Measurements & the stereo illusion

[jkeny]

You can't separate the the psychoacoustics from the physics, the stereo illusion from the cues that are the raw material the brain works with, particularly if you design to manipulate that illusion, to stimulate perceptions. These cues are not only the foundation of the stereo illusion, they are the only things recordings, playback equipment and even room acoustics can affect; a designer cannot rewire the brain's reaction, he can only give it cues to work with. Don't understand the cues? You don't understand the psychoacoustics.

Tim

But I would put it another way - if you don't understand the psychoacoustics then you don't know what are the cues of importance & what are the ones that can be disregarded. Stereo does not provide all the cues necessary so It appears to me that a better understanding of these cues & how to manipulate them/what to pay particular attention to, would help towards a better reproduction.

 

[Phelonious Ponk]

But I would put it another way - if you don't understand the psychoacoustics then you don't know what are the cues of importance & what are the ones that can be disregarded. Stereo does not provide all the cues necessary so It appears to me that a better understanding of these cues & how to manipulate them/what to pay particular attention to, would help towards a better reproduction.

Yes, an understanding of the time and location cues stereo provides, and knowledge of how to manipulate them can help us create a better stereo illusion. A rare moment of agreement.

Tim

 

[FrantzM]

But I would put it another way - if you don't understand the psychoacoustics then you don't know what are the cues of importance & what are the ones that can be disregarded. Stereo does not provide all the cues necessary so It appears to me that a better understanding of these cues & how to manipulate them/what to pay particular attention to, would help towards a better reproduction.

What do you truly means by psychoacoustics? Are we talking about perception mechanisms or something else? We don't know it all for sure. but what we know have help us come up with 2-ch stereo. If we need more cues we have Digital Signal Processing at our disposal something we, audiophiles are very reluctant to use.
In the end it looks like we know what is needed for accurate reproduction. I think you may have to be more specific about what more we need to know. Else this looks to me like a fishing expedition ...

 

[jkeny]

Yes, an understanding of the time and location cues stereo provides, and knowledge of how to manipulate them can help us create a better stereo illusion. A rare moment of agreement.

Tim

Ah, don't be too hasty in thinking we have agreement
I believe a lot more detail than simply "time & location cues" are needed - I presume you meant level, instead of location?

 

[jkeny]

What do you truly means by psychoacoustics? Are we talking about perception mechanisms or something else? We don't know it all for sure. but what we know have help us come up with 2-ch stereo. If we need more cues we have Digital Signal Processing at our disposal something we, audiophiles are very reluctant to use.

My point is that by understanding more about how we hear what we hear, we will be in a better position to reproduce a more natural sounding audio event.

In the end it looks like we know what is needed for accurate reproduction. I think you may have to be more specific about what more we need to know. Else this looks to me like a fishing expedition ...

Yes, we have 2 channel stereo which is good enough to fool some of the people, some of the time. Is that what you consider, done & dusted with regard to 2 channel audio reproduction? I'm of the opinion that there are some final areas to focus on which will bring benefits to the stereo experience & I'm trying to investigate what they might be. I know the main message in this thread has been that the primary benefits will come from attention to speakers - I agree but it's not my area of concern - I still believe that the reproduction chain before the speakers has some possibilities for improvement. These improvements will be achieved in a faster & more accurate manner if we can better understand the exact psychoacoustic model - the cues we pay particular attention to when building the audio scene. If you want me to nominate a particular area in psychoacoustics that is still being grappled with then audio scene analysis is probably the most active. But then what would I know, I'm told I need to read the basics.

What do you mean by fishing? Fishing for what, exactly? I don't get your meaning or implication - can you elucidate?

 

[Phelonious Ponk]

Ah, don't be too hasty in thinking we have agreement
I believe a lot more detail than simply "time & location cues" are needed - I presume you meant level, instead of location?

Nope. I meant location cues. Level is one of them, perhaps the most important, but it's not alone. And, John, I don't imagine we agree on much.

Tim

 

[Phelonious Ponk]

My point is that by understanding more about how we hear what we hear, we will be in a better position to reproduce a more natural sounding audio event.

Yes, we have 2 channel stereo which is good enough to fool some of the people, some of the time. Is that what you consider, done & dusted with regard to 2 channel audio reproduction? I'm of the opinion that there are some final areas to focus on which will bring benefits to the stereo experience & I'm trying to investigate what they might be. I know the main message in this thread has been that the primary benefits will come from attention to speakers - I agree but it's not my area of concern - I still believe that the reproduction chain before the speakers has some possibilities for improvement. These improvements will be achieved in a faster & more accurate manner if we can better understand the exact psychoacoustic model - the cues we pay particular attention to when building the audio scene. If you want me to nominate a particular area in psychoacoustics that is still being grappled with then audio scene analysis is probably the most active. But then what would I know, I'm told I need to read the basics.

What do you mean by fishing? Fishing for what, exactly? I don't get your meaning or implication - can you elucidate?

Are you testing products in development with CASA studies, John?

Tim

 

[jkeny]

Are you testing products in development with CASA studies, John?

Tim

CASA studies refer to what?

 

[jkeny]

Nope. I meant location cues. Level is one of them, perhaps the most important, but it's not alone. And, John, I don't imagine we agree on much.

Tim

Right, it seems that we can agree on very generalised statements such as this - probably only because it is vague & lacking in any detail (so not really agreement then) Much the same as a statement like - we are all just chemical reactions!

 

[Groucho]

My point is that by understanding more about how we hear what we hear, we will be in a better position to reproduce a more natural sounding audio event.

I disagree with your use of the word "reproduce". Just because it sounds natural it does not follow that it is true to the original recording. Take an anechoic recording and feed it through some digital reverb processing and you can produce all manner of natural sounding audio events all derived from the same recording. Feed the original recording into some omnidirectional or dipole contraption in your listening room and you may achieve something similar. But it will just be an acoustic version of a DSP effect.

 

[jkeny]

i should think you are getting about as good stereo illusion as possible since you are using single drivers, IMO.

Yes, I agree that those speakers do a great job of imaging as do my last pair - BBC LS3/5As.

Again, what brought me to starting this thread & asking this question is my experience with changing to some source or some playback software - I can hear a distinct improvement in the overall soundstage & more emotional involvement with the music - it is just more interesting. if it was solely my experience I would wonder about this experience but it is not that unusual with others also.

Having read some more background papers I believe this is the engagement that Dreisinger is talking about so maybe it's all about better coherence in the upper harmonics. Perhaps better replay chains render the upper harmonics of bass notes down to a deeper level i.e better lower level linearity?

 

[j_j]

As I'm sure you are aware, the progress I'm talking about is in the area of psychoacoustics!

Hence my pointing you to Blauert. What do you think psychoacoustics is, anyhow?

 

[j_j]

But I would put it another way - if you don't understand the psychoacoustics then you don't know what are the cues of importance & what are the ones that can be disregarded.

Again, please get a copy of Blauert, and read it.

And, actually, 2-channel stereo does a pretty good job of creating stimuli that present the right features to the ear in terms of localization, at least within the space between the two speakers.

 

[jkeny]

Here's a good description of what I'm talking about. Received via email a few days ago.. System he is using:
Windows Vista laptop with Foobar2000
Xindak Dac-8
Don't know what speakers he is using

I have blanked out the name of the device with XX to protect the innocent

When DAC connected to computer via USB, the sound is smooth, soft, detailed, good extension on high and low freq. extremes.

When DAC connected to XX USB converter, the overall sound signature remains the same. The soundstage seems the same in size, just instruments placement become a bit more accurate and more air between them. Slightly more bass presence. What is interesting, is that while the sound signature is retained, music listening gets much more involving. And this is really a big change in a listening pleasure, not subtle! It seems it has something to do with timing change, plus microdynamics maybe.

So, I'm not a master in audio equipment reviewing, leaving alone my language limitations. But this is something I really can not easily explain: sound is almost the same (smoothness, soundstage size, detail level), but music listening is more involving and pleasurable.

Just to give you an idea: before XX converter I was planning my listening sessions in advance, waiting for the mood to sit and listen, carefully selecting the album which should fit the mood, waiting for conditions - kids are in beds, quiet late evening etc.
Now with XX converter, as soon as I put headphones on and play any of my several hundred albums, at any time of the day, - I'm get so involved that I need quite an effort to stop!

Sure, it's all his expectation bias!!

 

[Phelonious Ponk]

John quoted:

So, I'm not a master in audio equipment reviewing, leaving alone my language limitations. But this is something I really can not easily explain: sound is almost the same (smoothness, soundstage size, detail level), but music listening is more involving and pleasurable.

John said:

Sure, it's all his expectation bias!!

Agreed. He didn't hear anything...and yet he managed to imagine a superior listening experience. Yes, that has expectation bias written all over it.

Tim

 

[jkeny]

Agreed. He didn't hear anything...and yet he managed to imagine a superior listening experience. Yes, that has expectation bias written all over it.

Tim

Thanks Tim, you set my mind at ease. For a moment there I really believed there was something to this. Much appreciate your insightfulness.

 

[microstrip]

What do you truly means by psychoacoustics? Are we talking about perception mechanisms or something else? We don't know it all for sure. but what we know have help us come up with 2-ch stereo. If we need more cues we have Digital Signal Processing at our disposal something we, audiophiles are very reluctant to use.
In the end it looks like we know what is needed for accurate reproduction. I think you may have to be more specific about what more we need to know. Else this looks to me like a fishing expedition ...

Frantz,

Psychoacoustics is needed to understand the perception mechanisms and understand how a system some people consider so flawed as stereo is able to attract our preferences and be so rewarding.

And unhappily I think we still do not know exactly what is needed for accurate sound reproduction, unless you are addressing only the mechanical reproduction of air vibrations, and ignoring the perceptual aspects. We know many aspects of it, but we are still far from understanding it all. Until we do not know the role of the small cues and understand how some systems enhance the positive ones and eliminate the negative ones, DPS is useless.

EDIT: After writing the post above I found John's post:

But I would put it another way - if you don't understand the psychoacoustics then you don't know what are the cues of importance & what are the ones that can be disregarded. Stereo does not provide all the cues necessary so It appears to me that a better understanding of these cues & how to manipulate them/what to pay particular attention to, would help towards a better reproduction.

It looks like we have been reading the same books and articles!

 

[Phelonious Ponk]

Thanks Tim, you set my mind at ease. For a moment there I really believed there was something to this. Much appreciate your insightfulness.

Here to serve, John.

Tim

 

[jkeny]

Frantz,

Psychoacoustics is needed to understand the perception mechanisms and understand how a system some people consider so flawed as stereo is able to attract our preferences and be so rewarding.

And unhappily I think we still do not know exactly what is needed for accurate sound reproduction, unless you are addressing only the mechanical reproduction of air vibrations, and ignoring the perceptual aspects. We know many aspects of it, but we are still far from understanding it all. Until we do not know the role of the small cues and understand how some systems enhance the positive ones and eliminate the negative ones, DPS is useless.

EDIT: After writing the post above I found John's post:

It looks like we have been reading the same books and articles!

Indeed, some people seem to forget that the full mechanisms of perception needs to be understood in order to more accurately create an illusion that fully appeals to our senses. The mistaken & predominant approach in audio reproduction is to try to make the waveform emanating from the speakers as accurate as possible to the waveform captured by the microphone(s). This has brought us a long way to where we are today but the measurements used to achieve this has a disconnect with our perception i.e current measurements can't tell us exactly how a device will sound to us as listeners.

This situation divides opinion into those who say it's all delusion if it can't be measured & those who say more sophisticated measurements are needed to better match what we hear. Note more sophisticated doesn't necessarily mean individual measurements at a more detailed level (although this may also be required) but measurements that better reflect the model of how our sense of hearing works. Without the full detail of this model, our progress will be slow often resulting in hearing differences in sound that currently defy the "industry standard" measurements ability to show where that difference comes from. It's a rather easy & lazy statement to say that "if there's a audible difference then its measurable" - it's a simplistic statement - so simplistic that it's often used in a disingenuous way.

 

[jkeny]

Tom, your blindness is caused by the fact that you are consumed by the detail that measurements provide. Yet you fail to comprehend why we measure - in order to better understand our world & how we perceive it.
I'm not saying that it's not possible to measure things we hear, I'm saying that we may well need to connect the measurements of various elements in the the sound spectrum along a time axis & how they interrelate in order to understand differences between different reproductions & why they are obviously different to our hearing. What these parts are, what level they need to be measured down to, what the inter-relationship between them is all still needed to be worked out i.e the psychoacoustic model needs to be worked out which will direct our measurements