• WANTED: Happy members who like to discuss audio and other topics related to our interest. Desire to learn and share knowledge of science required. There are many reviews of audio hardware and expert members to help answer your questions. Click here to have your audio equipment measured for free!

Master Thread: Are measurements Everything or Nothing?

fpitas

Master Contributor
Forum Donor
Joined
Jul 7, 2022
Messages
9,885
Likes
14,245
Location
Northern Virginia, USA

Katji

Major Contributor
Joined
Sep 26, 2017
Messages
2,989
Likes
2,275
Well, yes. But I dont think anyone wants a playback gizmo that tries to reconstruct "real" from a stereo recording do they? I get why people want Mch recording AND playback but is anyone asking for Mch playback from stereo?
No,no. ...
[Maybe you saw it / heard it...]
MacRumors podcast yesterday, on Spatial Audio, interview format... good conversation about that aspect of it - the reprocessing/conversion of existing material (stereo, not produced for spatial audio), said he doesn't know why Apple thought they should immediately start doing that to everything, that it puts people off, gives a bad impression... [i think it's at around 60-70% in the podcast.]
However, that was not a major point in the whole story.

I somehow thought it would be good, and it turned out to be better than i might've hoped.
I only got to around 80% through, so I don't know whether it gets to some conversation sbout the VR/AR headset. Most important.

Although to begin with, and here now, just the spatial audio thing is important...with the head movement thing, the head tracking. Re creating the impression of reality. not just left and right, and up and down, it has to be full 360 degrees, it must work right / be real when you turn so the music - the musician is directly behind you. So that is supposed to be quite a major advancement in reproducing the "real."

* So yeh, the conversation there about it needing to be produced for spatial audio is good.

Spatial Audio Deep-Dive ft. Jonathan Morrison

https://omny.fm/shows/the-macrumors-show/spatial-audio-deep-dive-ft-jonathan-morrison
 

gavagai

Member
Joined
Jul 17, 2022
Messages
21
Likes
47
Location
France

Reality doesn’t exist until you measure it, quantum parlor trick confirms​



Yes and the article's rigor is unbelievable from the logical point of view. I quote :
"A quantum particle can exist in two mutually exclusive conditions at once. For example, a photon can be polarized so that the electric field in it wriggles vertically, horizontally, or both ways at the same time—at least until it’s measured."

It's quantum mechanics BS. There's literally and editorial industry for that.

And strangely enough, there is a strong connection with audio stuff. Because all the BS are rooted upon the Heisenberg's uncertainty principle, which is almost always stated like this "You can't know the position and the energy of a particle and the same time.". And the reason for that is the infamous wave-particle duality. A particle is in fact and at the same time a particle and a wave. It has and ambiguous nature, and it's this ambiguous nature that prevent us to know at the same time the position and the energy. Pretty strange strange isn't ?

But NO.

The real principle is "A particle can't have a precise position and a precise energy at the same time.", and the reason is that "particles" are in fact waves. And a wave can has two extreme opposite states that must be grasped by the mean of two domains : frequency domain and time domain, with Fourier transform between the two.
A wave can be :
  • A soliton, a solitary wave. A solitary wave has a precise position in the time domain (the position of the unique peak of the wave - think about impulse response in audio) , but it is not defined at all in the frequency domain (it doesn't have a frequency). Hence it's energy is not defined.
  • It's a "real" wave with multiple peaks. A "real" wave doesn't have a position in the time domain (because of the multiple peaks), but it has a well defined frequency, hence energy.
The wave-particule duality and almost quantum mechanics BS are not facts, but philosophical speculations build upon a logical fact, that is that time domain and frequency domain are the logical form by which science can grasp the fact that a wave is a unique thing with two very different aspects.

This logical form is the same in all wave involving domains, like audio stuff. So if you're involved in the audio stuff, maybe you can take all extraordinary claims made by articles about quantum mechanics with, at least, a grain of salt.
 
Last edited:

mglobe

Addicted to Fun and Learning
Forum Donor
Joined
Mar 18, 2022
Messages
501
Likes
874
Location
Texas
I just recently built a system after being in the home theater desert for 40 years. ASR became a go-to resource for me in deciding what to buy, with my speakers (KEF R3's) being the only thing I actually listened to prior to purchase. I'm very much on the objectivist side of the population. But one thing I've been thinking about is that if Dr Toole's work showed that people preferred speakers with a particular set of measurement characteristics (hope I have that right, I've honestly not read his publications yet), that means that they were able to identify superior speakers by listning.

So IF an experienced reviewer can separate their bias (be it brand, appearance, design, price,...) their subjective observations should have some validity. I do realize that the possibility of a listener (or reviewer) being bias-free is a huge assumption, particularly when a reviewer is likely to be influenced by advertising income.

I'm very data driven (I work data for a living) and I would never buy speakers without an audition unless I had the kind of data that Amir provides here. But I can't discount the value of a properly done audition.
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,915
Likes
2,070
Location
Tampa Bay
I just recently built a system after being in the home theater desert for 40 years. ASR became a go-to resource for me in deciding what to buy, with my speakers (KEF R3's) being the only thing I actually listened to prior to purchase. I'm very much on the objectivist side of the population. But one thing I've been thinking about is that if Dr Toole's work showed that people preferred speakers with a particular set of measurement characteristics (hope I have that right, I've honestly not read his publications yet), that means that they were able to identify superior speakers by listning.

So IF an experienced reviewer can separate their bias (be it brand, appearance, design, price,...) their subjective observations should have some validity. I do realize that the possibility of a listener (or reviewer) being bias-free is a huge assumption, particularly when a reviewer is likely to be influenced by advertising income.

I'm very data driven (I work data for a living) and I would never buy speakers without an audition unless I had the kind of data that Amir provides here. But I can't discount the value of a properly done audition.
Well I agree here. I am very data driven as I am a low-level software engineer and deal with lots of assembly and stuff on a daily basis. Data is the thing that I look at most. But I also understand that one set of data does not accurately represent the entire picture and also data can be manipulated. We recently saw this happen with the smsl M500 I believe it was, as people noticed that playing out of only one channel would cause a severe reduction in performance. As the tones were summed across all the channels to minimize the Distortion in measurements. However when playing music both channels are not likely to be exactly the same so this would result in worse performance.
There are also things like crosstalk which can affect the performance of a DAC or amp. However I have noticed over the years that the linearity test is in my opinion the most important one to determine how something actually sounds. As I have been able to personally make predictions about a device's performance by only hearing it first and I have picked out poor linearity measurements on more than one occasion.

When it comes to things like headphones it is very hard for me to say I would purchase something serious without an audition first. Despite measurements showing how clear the headphones or speakers could be; they don't necessarily represent exactly how they will sound. Since everyone's room, head and environment are different.
So even though the headphones May perform well you may really dislike the pads or the weight or they don't fit your ears well. Speakers are the same way in that you may not like the way they look, they are not the right size for you and your room, they have too high of power requirements versus your setup. Maybe you also don't want to change the entire setup.
Thankfully this forum has a good mixture of information of different types. We don't just have one person publishing measurements and no other information at all.
 

Blumlein 88

Grand Contributor
Forum Donor
Joined
Feb 23, 2016
Messages
20,952
Likes
38,082
Well I agree here. I am very data driven as I am a low-level software engineer and deal with lots of assembly and stuff on a daily basis. Data is the thing that I look at most. But I also understand that one set of data does not accurately represent the entire picture and also data can be manipulated. We recently saw this happen with the smsl M500 I believe it was, as people noticed that playing out of only one channel would cause a severe reduction in performance. As the tones were summed across all the channels to minimize the Distortion in measurements. However when playing music both channels are not likely to be exactly the same so this would result in worse performance.
There are also things like crosstalk which can affect the performance of a DAC or amp. However I have noticed over the years that the linearity test is in my opinion the most important one to determine how something actually sounds. As I have been able to personally make predictions about a device's performance by only hearing it first and I have picked out poor linearity measurements on more than one occasion.

When it comes to things like headphones it is very hard for me to say I would purchase something serious without an audition first. Despite measurements showing how clear the headphones or speakers could be; they don't necessarily represent exactly how they will sound. Since everyone's room, head and environment are different.
So even though the headphones May perform well you may really dislike the pads or the weight or they don't fit your ears well. Speakers are the same way in that you may not like the way they look, they are not the right size for you and your room, they have too high of power requirements versus your setup. Maybe you also don't want to change the entire setup.
Thankfully this forum has a good mixture of information of different types. We don't just have one person publishing measurements and no other information at all.
Pretty much all sigma delta dacs have perfect linearity.
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,915
Likes
2,070
Location
Tampa Bay
Pretty much all sigma delta dacs have perfect linearity.
Maybe most recent ones tested here... But many had issues just a few years ago. Poor crosstalk, poor overall gain stage / buffer layout, etc... Some devices also using ancient DAC chips with poor power control...
 
D

Deleted member 49736

Guest
Measurements are meaningless beyond a point. That point is floating, it’s arbitrary for everyone. Mankind does not have a firm enough understanding of psychoacoustics to develop a mathematical formula that can accurately calculate the aforementioned tipping point for an individual and their hearing capabilities.

The only time measurements matter is when a device clearly measures so poorly, that it’s guaranteed to adversely affect the performance of the stereo system.

There are so many factors that go in to the overal performance of a stereo system, it’s beyond the mental capacity of human beings to have all of these factor in mind simultaneously. A device with excellent measured performance can easily be squandered.
 
D

Deleted member 49736

Guest
I'd hate to undergo heart surgery while dependent on a heart-lung machine that someone had arbitrarily set to a floating point of reference as to quality because the surgeon didn't think they needed any better. Although audio is not life-threatening, my point is that there are levels of engineering and manufacturing excellence beyond what we can hear

You actually proved my point. Doctors have a strong enough understanding of the pulmonary system’s need for oxygen to make the necessary precision calculations to keep you alive in that scenario. Your hypothetical death would be a function of human error, not an inadequate machine.

You argue that better measurements yield something we can’t hear. That literally means it is irrelevant. The point of this thread is how measurements matter relative to the performance of a stereo system. Stereo systems are qualified by human hearing, not some arbitrary compulsion to have the “best” engineered electronics within your stereo.

Perfect measuring components are not guaranteed to yield the most enjoyable stereo system.
 
D

Deleted member 49736

Guest
my point is that there are levels of engineering and manufacturing excellence beyond what we can hear or see or feel.
You argue that better measurements yield something we can’t hear.
No, I didn't say that. I didn't even insinuate that.

Yes, you explicitly said it. You also brought seeing and feeling into a discussion about listening.

Again ..... nope. A doctor's understanding of the patient's needs can only be satisfied on a machine that is correctly calibrated and functions as it should. Tests and measurements assure that, and my point is that it is necessary for those tests and measurements to go far beyond human perception.

An engineer needs tools just the same as a carpenter or electrician. Biomedical devices are built by engineers, not doctors. The device arrives in the hands of the doctor calibrated to support the relevant criteria necessary to operate in the intended use case. Just the same as Harman uses measurements to develop the "best" measuring products. That doesn't mean 9 out of 10 people won't prefer a Focal that measures "worse". Keeping a person alive has absolutely zero psychological confliction. There is only dead or alive. It is very easy to get caught in a position where you're unable to pick a preference between two speakers. They could both exhibit excellent qualities and sound vastly different. They could both measure nearly perfect and sound vastly different.

Incorrect. What you should have said is, "That means it is irrelevant to someone who uses only the limits of their perception to make judgements." In this case, I imagine that limitation would be your hearing, but I might be wrong.

Wow, very rude to tell me what I "should have said". Are you god? I meant what I said and it is true. You proved my opinion.

Your incoherent argument straddles the line between two important questions:

1. Do measurements matter to engineers (in all fields)

2. Do measurements directly corelate to the listening experience

I'm here to answer question 2. I can tell you are not an engineer, so you are also here to answer question 2. Leave question 1 behind, the answer is obvious.

As an aside, you do realize it is clinically proven that most humas fall victim to rejecting valid evidence put in front of them for the vain and insubstantial reason that it just doesn't support their preconceived and uneducated views? Dunning-Krueger hard at work.
 

theREALdotnet

Major Contributor
Joined
Mar 11, 2022
Messages
1,213
Likes
2,096
As an aside, you do realize it is clinically proven that most humas fall victim to rejecting valid evidence put in front of them for the vain and insubstantial reason that it just doesn't support their preconceived and uneducated views?

Yes, that happens a lot. For example when someone buys an expensive loudspeaker and then fights tooth and nail to defend its supposed audio quality, despite objective evidence (measurements) to the contrary.

Dunning-Krueger hard at work.

However, that is not the Dunning-Kruger effect. Your post on the other hand…
 

Jimster480

Major Contributor
Joined
Jan 26, 2018
Messages
2,915
Likes
2,070
Location
Tampa Bay
My PhD disagrees with you. D-K is incompetency in judgement.
PhD means exactly squat. I have absolutely zero respect in any way shape or form for college degrees. Colleges award honorary degrees as well to morons who donate money or are affiliated with specific political groups. When you invoke something as worthless as a degree to try to win an argument it only shows that you are actually losing the argument and have no further relevant information to provide which supports your viewpoint.

I worked at a startup company where every person there had degrees from prestigious universities and everybody didn't know how to do their jobs. Most of them relied on me who was actually the only engineer building the software platform that ran the entire business. When I told them facts that they did not like they often invoked their degrees to try to say that what they were telling me was the actual truth.
When I pulled relevant data from the databases to prove things like us not having meaningful numbers of clients; the people with degrees would then pull arbitrary data and claim that this data was actually more valid because they have fancy degrees that make them qualify to decide what data is valid.

I went on to focus more on my own company and made it very profitable and left this company which went out of business and was purchased by one of its clients. That client then went out of business as well and was billed out by Mubdala Investments.
Today some of these supposedly brilliant people still work in this new conglomerate company called "Reef". Which claims to be the biggest parking company in the world and yet has no functioning platforms and a barely working website.
Recently one of my ex co-workers came to brag to me about how big and successful their company had gotten and when I asked him how come the app does not work and the website doesn't have any relevant information he could not give me an answer.
When I asked him how the company specifically makes money because nothing written suggests that it has any actual Financial viability he could also not answer the question.
He told me that they make money through partnering with other people who white label products from influencers in order to run ghost kitchens.
I asked him which successful ghost kitchens they had actually launched and where I could find some of the products and he could also not answer the question.

He then went on to tell me that they are just so successful and he is involved in so many things that he just doesn't have time to know about these things that I'm asking him. And that I should just be happy that something that I originally worked at became so big and great.
Digging into some financial information that I was able to obtain publicly suggests that the company is bleeding money and has no real viability. That they spend huge amounts of money for these influencer contracts and advertisements which only generate money at specific events where they never have an ROI...

This post has become too long now but I hope you get the point. In the case of audio products; better measurements means better performing products. More overhead in measurements versus audibility leads to more of a cushion for performance in other situations not in the measurement. Better engineering is better engineering no matter how you look at it.
How good your hearing is or what's your preferences are do not add or subtract from the quality of engineering in a specific product in standardized measurements.
 
D

Deleted member 49736

Guest
PhD means exactly squat. I have absolutely zero respect in any way shape or form for college degrees.
People who are admitted to the hospital via the emergency room and leave the hospital healthy are literally living proof to the contrary. They don’t just give PhD’s to anyone who can show up to class and turn in homework on time.

I’m sorry you were the kid that had to carry the group project. I was also that kid. It sounds like as an adult you’re in an industry full of non-professional people. In my world, my colleagues are dependable and -dare I say it- inspiring.

My point was simply that D-K is misunderstood by the masses, because they are victims of it.
 

Geert

Major Contributor
Joined
Mar 20, 2020
Messages
1,973
Likes
3,644
Mankind does not have a firm enough understanding of psychoacoustics to develop a mathematical formula that can accurately calculate the aforementioned tipping point for an individual and their hearing capabilities.

We don't absolutely need a formula for individual hearing capabilities. There's a ton of research that established hearing thresholds and even preference via tests. And audio is not the only engineering area where measured performance or load of certain elements is used during design.

The only time measurements matter is when a device clearly measures so poorly, that it’s guaranteed to adversely affect the performance of the stereo system.

This statement confirms we have some understanding of what performance level is required to avoid issues.

There are so many factors that go in to the overal performance of a stereo system, it’s beyond the mental capacity of human beings to have all of these factor in mind simultaneously.

That's the second time you envision an unrealistic and unnecessary objective to support your beliefs. When engineers design a mile long bridge they also don't have a single mathematical formula, nor do they have all factors in mind simultaneously. They design and measure all sub components and then put all the pieces together. And rest assure they will measure the end result, even when normal users think it looks fine.
 
Last edited:

solderdude

Grand Contributor
Joined
Jul 21, 2018
Messages
16,159
Likes
36,902
Location
The Neitherlands
Perfect measuring components are not guaranteed to yield the most enjoyable stereo system.
But neither are less perfect measuring components. ;)

Most enjoyable suggests personal preference.
That's where the correlation between measurements and perception usually ends (when not testing blind, level matched and with statistical relevant nummer of attempts)
 
Last edited:
D

Deleted member 48726

Guest
Measurements are meaningless beyond a point. That point is floating, it’s arbitrary for everyone. Mankind does not have a firm enough understanding of psychoacoustics to develop a mathematical formula that can accurately calculate the aforementioned tipping point for an individual and their hearing capabilities.

The only time measurements matter is when a device clearly measures so poorly, that it’s guaranteed to adversely affect the performance of the stereo system.

There are so many factors that go in to the overal performance of a stereo system, it’s beyond the mental capacity of human beings to have all of these factor in mind simultaneously. A device with excellent measured performance can easily be squandered.
Psychoacoustics.. -It is the perceived, subjective evaluation of the sound -right?

The goal is to purchase, within ones budget, the best neutral sounding equipment which measurements helps to clarify and uncover. What one may do to the sound afterwards IR to EQ to suit ones preference is individual to a taste preference. So why should one not pursue the best measuring devices to eliminate the risk of hearing flaws, which may or may not be within the capability of the individuals hearing?

Reading numerous posts about this, I can easily deduct that peoples hearing differ to the extreme, but objectively the best starting point would be to invest in the best possible measuring equipment within the budget as a starting point. -So I think you are wrong in stating that measurements only matter when measuring poorly and that this statement even contradict your last sentence, that says excellent measuring devices can be squandered. -Because why can it then, in your own words, even be "guaranteed" to "adversely affect the performance..." when measuring poorly.. -That is subjective. But the measurements are not.

Some meaning in the above may be lost in translation due to the poster being a non-native english speaker. :)
 
Top Bottom