are some linear FR headphones that fail to thrill


Graph also indicates that there are many non expensive, good sounding, good / linear FR headphones out there (the portion labeled as "The point the article was trying to make").

It goes without saying I'm interested if I misunderstood something.

Are the best headphone the most expensive? No (well, yes).

Could we have an example of this (also is it no or yes)?

This "study" was an exercise in stupidity.

Perhaps to leave the heavy guns as a last resort? For one might say the same for your car analogy. Four wheels might pertain to functionality, but not quality.

 

[GrizzledGeezer]

There are people -- among whom Sean Olive is perhaps the best known -- who insist that amplitude response (FR) is the only audible difference among components. These people have spent the last 50 years performing meaningless tests intended to use "science" to humiliate serious listeners into denying they hear what they (probably) hear. One of these people even goes so far as to claim -- in so many words -- that the sonic errors introduced by any amplifier with low measured distortion are essentially inaudible. (He has obviously never heard a Crown K-series amplifier.)

...and yet no audiophool ever found anything the least bit strange in the fact that the ENTIRE overpriced audio industry is somehow situated precisely in the immeasurable part of the audio spectrum... Golden rule of audiophoolery: if it's overpriced, it has to be immeasurable.

Might I point out that "unmeasurable" and "immeasurable" hardly mean the same thing? Furthermore, whatever is audible must be measurable, but not necessarily the other way 'round.

It should be obvious (???) that a speaker's sound quality is primarily determined by how quickly (and "precisely") it stops moving when the driving signal is removed. Peter Walker made a strong case for this over 35 years ago, but no one has ever definitively demonstrated it. Olive (et al) will counter-claim that drivers with the same amplitude and phase response must have the same "transient response" (which is what you're taught in school), and thus sound the same, but again, this has never been subjected to careful research.

People (including myself) do not prefer planar drivers because they "like" their sound, but because they think it's more accurate. Which is the whole point of high fidelity sound reproduction -- the closest approach to the original sound.

At the risk of offending anyone, it's not my fault that people don't understand how science works, and are mostly interested in "proving" what they already "know" to be true.

 

[killdozzer]

My signature not being the case here, thx for the "heads up" anyway.

Furthermore, whatever is audible must be measurable, but not necessarily the other way 'round.

This is very often my point. It should also be obvious (???) that there's a huuuge gap waiting to be filled with a measuring device that measures other aspects of sound. Such a huge gap that one would probably end up filthy rich upon inventing it. I do find the lack of interest in inventing such a device strange. If so much as one fourth of your income would depend on proving to buyers there is an audible difference, you'd be crazy not to invest in such a device.


Anyhow, this has nothing to do with my post.

It should be obvious (???) that a speaker's sound quality is primarily determined by how quickly (and "precisely") it stops moving when the driving signal is removed.

Among other features, right? Why do you think someone would dedicate himself to undermining the fact you stated? Driver not stopping instantly creates easily perceivable distortion, how could one argue that?

no one has ever definitively demonstrated it.

This is very important, right?

people don't understand how science works

There's no such thing as science without measuring and providing proof. Until that happens all we have is some strong built cases.

Now, other than clearly demonstrating that you're pissed, how does your comment relate to mine? Since you quote me, I'd expect to see some; this is why you're wrong or this is why you're right type of info. I only saw that in regards to my signature.

My question was without affectation and I fail to see why you treated me with yours.

I thought I covered all types of products important for this debate. The only thing I could find to be the possible reason of someone's disdain is if someone would put forward that most high priced headphones sound good and that the correlation is more indicative.

If that's what you're saying, than we have something to talk about. If it's just that you don't like some Olive guy, I can't help you.

 

[William Sommerwerck]

It seems you are over reacting to my having mispositioned my response. As to "affectation", I have no idea what you mean.

It should also be obvious (???) that there's a huuuge gap waiting to be filled with a measuring device that measures other aspects of sound [than frequency response]. Such a huge gap that one would probably end up filthy rich upon inventing it. I do find the lack of interest in inventing such a device strange. If so much as one fourth of your income would depend on proving to buyers there is an audible difference, you'd be crazy not to invest in such a device.

You can't build such a device without knowing what errors you're trying to measure! Audio reactionaries have been successfully blocking attempts to find out what they are. They're not things that are going to yield to a few weeks of trivial investigation. We're talking about years of difficult, annoying research that no one wants to pay for.

Why do you think someone would dedicate himself to undermining the [presumed] fact you stated? Driver not stopping instantly creates easily perceivable distortion, how could one argue that?

Because it blurs the sound. It doesn't introduce distortion. As I stated, the reactionaries believe that all systems with the same measured amplitude and phase response necessarily sound the same.

Bottom line... Science proceeds by asking questions, not by assuming you already know the answers. Olive, et al, don't understand this, so they continue to muddy the waters.

I'd like to let this drop for the time being.

 

[killdozzer]

It seems you are over reacting to my having mispositioned my response. As to "affectation", I have no idea what you mean.

It should also be obvious (???) that there's a huuuge gap waiting to be filled with a measuring device that measures other aspects of sound [than frequency response]. Such a huge gap that one would probably end up filthy rich upon inventing it. I do find the lack of interest in inventing such a device strange. If so much as one fourth of your income would depend on proving to buyers there is an audible difference, you'd be crazy not to invest in such a device.

You can't build such a device without knowing what errors you're trying to measure! Audio reactionaries have been successfully blocking attempts to find out what they are. They're not things that are going to yield to a few weeks of trivial investigation. We're talking about years of difficult, annoying research that no one wants to pay for.

Why do you think someone would dedicate himself to undermining the [presumed] fact you stated? Driver not stopping instantly creates easily perceivable distortion, how could one argue that?

Because it blurs the sound. It doesn't introduce distortion. As I stated, the reactionaries believe that all systems with the same measured amplitude and phase response necessarily sound the same.

Bottom line... Science proceeds by asking questions, not by assuming you already know the answers. Olive, et al, don't understand this, so they continue to muddy the waters.

I'd like to let this drop for the time being.

??

I wasn't even talking to you. Do you have multiple accounts?

 

[KEW]

??

I wasn't even talking to you. Do you have multiple accounts?

Oopsidaisy!

I think he mixed up his "safe" account with his "flame on" account!
Good call

 

[killdozzer]

I really don't have a horse in this race. I was genuinely interested in how you think, what do you know and what are your arguments. I'm not pro-Olive nor counter-Olive. But:

Audio reactionaries have been successfully blocking /.../

I think I can safely say:

We're talking about years of difficult, annoying research that no one wants to pay for.

No. We're talking about THEE crucial research that would once and for all prove better sounding amps deserve higher price, 60k TT's are worth every cent, perhaps even 15 or 20k cables are as well. The market for these products would spread like wild fire. There are a lot of people who have the money but don't want to flush it down the toilet. If it isn't flushing down the toilet, every overpriced audio company should be interested in such a research and such a device. It would earn them crazy dosh. Even I could afford at least 200-300$ Nordost, but I'm not going to because someone couldn't be bothered with doing the research (coming from the mouth of someone who keeps explaining what science is).

And:

No. We're talking about the fact that through the act of paying the asking price, you're behaving as if the difficult, annoying research was paid for, done and conclusively shows that your opinions are confirmed and this is not the case. But don't let that worry you, it is the most common trap where you'll find the company of all the research deniers.

My actual position is this: until it's proven, don't act (pay) as if it's proven. Wait for the answer.

Until you get the answer, it's possible you're paying for something that is in your head. Doesn't that possibility worry you?

I'd like to let this drop for the time being.

Your dropping skills are below poor. When you want to drop a subject you don't come out with "audio reactionaries". You just drop it.

And all of this has been such a waste of breath and time and waaaay off topic.

I'd still like to know if I misunderstood the point of the article. There's one more possible flaw I forgot to mention; if someone wants to say that the first proposition (close relation of SQ and FR) is flawed and thus every analogy must be flawed.

 

[swspiers]

Wow. I thought I was the only one skeptical about the assumptions, hypotheses, and methodology of Sean Olive. Glad to see other people share similar concerns...