• Welcome to ASR. 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!

Questions on Measurement

sandwg

New Member
Joined
Feb 4, 2026
Messages
1
Likes
1
I was just reading a post on ASR and a reviewer (more like first impressions....) of an amp writes "amp A has greater sound stage than amp B". If the only variable in the equation is the amp, I assume something sent over the speaker wire is different or it's in the reviewers head. And if it's in the wire, it should be measurable. Some graph, curve, plot should be different. Am I distilling this down too much?

Comp Sci / Software engineer. They made me take a circuits class once. I have no idea how to use an oscilliscope.

TIA
 
Sound stage of stereo is an illusion created by the mind of listener. One would need quite strict testing protocols to assess an illusion. Starting by verbal definition, how to subjectively quantify illusion. What is small, average or big sound stage. This would be a lot of work for nothig. You can safely assume, that a casual reference to sound stage of an amp has no real meaning.
 
If the only variable in the equation is the amp ...

If the comparison is not made under strict controls, there will have been other variables. The listener must match levels, must have no "tells" as to which device is active, and other rigorous controls ... which are discussed here:



Even slight changes in head position, which might seem totally unimportant to the listener, can result in detectable differences in some rooms.
 
Definitely can be heard; my Apos Gremlin has noticeably wider stereo stage than my MOTU Ultralite's built-in headphone amp. Very 3D and holographic. In comparison, my Pietus Maximus has narrower stage than the MOTU, but the panning (positioning) of sounds are very precise on the horizontal plane, and it also does depth layering very well (if the music has depth layers, of course) while the other amps don't really do any of that -- everything is just on a rather flat plane in front of you.

How these relate to measurements, I have no idea. Many things about audio either cannot be measured yet, or to be more precise, we don't quite know how to explain these perceptual phenomena with the numbers our instruments are measuring. Sure, if you measure the electrical differences going into the drivers, the waves will wiggle differently, but how you translate those differences into "this amp has a wider stereo image", well, good luck with that :)

Today, you won't find the answer in the graphs and numbers -- you gotta listen to different amps and hear it for yourself.

Remember, perception first, then measurements. Measurements and numbers attempt to explain what we hear, but they're models. Models are not reality, and the the typical Audio Precision measurements just document gross phenomena; they won't tell you much about what you hear when you listen to a given headphone, amp, DAC, etc. (equipment measuring "the same" can sound different, and equipment measuring like crap can sound great, and "perfectly measuring" stuff can leave you rather cold... or not :) )
 
Remember, perception first, then measurements. Measurements and numbers attempt to explain what we hear, but they're models. Models are not reality, and the the typical Audio Precision measurements just document gross phenomena; they won't tell you much about what you hear when you listen to a given headphone, amp, DAC, etc. (equipment measuring "the same" can sound different, and equipment measuring like crap can sound great, and "perfectly measuring" stuff can leave you rather cold... or not :) )

Remember, perception first, then measurements.

This is true, but perhaps not in the way you might think. The perception needs to precede the measurement in time. We couldn't measure the orbit of Pluto's moon Charon until after 1978, which was the date it was discovered.
However, if you mean to say that perception is of primary importance and measurements are secondary, then I think you are very wrong. The reason is that the human senses - in hearing and in seeing - can be fooled terribly. see:


I believe that we're all familiar with visual illusions, so we need not go down that rabbit hole on a site dedicated to audio.

Measurements and numbers attempt to explain what we hear

There is no "attempt" about it. The recording industry uses measurements to record, mix and master. Understanding and controlling the difference in mics, gain riding, compression, panning, and the differences that mastering can accomplish were created and are controlled by measurements.
As for numbers ... they are simply the language of measurements.
What numbers and measurements CANNOT do is predict some other person's preference. If you take measurements of the recordings you like, you can create an information base that you can use to predict what you will like in the future. But that information base, which is a (again) a numerical basis for your preference, is unique to you, and is not useful to another person.
So in a sense, you are correct. measurements and numbers cannot explain what WE hear. But they can explain what YOU hear, if you learn how to use them.

Models are not reality

No, they are not. But humans cannot access reality :


the typical Audio Precision measurements just document gross phenomena

AP units document information to a far, far finer degree than humans can. They can show us differences between electronic circuits that are reproducible yet so negligible that humans cannot hear those differences.

they won't tell you much about what you hear when you listen to a given headphone, amp, DAC, etc

Lumping transducers such as headphones and speakers in a group along with electronic circuits such as amplifiers and DACs is not productive. Electronic circuits are designed (at least the ones that are competently designed) to output an electronic signal that is unchanged from the input except in power characteristics. The methods for doing this successfully have been understood and implemented for many years. If this were not so, we could not build circuits which are predictably successful at transmitting information over cell phones or digital files. Digital technology allows us to successfully communicate with the Mars Rovers and allows satellites to take pictures of the planets and transmit those pictures back to Earth.
Compared to that, audio recording and reproduction is a walk in the park.

As for transducers - particularly headphones - the placement of the transducer in the cup (in relation to the ear canal), differences in frequency response, and the differences in headphone impedance in relation to the circuit driving them can account for a great many differences in "sound". However ... all these things can be identified.

equipment measuring "the same" can sound different

If they actually sound different, then the measurements will show the reasons. If the measurements don't show the reasons, then the measurements were not administered to the necessary degree.
One example of this is the old 1/3 octave speaker response curve, with a frequency cutoff at 200 cycles. Newer response curves are taken with 1/12th, and even 1/20th octave. They show differences in much greater detail.


equipment measuring like crap can sound great, and "perfectly measuring" stuff can leave you rather cold... or not

This is simply a description of preference ... nothing more. Preference amounts to subjective opinion, and things that are subjective cannot reliably transfer information from one person to another because subjective opinions are not reproducible.
 
Last edited:
no idea how to use an oscilliscope
To begin with, you don't need an oscilloscope. One can formalize the soundstage's width somewhat. For example, listen to panned test signals and point finger in the direction of the phantom sound source. If a difference in azimuths is detected, it will be possible to look for the causes.
 
However, if you mean to say that perception is of primary importance and measurements are secondary, then I think you are very wrong. The reason is that the human senses - in hearing and in seeing - can be fooled terribly. see:
In audio, yes, perception has primary importance. I'm very much familiar with sensory illusions, but from my perspective, you have it backwards. Fooling or not fooling the senses, all what matters is how the music makes you feel.

Why do we bother with audio equipment at all? To enjoy music. Do you enjoy the music? That's all that matters.

The ear is not a measurement microphone, and the brain is not a tape recorder.

Anything related to audio is 50% engineering and "sciencey" stuff, and the remaining 50% is art. That goes for pretty much any engineering based discipline where the final output is "consumed" by our human senses.

If they sound different, then the measurements will show the reasons. If the measurements don't show the reasons, then the measurements were not administered to the necessary degree.
Look, I have an engineering degree and it's very empowering to think and feel one knows all the answers and can condense every phenomena into a few formulas and measurements. I had that phase too about 20-30 years ago. Then the more I learned about audio over the following decades, the less sure I felt about all these things. It's a journey.

Aspects of audio equipment production, design, and measurement are "understood", sure, how electricity "works" is understood, but how the more subtler aspects of amp/DAC/headphones design translate into "enjoyable sound" and various other hard to pin down perceptual phenomena, those aspects are not fully understood.

As one of my professors at uni in our very first lecture on acoustics said: measurements and electrical design are a starting point, not the final word. Sure, we shouldn't design equipment that blows up when you power it up. But beyond that basic engineering criteria, well-measuring equipment sometimes sound terrible, and horrible measuring equipment sometimes sounds awesomes. Your ears are always the final judge.

My friends who design amps and speakers surely use some engineering principles, but after a certain point it's about just trial and error and listening very carefully over long periods of time and training your ears to hear the subtleties. Things that "should be the same" according to the math can sound different in real life.

There is no "attempt" about it. The recording industry uses measurements to record, mix and master. Understanding and controlling the difference in mics, gain riding, compression, panning, and the differences that mastering can accomplish were created and are controlled by measurements.
Yeah, so I've been creating and producing music for over a few decades now as a hobby. You can be 100% sure that recording and mixing engineers don't stare at specs all day when producing music. In fact, probably it's more accurate to say what they do is closer to art than pure engineering. Most guys treat the various processors as black boxes; if it does something interesting to the sound, and they like it, it's good. They go by feel. Sure, you can take a peak at a spectrum analyser or whatever, but 90% of the process is driven by feel, starting from the recording stage. Recording guys usually have a closet full of mics, all interact with various sources in subtly different ways. It's common to try 10-20 mics with a particular singer, searching for a good synergystic combo. The best one is not determined by "measurements" (although the mics response can surely can be measured, but how does that help?), but simply by listening to the recorded sound and letting your feelings dictate what is the "best" mic for that particular job.

Same with micing guitar amps; the "true and tested" way is to move them around until something sounds cool. Sure, certain guys have their favourite techniques and mics, but either they stumbled upon a winning combo by trial and error (and listening), or the technique was passed down to them by some more experienced guy.

There is very, very little "measurement" going on in the actual music production process.

Anyway, there's lots of very engineering focused types here, I know, and my intention is not to "argue" all day about the engineering vs art aspects of audio or convince you. But being laser focused on the engineering aspects is myopic; this is not about creating factory equipment or an EEG machine but about audio reproduction, where the ultimate judge is your human perception, not a spectrum analyser or a computer evaluating the data.

Think of it this way: we can measure all sorts of health metrics of a human, like EEG, hormonal balance to some degree, etc. but you can't derive from those metrics how the individual really feels. The interactions are just too complex.

To the original poster, listen to different equipments if you get a chance, look at measurements too, if you like, and draw your own conclusions. I'm not saying ignore the engineering aspects, that's foolish, but keep an open mind. Like my prof said, don't assume the scientific theories and measurements will tell you the whole picture, experiment. To me, the answer is obvious: yes, there are very noticeable soundstage differences between amps. Surely it's measurable in a very abstract sense, but I don't think the current simplistic Audio Precision analysers metrics using test signals will reveal the truth about how you perceive complex material. The best way is just to listen to over weeks or months to a particular gear. Some aspects are immediately noticeable even to the untrained ear, but subtler aspects might take a long time to notice. By listening, you're effectively retraining your brain and create new neural pathways which simply don't exist in people who haven't developed an intense interest in audio. That's true for any human endeavour, not just audio.

I recommend watching the Wavetheory youtube channel about this "science vs art" debate. The guy is in a unique position given he's an actual trained scientists (unlike most people here) AND he's an audiophile as well.
 
Last edited:
You are exactly right.
But please don’t fall into the trap of forgetting that sighted bias affects perception (significantly more than most people realise)
Yeah, those things exist, and there have been quite a few times I've been tweaking an EQ of a muted channel while mixing and "hearing the difference" until I realise what's going on. Can fool me for a good 10-20 seconds at a time, still. I also realise some people might be more susceptible to auditory placebo than others. Then some people are just not that good at picking out differences, no matter how hard they try. I know I wouldn't be a world-class basketball player no matter how hard I tried; we all have different abilities.

I'm a fan of A/B testing (volume matched), but I've also realised some subtler aspects are not immediately obvious even when doing that if you don't know what to listen for (but those aspects can still subtly influence you enjoment, even though you're not "consciously" aware of them, at least not intially).

Yes, I can 10/10 pick my ESS vs AKM DAC on my best amp (the Pietus) in a blind test, but only using my Verum 2, and only on some dense full-bandwidth materials, and only on good days. When I'm tired, everything sounds the same, so it's a brain bandwidth thing. So it's definitely subtle, but yeah on good resolving components these differences can be heard. I also agree they're not night and day, but if you're really into audio, you start caring about last 5-10% differences. Most people wouldn't care.
 
Last edited:
all what matters is how the music makes you feel.

Correct. It's how the MUSIC makes you feel ... not how the reproduction equipment makes you feel. Beethoven's 7th Symphony is a truly magnificent composition, no matter which equipment I hear it on.
And if I'm put off by an incompetent performance of that symphony, that's on the orchestra and conductor, not the equipment.

this is not about creating factory equipment or an EEG machine but about audio reproduction, where the ultimate judge is your human perception, not a spectrum analyser or a computer evaluating the data.

Again, you are conflating two entirely different things. Audio recording and reproduction equipment has a great deal in common with an EEG machine, a computer or a spectrum analyzer. They are pieces of electronic equipment designed to do a job, and do it well. So is audio equipment. In fact, the ideal audio equipment is designed to do one job: to accurately reproduce the input signal ... nothing added, nothing subtracted.

You are conflating that job that the audio equipment does, with your appreciation of the information that it transmits (the music).

Not the same.
 
Correct. It's how the MUSIC makes you feel ... not how the reproduction equipment makes you feel. Beethoven's 7th Symphony is a truly magnificent composition, no matter which equipment I hear it on.
And if I'm put off by an incompetent performance of that symphony, that's on the orchestra and conductor, not the equipment.
Changing goalposts. I've seen that a lot here :)

But okay, I'll play along: it seems we should only have a single recording of the 7th then, am I understanding you right? One which is perfect, both at the music reproduction (orchestra) level, and the recording level, and we're done with this piece of music.

Except there are lots of various interpretation of said music, by different conductors, recorded in different halls, having different acoustics, with different recording techniques, and so on. Classical music enthusiast all have their favourite renditions and recordings of the same symphony.

Drawing a line at the recording level seems rather arbitrary. The preference in different presentations continues with the amps, transducers, DACs, etc.

Of course, it's fine if you draw the line there. What rubs me the wrong way is people dogmatically telling other that "you're doing it wrong" if you want to tailor your audio reproduction system to your preferences. Is there a law against that? :) Why is it "wrong"? :)

For example, I like transducers with a "BBC-dip" around 2-3 kHz. Especially when listening to metal. People telling me that a ruler flat frequency response is "correct" is just not helpful; I don't enjoy listening to that sound, plain and simple. In hifi shops knowledgeable salesman used to recommend speakers to people based on their genre preferences, which is the "correct" way to go about things if you're after maximising your enjoyment, not adhering to some "rules flat response curves are good" arbitrary ideal.

So is audio equipment. In fact, the ideal audio equipment is designed to do one job: to accurately reproduce the input signal ... nothing added, nothing subtracted.
Except there is no ideal equipment and there is no true "transparency to the source" (as per above). If you were involved in modern music production, you'd understand it a lot better.

Also, changing goalposts again. It started with "everything you should know about audio is in the measurements". Not really. In fact, far from it.

All gear presents music in different ways. If people enjoy one presentation better then the other, let them. One recurring theme of this forum is whack people on the head with "scientific sounding" arguments that basically gaslight people ("no you cannot hear that"), and/or telling people "you're enjoying your music wrong". Right :)

Let's keep an open mind, and let people enjoy what they enjoy, without dogmatism.

Also, listen to equipment once in a while between obsession over graphs :) You might discover some interesting things that way :)
 
Last edited:
I assume something sent over the speaker wire is different or it's in the reviewers head. And if it's in the wire, it should be measurable. Some graph, curve, plot should be different. Am I distilling this down too much?
Welcome to ASR! I would not say you've distilled it down too much.

Things are often in reviewers' heads, and much ink is spilled trying to assert alternative explanations when nothing shows up 'in the wire'. Why the audio world can't just accept this and move on, who knows?

For soundstage in particular, you would look for something like crosstalk, THD, IMD, or possibly frequency response as a culprit in a different soundstage.

If you find no meaningful differences in those things (or anything else) at normal listening levels, we can usually assume it's in the reviewer's head.

What rubs me the wrong way is people dogmatically telling other that "you're doing it wrong" if you want to tailor your audio reproduction system to your preferences. Is there a law against that? :) Why is it "wrong"? :)
There's absolutely nothing wrong with that. What's "wrong" is when someone asserts that the equipment is driving preference in how the music sounds, when the equipment isn't actually making the music sound different. If you look at the output of two different pieces of equipment and they are functionally identical, it would be wrong (as in incorrect) to say there's an *audible* basis for preference.

All gear presents music in different ways.
Nope, not all... there is a pretty wide selection of gear that presents music identically in practice.

Some stuff just has no character of its own. Some water tastes different than other water, but all distilled water tastes the same. It might taste different to you depending on how your day is going or what you had for lunch, but that's you and not the water. Transparent amps and DACs are much like this.

As an aside, I would suspect that if you are hearing a qualitative 5-10% difference between DACs there's a gain or impedance mismatch somewhere, or maybe one is using a very different filter setting... that's a very large subjective divergence for two DACs.

(significantly more than most people realise)

Agree... as a former marketer of audio equipment, I would bet money that I could set up a sighted A/B test that got people to say one piece of gear sounded at least 50% better than another. Without actually swapping anything. I don't think it would be that complicated. I once got similar results totally by accident via a firmware update.

It's just so pervasive that once you see how easy it is to drive "placebo effect" you never trust another review again. The human brain leads the ear around like a dog on a leash.
 
Last edited:
People can like whatever they like. I like cheesecake and jamon much more than I like broccoli, for instance.

But that's not what happens here at all. What happens here--and annoys people here on a regular basis--is more like if I were to log into a nutrition science forum and argue that all the science that says broccoli is more nutritious is no more accurate or valid than is my preference for cheesecake and jamon, which I believe are better for me because I like them more and those science nerds should put down their measurement tools and actually try tasting food for a change.
 
Changing goalposts. I've seen that a lot here :)

But okay, I'll play along: it seems we should only have a single recording of the 7th then, am I understanding you right? One which is perfect, both at the music reproduction (orchestra) level, and the recording level, and we're done with this piece of music.

Except there are lots of various interpretation of said music, by different conductors, recorded in different halls, having different acoustics, with different recording techniques, and so on. Classical music enthusiast all have their favourite renditions and recordings of the same symphony.

Drawing a line at the recording level seems rather arbitrary. The preference in different presentations continues with the amps, transducers, DACs, etc.

Of course, it's fine if you draw the line there. What rubs me the wrong way is people dogmatically telling other that "you're doing it wrong" if you want to tailor your audio reproduction system to your preferences. Is there a law against that? :) Why is it "wrong"? :)

For example, I like transducers with a "BBC-dip" around 2-3 kHz. Especially when listening to metal. People telling me that a ruler flat frequency response is "correct" is just not helpful; I don't enjoy listening to that sound, plain and simple. In hifi shops knowledgeable salesman used to recommend speakers to people based on their genre preferences, which is the "correct" way to go about things if you're after maximising your enjoyment, not adhering to some "rules flat response curves are good" arbitrary ideal.


Except there is no ideal equipment and there is no true "transparency to the source" (as per above). If you were involved in modern music production, you'd understand it a lot better.

Also, changing goalposts again. It started with "everything you should know about audio is in the measurements". Not really. In fact, far from it.

All gear presents music in different ways. If people enjoy one presentation better then the other, let them. One recurring theme of this forum is whack people on the head with "scientific sounding" arguments that basically gaslight people ("no you cannot hear that"), and/or telling people "you're enjoying your music wrong". Right :)

Let's keep an open mind, and let people enjoy what they enjoy, without dogmatism.

Also, listen to equipment once in a while between obsession over graphs :) You might discover some interesting things that way :)

You have been given all adequate information. You reply with sophistry, deflection and arrogant stereotypes.
You are either deliberately dense, a ragebaiter or a troll. Regardless, I'm putting you on "ignore".
 
Last edited:
To begin with, you don't need an oscilloscope. One can formalize the soundstage's width somewhat. For example, listen to panned test signals and point finger in the direction of the phantom sound source. If a difference in azimuths is detected, it will be possible to look for the causes.
Well, except for Haas Effect panning...
 
You are either deliberately dense, a ragebaiter or a troll. Regardless, I'm putting you on "ignore".
I'm sorry to see you're so close-minded that you're not open to a discussion that challenges your presently held fixed beliefs. Also sorry to see you're feeling you need to resort to personal insults -- that's a hallmark of a person running out of valuable things to say on the subject.

But it's all good; I hope my posts will make some people on this forum think about audio in a more open-minded manner. Again, I'm not here to argue :) It's a public forum, so we're all free to share our opinions. I'm fine with people disagreeing with me, however, we are free to express our opinions here, otherwise what's the point of a forum like this?
 
Back
Top Bottom