[ralphp@optonline]

I'm interested to know whether by measuring we can predict how a piece of audio gear will perfom.
For example, I am told that the iPhone measures very well but it sounds flat and lifeless to me compared with my external amp/DAC. This may well be placebo.
The Mojo gets 10.000 posts. In the main extremely favourable. I cannot hear any difference from the iPhone. Is it me?

It's not the iPhone that sounds flat and lifeless but the combination of the iPhone and the headphones/earphones that does not sound as good as the combination of iPhone -> external amp/DAC -> headphones/earphones. Change any piece of equipment and the overall sound will be different. Try the iPhone with different sets of headphones until you hear a set that sounds good to you. In the end how it sounds to you is only "measurement" that matters.

 

[krismusic]

It's not the iPhone that sounds flat and lifeless but the combination of the iPhone and the headphones/earphones that does not sound as good as the combination of iPhone -> external amp/DAC -> headphones/earphones. Change any piece of equipment and the overall sound will be different. Try the iPhone with different sets of headphones until you hear a set that sounds good to you. In the end how it sounds to you is only "measurement" that matters.

So you would say that the iPhone should give excellent results with the right phones?
The reason that I like the amp/DAC is that it introducing a distortion that I happen to like?

 

[ralphp@optonline]

So you would say that the iPhone should give excellent results with the right phones?
The reason that I like the amp/DAC is that it introducing a distortion that I happen to like?

Yes with the proper headphones/eaphones the iPhone should produce excellent results. And is it more likely that amp/DAC is producing less distortion with your head/earphones than with the iPhone alone. And the distortion when using just the iPhone is being caused by a poor match between the iPhone and your head/earphones and not because the measurements of the iPhone are somehow "wrong".

 

[mmerrill99]

The debate doesn't seem to go anywhere but it depends if you believe that everything about audio performance can be measured.

And again, congrats - this is the most interesting of the last couple of posts and it is the most crucial difference between me & the people who accuse me of "magic thinking".

What some call "magic thinking" I call auditory processing & the only "magic" is that we don't fully understand it's operation. When something is not fully understood, it's common to consider it "magic" but this is a fault of the arrogance of assuming that we know all there is to know so anything outside of this is "magic"

So, let's go back to the idea of "loudness" - this is a psychoacoustic phenomena which isn't the same as "intensity" (the external measure of sound). Loudness is the internal subjective experience of how loud a signal is. So what is being said here? Does the "measurable characteristic, intensity correlate to our perception & "define" what we hear? No, it doesn't. According to JJ johnston, "loudness and intensity can mostly be related by a complex calculation, Johnston noted several important caveats;

• The calculation of the relationship between intensity and loudness is highly complex.
• In the most severe cases, perceived loudness does not closely track measured intensity. "

What is measured is not correlated to what is heard. Is this "magic thinking"? No! Why? Because it is something that JJ Johnston says is "highly complex". Does anyone here know the details of this "highly complex" relationship calculation? Yes? Please explain it.

So why is sound stage perception considered to be "magic thinking" rather than a "highly complex" mixture of ITD, ILD, & other factors yet to be teased out?

This goes back to the post of Krismusic "Do you believe that everything about audio performance can be measured"

Some people take such a statement as an affront to rational thought but logic dictates that if it applies to "loudness" then why would soundstage be any different?

 

[mmerrill99]

So you would say that the iPhone should give excellent results with the right phones?

The reason that I like the amp/DAC is that it introducing a distortion that I happen to like?

Yes with the proper headphones/eaphones the iPhone should produce excellent results. And is it more likely that amp/DAC is producing less distortion with your head/earphones than with the iPhone alone. And the distortion when using just the iPhone is being caused by a poor match between the iPhone and your head/earphones and not because the measurements of the iPhone are somehow "wrong".

Yes could well be.Can you detail the distortion characteristics that give rise to this flattened sound? What exactly is the flattening, non musical sound that Krismusic is describing - can you put some meat on this bone?

 

[sonitus mirus]

Sound stage, with regards to headphones, seems to be a myriad of different things to people.  I don't particularly hear much of a difference at all and care very little about sound stage, while some use this as an important deciding factor in what to purchase.  It might be part frequency response, delay, isolation, ear shape, and fit.  When people think they are hearing a subtle difference, they often attribute this to sound stage.  I simply gloss over any review that uses this feature to describe a pair of headphones.  It is a practically meaningless attribute to me.  Since I listen to speakers almost all of the time, maybe a real sound stage makes it difficult to overcome whatever it is that headphones seem to resemble to certain folks.

 

[mmerrill99]

Sound stage, with regards to headphones, seems to be a myriad of different things to people.  I don't particularly hear much of a difference at all and care very little about sound stage, while some use this as an important deciding factor in what to purchase.  It might be part frequency response, delay, isolation, ear shape, and fit.  When people think they are hearing a subtle difference, they often attribute this to sound stage.  I simply gloss over any review that uses this feature to describe a pair of headphones.  It is a practically meaningless attribute to me.  Since I listen to speakers almost all of the time, maybe a real sound stage makes it difficult to overcome whatever it is that headphones seem to resemble to certain folks.

Is this "flattening" that Krismusic mentions, a flattening of soundstage? Maybe Kris could tell us?

 

[castleofargh]

I see many ideas and problems/conditions in the last posts.
1/ can we count soundstage as real when the way it's manufactured in music records has little to do with the real position of the instruments while recording? most of the job is done with more or less simple panning while mixing. so is there a point in trying to judge the realism of a manufactured effect?
2/ as headphones output the sound wrongly(left sound not reaching the right ear like any real world sound should, no body refraction, non flat signature...), is the perceived soundstage/headstage we get from them, related to the original soundstage the sound engineer created? or isn't it mostly our brain trying to create a soundstage out of conflicting cues that shouldn't make a soundstage?
3/ are we sure that because 10 people find the soundstage on the hd800 to be bigger than on a hd600, all 10 people actually perceived the same magnitude of increase in soundstage? or the same soundstage at all?(see my topic about me feeling like frontal sounds go up on headphones, obviously not everybody feels that way).
4/ my point about soundstage in my previous post was(I guess ^_^) that it's not a sonic characteristic, but our interpretation of sonic characteristics. and as soon as interpretation is involved, the result may be subjective and hard to quantify. but even if we could quantify it, it would still, IMO not be an audio characteristic.
5/ if I can hear a difference can I make a measurement that will show it? my opinion on this is yes 100%(with proper measurement gear). we may not always be able to identify the cause of the difference, but we will measure "a" difference. given the resolution of devices and the resolution of the ear, I can't think of a situation where we would fail to show a difference in the output sound while hearing it.

 

[mmerrill99]

I see many ideas and problems/conditions in the last posts.
1/ can we count soundstage as real when the way it's manufactured in music records has little to do with the real position of the instruments while recording? most of the job is done with more or less simple panning while mixing. so is there a point in trying to judge the realism of a manufactured effect?

Are you saying that soundstage in recordings is ALWAYS manufactured? I doubt you mean this.

2/ as headphones output the sound wrongly(left sound not reaching the right ear like any real world sound should, no body refraction, non flat signature...), is the perceived soundstage/headstage we get from them, related to the original soundstage the sound engineer created? or isn't it mostly our brain trying to create a soundstage out of conflicting cues that shouldn't make a soundstage?

Is binaural recording & headphone TRICK or an attempt at recording in a way that better captures the signals that would normally reach both ears if listening to the "real event"?

3/ are we sure that because 10 people find the soundstage on the hd800 to be bigger than on a hd600, all 10 people actually perceived the same magnitude of increase in soundstage? or the same soundstage at all?(see my topic about me feeling like frontal sounds go up on headphones, obviously not everybody feels that way).

Does it matter if each of the10 perceives the "same magnitude of increase in soundstage"? Is this not trying to evaluate us a measuring devices & asking if we are calibrated the same?

4/ my point about soundstage in my previous post was(I guess ^_^) that it's not a sonic characteristic, but our interpretation of sonic characteristics. and as soon as interpretation is involved, the result may be subjective and hard to quantify. but even if we could quantify it, it would still, IMO not be an audio characteristic.

By "not an audio characteristic" do you mean that it's a figment of our imagination & not based on some characteristics contained in the signal stream?

5/ if I can hear a difference can I make a measurement that will show it? my opinion on this is yes 100%(with proper measurement gear). we may not always be able to identify the cause of the difference, but we will measure "a" difference. given the resolution of devices and the resolution of the ear, I can't think of a situation where we would fail to show a difference in the output sound while hearing it.

OK. Can you show a set of such measurements that define a real (not studio manufactured) soundstage?

 

[ralphp@optonline]

Yes could well be.Can you detail the distortion characteristics that give rise to this flattened sound? What exactly is the flattening, non musical sound that Krismusic is describing - can you put some meat on this bone?

Flattened sound as I understand it would be simple treble roll off, no more, no less. No magic. And treble roll off can be easily measured. No magic.
 
Edit: the original post said: "sounds flat and lifeless" - mmerrill99 said "flattened". Flat and lifeless indicates some treble roll off. No magic needed.
 
And please stop conflating the part of the audio that happens between the listener and the speaker/headphone with what happens in the electrical audio equipment, in other words, what happens within the equipment up until the final output of the speakers. As I said before, one is a complex acoustical phenomena and the other is well understood electronics. And there is no magic in either one.
 
Mod Edited : to remove unnecessary addendum

 

[castleofargh]

1/outside of binaural and maybe a few special cases using particular techniques, the very very vast majority of instruments are recorded in mono. even when several microphones are used they are usually used for purpose unrelated to making a stereo recording. not that we couldn't do so, just that most sound engineers decide not to focus on that.
 
2/ binaural is the best chance we have at actual positioning with headphones. there can still be problems with the headphone's signature, distortions, and individual HRTF of the listeners, but yes that corrects the biggest problem of headphones when it comes to reproducing a real band in a real space.
 
3/ yes. as we're talking about perceived soundstage. and it will be an individual interpretation, we need to measure the human and see how much variations can occur from one dude to the other. and that's not sound anymore. we're on the subjective side of things where everything is complicated because people are different and senses aren't used independently by the brain, and past experiences impact everything. IDK how much variations we can get from people to people. stuff on the left will still feel on the left, reverb like we're in the toilet will still feel like a small room, and other stuff like that, but as far as I know not everybody is as good at making 3D visualizations, maybe that has an influence too for all I know. the human element just adds too much variables that have nothing to do with sound.
 
4/both. of course it depends on the sound stimuli, but it also depends on what the person will actually hear, plus like any subjective notion, the final result can be altered with external factors. imagining myself in front of the guys, instead of just waiting to feel like they're here in my room, closing my eyes, or looking at the DVD of the live. all those IMO have the potential to alter my perception of the soundstage, in conjunction with the actual sound. so do I measure that? no, this is one more stuff we leave to subjectivity and it goes beyond the sound itself.
 
5/ I don't get what you're asking for? I'm talking about audible variations that can be measured. if a guy listens to 2 amps or whatever and gets a change in soundstage(even in a blind test ^_^), then I'm confident that we can measure a difference between both outputs.

 

[mmerrill99]

Yes could well be.Can you detail the distortion characteristics that give rise to this flattened sound? What exactly is the flattening, non musical sound that Krismusic is describing - can you put some meat on this bone?

Flattened sound as I understand it would be simple treble roll off, no more, no less. No magic. And treble roll off can be easily measured. No magic.

Edit: the original post said: "sounds flat and lifeless" - mmerrill99 said "flattened". Flat and lifeless indicates some treble roll off. No magic needed.

OK, so knowing the output impedance of the iPhone 6 & the headphones impedance & sensitivity, it should be possible to predict what headphones will be flat & lifeless, no? Care to take a shot at those headphone values that won't work with the iPhone & then let's see if Kris's headphones match that criteria? It will also help him choose headphones which will work with the iPhone 6.
After all, what you have just said is a testable claim so let's test it.

please stop conflating the part of the audio that happens between the listener and the speaker/headphone with what happens in the electrical audio equipment, in other words, what happens within the equipment up until the final output of the speakers. As I said before, one is a complex acoustical phenomena and the other is well understood electronics. And there is no magic in either one.

Mod Edited : to remove unnecessary addendum

I said this before - if this section is restricted to just discussing the electrical side of Sound reproduction then the section is incorrectly named. I didn't see this sign over the door but correct me if I'm wrong.

 

[mmerrill99]

1/outside of binaural and maybe a few special cases using particular techniques, the very very vast majority of instruments are recorded in mono. even when several microphones are used they are usually used for purpose unrelated to making a stereo recording. not that we couldn't do so, just that most sound engineers decide not to focus on that.

Really? I believe you are overstating things way too much here. I'm pretty sure that there are a vast number of recordings which use room miking. Maybe someone with professional recording experience can chime in here?

Correct me if I'm wrong but what you seem to be maintaining here is that room ambience & soundstage are solely manufactured in the recording studio & not part of the original recording?

2/ binaural is the best chance we have at actual positioning with headphones. there can still be problems with the headphone's signature, distortions, and individual HRTF of the listeners, but yes that corrects the biggest problem of headphones when it comes to reproducing a real band in a real space.

So you're original point - that "isn't it mostly our brain trying to create a soundstage out of conflicting cues that shouldn't make a soundstage?" - seems incorrect to me. Surely some of the techniques used in binaural are equally applicable to standard recordings thus producing a more realistic soundstage? It's not therefore a confusion of the brain to conflicting cues?

If what you meant was that in using headphones, the soundstage stays fixed when we move our head & this is unnatural & somewhat confusing, then I would agree but it's not the reason for hearing a soundstage in the first place. I would also agree that headphone soundstage can be less natural than speaker soundstage but again I don't believe that it's the confusion of the brain that is creating the soundstage in the first place.

3/ yes. as we're talking about perceived soundstage. and it will be an individual interpretation, we need to measure the human and see how much variations can occur from one dude to the other. and that's not sound anymore. we're on the subjective side of things where everything is complicated because people are different and senses aren't used independently by the brain, and past experiences impact everything. IDK how much variations we can get from people to people. stuff on the left will still feel on the left, reverb like we're in the toilet will still feel like a small room, and other stuff like that, but as far as I know not everybody is as good at making 3D visualizations, maybe that has an influence too for all I know. the human element just adds too much variables that have nothing to do with sound.

Well, here's an interactive demonstration of soundstage construction & the tradeoff between ITD & ILD. Maybe you can test yourself to see if you hear what you are supposed to hear? I doubt any subjective imagination will add complexity to an individual's perception of left/right up/down?

4/both. of course it depends on the sound stimuli, but it also depends on what the person will actually hear, plus like any subjective notion, the final result can be altered with external factors. imagining myself in front of the guys, instead of just waiting to feel like they're here in my room, closing my eyes, or looking at the DVD of the live. all those IMO have the potential to alter my perception of the soundstage, in conjunction with the actual sound. so do I measure that? no, this is one more stuff we leave to subjectivity and it goes beyond the sound itself.

OK, I'm finding it hard to keep up with what I perceive as your changing viewpoint on this. First you said "usually people decide that sound is very complex and full of unknowns, because they project human constructs like pleasure or even more ludicrous like quantifying a perceived soundstage, and hope for a machine to tell them all of it in a simple graph... except that has very little to do with the audio signal and a lot to do with the person listening." Then "it's not a sonic characteristic, but our interpretation of sonic characteristics. and as soon as interpretation is involved, the result may be subjective and hard to quantify." And now the above " it depends on the sound stimuli, but it also depends on what the person will actually hear,"

So, again, correct me if I'm wrong but what you seem to be saying in this & your points 1-3 above is that soundstage has largely nothing to do with any signals on the recording - it's mostly a construct of the listener? Again, let me be clear here & get this straight - all our auditory perceptions are a construct of the brain - what I believe you mean here is that perceived soundstage is very loosely based on weak characteristics within the signal stream & mostly enhanced by our imagination to arrive at a perceived soundstage which bears little resemblance to what the signal stream would justify? Am I right in this restatement of your position?

5/ I don't get what you're asking for? I'm talking about audible variations that can be measured. if a guy listens to 2 amps or whatever and gets a change in soundstage(even in a blind test ^_^), then I'm confident that we can measure a difference between both outputs.

Well, I mostly see stated that soundstage differences are not a measurable entity & yours was one of the few statements that bucked this trend so I was interested in some examples of these measurements, if you have any. I have not seen any such examples ever produced by anybody up to now so it piqued my interest that you have such confidence in there being a measurement that will show this.

 

[gregorio]

Does anyone here know the details of this "highly complex" relationship calculation? Yes? Please explain it.

 
Simply explained, the signal is put through a K-weighted filter and the RMS of the result is measured. The important part is the k-weighted filter, which was designed by the ITU after compiling and averaging many decades of studies. The thing to remember is that it's a relative rather than an absolute measurement of loudness and even then, as it's based on a broad average, it may or may not align with any particular individual's perception. Furthermore, it only works under certain conditions, in relatively small rooms.
 

Are you saying that soundstage in recordings is ALWAYS manufactured? I doubt you mean this.

 
Why do you doubt this? Soundstage is always manufactured.
 

By "not an audio characteristic" do you mean that it's a figment of our imagination & not based on some characteristics contained in the signal stream?

 
Effectively yes, it's a figment of our imagination and not based on a property or characteristic contained in the sound waves.
 

OK. Can you show a set of such measurements that define a real (not studio manufactured) soundstage?

 
No, you're not getting it, there is no such thing as a "real" soundstage. Soundstage is a human concept, an artificial construct of the brain, a perception. It's not a physical property of the sound waves! What you are asking for is akin to asking for a set of measurements which define a real dragon. The only way we could create a set of measurements for a dragon would be to get a bunch of people to draw a dragon next to something of a known size, say a car, then work out the dimensions of each of the drawn dragons and finally calculate an average. Then we would have a set of measurements. This is, in effect, how we measure loudness. Soundstage is a more tricky problem though, because there's not even consensus on what soundstage actually is, and until we have that, we can't even start to think about a way of creating an averaged model to measure it. That's why I used loudness as an example, there's general consensus over what loudness is, although not general consensus over exactly what loud is and that's why we can only model relative loudness and not absolute loudness.
 
You are going to struggle to grasp these concepts until you understand that perception and reality are two different things which are either unrelated or related in a complex way which varies between individuals. Our brains create complex perceptual structures or models, the functional design of which is allow us to make sense of the world around us. The mistake made by many is a failure to understand this principle design goal and instead believe that the design goal of perception is to represent reality as accurately as possible and there are some/many who don't even believe there is any difference between perception and reality. The difficulty facing these latter groups of people is that perception is all they have ever experienced, reality cannot be experienced and therefore we can't compare reality and perception, except indirectly, through knowledge and understanding of the concepts, rather than through physical experience. At it's heart this is what science is; in an attempt to model reality, science attempts to separate reality from perception. Perception is easy, so easy we barely have to think about it and so common we invent labels to describe shared perceptions but if we want to really understand what's going on, we only have two choices; accept that reality and perception are two different things or invent some "magic" in an attempt to justify that they're the same thing! Science dictates that sound waves only have two intrinsic properties, amplitude and frequency. What we perceive when we listen is only loosely related to these two intrinsic properties, sometimes not related at all or related so complexly it's near impossible to work out if there's any relationship and because it's a perception rather than reality it can vary significantly from person to person.
 
Much of what is attributed to properties of sound waves does not exist in reality, regardless of how trivially easy to accomplish or how commonly shared those perceptions are. We've already mentioned loudness and soundstage, to these we can add a whole host of other commonly shared and labelled perceptions such as; pitch, musicality and even "music" itself, to name just a few, none of which exist in reality! Audiophiles are commonly unable to understand this difference, relying solely on personal experience rather than on knowledge/understanding and that's why the response when trying to justify personal experience against the science must ultimately must come down to "magic". Of course the response to that accusation is always along the lines of; It's only magic because science isn't yet able to measure properties beyond just amplitude and frequency but when it does, it will cease to be magic and it will become science. Unfortunately for audiophiles, we are not talking about the type of theoretical scientific model which we know is flawed but is just the best we currently have, we are talking about a precise, proven mathematical understanding of sound waves which has been around for nearly 200 years. So in this case, for magic to become science, this mathematical proof would have to be proved incorrect, a feat which nearly 200 years worth of the world's top mathematicians have failed to achieve. There's additionally a simple logical proof, borne out of the practical application of the mathematical proof; As amplitude and frequency are the only properties of sound waves we know about, they are the only properties of sound waves which we are able to measure and therefore record. In other words, if there is some other "magic" in there, we can't record (or reproduce) it! So, as this "magic" cannot and does not exist in the recordings these audiophiles are listening to, the only logical conclusion is that if it exists at all, it must exist somewhere other than the recording and the only logical place that could be is in their perception!
 
G

 

[ralphp@optonline]

  Much of what is attributed to properties of sound waves does not exist in reality, regardless of how trivially easy to accomplish or how commonly shared those perceptions are. We've already mentioned loudness and soundstage, to these we can add a whole host of other commonly shared and labelled perceptions such as; pitch, musicality and even "music" itself, to name just a few, none of which exist in reality! Audiophiles are commonly unable to understand this difference, relying solely on personal experience rather than on knowledge/understanding and that's why the response when trying to justify personal experience against the science must ultimately must come down to "magic". Of course the response to that accusation is always along the lines of; It's only magic because science isn't yet able to measure properties beyond just amplitude and frequency but when it does, it will cease to be magic and it will become science. Unfortunately for audiophiles, we are not talking about the type of theoretical scientific model which we know is flawed but is just the best we currently have, we are talking about a precise, proven mathematical understanding of sound waves which has been around for nearly 200 years. So in this case, for magic to become science, this mathematical proof would have to be proved incorrect, a feat which nearly 200 years worth of the world's top mathematicians have failed to achieve. There's additionally a simple logical proof, borne out of the practical application of the mathematical proof; As amplitude and frequency are the only properties of sound waves we know about, they are the only properties of sound waves which we are able to measure and therefore record. In other words, if there is some other "magic" in there, we can't record (or reproduce) it! So, as this "magic" cannot and does not exist in the recordings these audiophiles are listening to, the only logical conclusion is that if it exists at all, it must exist somewhere other than the recording and the only logical place that could be is in their perception!
 
G

Nicely stated and I fear that it will only fall on deaf ears.