Audio Blind Testing

[gmgraves]

2 hours ago, STC said:

 

Bit that is not answering various demos of live sound vs recorded sound performed in concert hall where the audience couldn’t tell the difference. So what live performance should be the reference?

 

When I was a teen, Acoustic Research (AR) had a showroom on Broadway (Avenue of the Americas?) in NYC. They held live vs recorded demos several times a day. What I noticed was that these demos were carefully set up with a string quartet that played quietly without much dynamic contrast. At first, youngsters like me could always tell when the tape was playing and the quartet was playing because of the tape hiss. It was apparent to me because I could hear it clearly. Oldsters... not so much. Then they got smart and left the tape playing even when there was nothing on it so that there was tape Hiss even when the musicians played live. Then, later they did it another way. They took the musicians and a pair of AR3ax outdoors and recorded both the musicians and the speakers playing a recording of the musicians  onto a second tape, and played that in their showroom. 

 

The point is that it isn't hard to fool people if you rig the test sufficiently. AR rigged the test sufficiently to say that people couldn't tell the live from the playback through AR's speakers. Well, believe me with a nice jazz group sporting trumpets, trombones, saxes and a drum kit , they would have fooled no one.  

[gmgraves]

2 hours ago, audiventory said:

 

I don't met some researches of the same instrument recognition. But Stradivarius sound differently then cheap violin, isn't it?

 

An accurate stereo system will let you clearly hear the difference between a well-recorded Stradivarius, a Guarneri, and an Amati as well as that cheap violin and in today's world do that without fooling you that are listening to these instruments live! 

[audiventory]

2 hours ago, fas42 said:

Which means a reproduction system only has to reach a certain quality, that corresponds roughly to the live thing - and we "hear the illusion".

 

I'm agree. Objectively (measurable with some precision) ideal audio system can re-produce the sound hologram next the ears.

But it don't guarantee, that we listen same sound after the ideal reproduction.

[audiventory]

7 minutes ago, gmgraves said:

An accurate stereo system will let you clearly hear the difference between a well-recorded Stradivarius, a Guarneri, and an Amati as well as that cheap violin and in today's world do that without fooling you that are listening to these instruments live! 

 

I have some experience with oboes as oboist long long time ago. Though, I'm not sure, that currently I can exactly distinguish the oboe models. Pianos have very different sound. It is reasons, why I think, that experts can detect instrument kind even by sound.

Modern audio systems produce recorded sound very good (by distortions). I really love it. But, for me, it is not same to environmental "live" sound by wave field issues. Wave field reproduction is principial and very important thing currently in my opinion. Because, I think, it allow us to move forward next big step in HiEnd audio.

[STC]

30 minutes ago, gmgraves said:

 

When I was a teen, Acoustic Research (AR) had a showroom on Broadway (Avenue of the Americas?) in NYC. They held live vs recorded demos several times a day. What I noticed was that these demos were carefully set up with a string quartet that played quietly without much dynamic contrast. At first, youngsters like me could always tell when the tape was playing and the quartet was playing because of the tape hiss. It was apparent to me because I could hear it clearly. Oldsters... not so much. Then they got smart and left the tape playing even when there was nothing on it so that there was tape Hiss even when the musicians played live. Then, later they did it another way. They took the musicians and a pair of AR3ax outdoors and recorded both the musicians and the speakers playing a recording of the musicians  onto a second tape, and played that in their showroom. 

 

The point is that it isn't hard to fool people if you rig the test sufficiently. AR rigged the test sufficiently to say that people couldn't tell the live from the playback through AR's speakers. Well, believe me with a nice jazz group sporting trumpets, trombones, saxes and a drum kit , they would have fooled no one.  

 

Thank you. You just confirmed that without the hiss it became harder to distinguish. Will they be doing another demo with DSD recording?

[audiventory]

1 hour ago, STC said:

In nature there is no stereo sound and therefore you cannot reproduce live performance accurately with two speakers stereo. However, if you were to channel each sound to a single speaker and placed them accordingly as in live performance, it should sound as good as live performance.

 

Channel number is not base of sound hologram reproduction.

 

For design hologram reproduction system we define reproducing field precision.

 

And after we account room configuragtion/construction.

 

The design algorithm have output:

1) Channel number;

2) Channel speaker locations.

[STC]

3 hours ago, audiventory said:

 

Channel number is not base of sound hologram reproduction.

 

For design hologram reproduction system we define reproducing field precision.

 

And after we account room configuragtion/construction.

 

The design algorithm have output:

1) Channel number;

2) Channel speaker locations.

 

In short, 2 speakers are not enough for realistic live performance. 

[STC]

3 hours ago, audiventory said:

 

I'm agree. Objectively (measurable with some precision) ideal audio system can re-produce the sound hologram next the ears.

But it don't guarantee, that we listen same sound after the ideal reproduction.

 

Many known product which can do this?

[fas42]

13 minutes ago, STC said:

 

In short, 2 speakers are not enough for realistic live performance. 

 

When the replay through them is not of adequate quality, that's true. In the currently rare instances when it is, then the condition are met, for at least a percentage of the listeners - where realistic means, for example, that blinded listeners are not capable of determining what the true situation is. This also covers a wide variety of musical styles, and instruments - not just the "easy stuff".

[audiventory]

25 minutes ago, STC said:

In short, 2 speakers are not enough for realistic live performance. 

 

Depend on desired precision of sound hologram.
 

25 minutes ago, STC said:

Many known product which can do this?

I don't know ideal systems. There are issues, that I don't know: how to implement currently.

[STC]

 

 

1 minute ago, audiventory said:

I don't know ideal systems. There are issues, that I don't know: how to implement currently.

 

So you are talking about a system that you have not heard? How can you even be sure that it will sound like live performance?

 

 

 

 

[STC]

14 minutes ago, fas42 said:

 

When the replay through them is not of adequate quality, that's true. In the currently rare instances when it is, then the condition are met, for at least a percentage of the listeners - where realistic means, for example, that blinded listeners are not capable of determining what the true situation is. This also covers a wide variety of musical styles, and instruments - not just the "easy stuff".

 

Such rare occasion can happen with 2 speakers when each channel is playing distinct solo instrument. 

 

For an example, Sonny Rollins I’m an old cowhand. The drums on right and sax on the left. Two mono sound from the speakers just like two mono sound in live performance where the sax is on left and drums on the right. 

[fas42]

4 minutes ago, STC said:

 

Such rare occasion can happen with 2 speakers when each channel is playing distinct solo instrument. 

 

For an example, Sonny Rollins I’m an old cowhand. The drums on right and sax on the left. Two mono sound from the speakers just like two mono sound in live performance where the sax is on left and drums on the right. 

 

So if there is a vocalist recorded so that he is positioned midway between those two instruments, speakers - this immediately won't sound realistic? Why would this be the case?

[STC]

Simple! Ask why dialogues sound better with center speaker than the side speakers. 

 

If Tracy chapman were to sing Behind the wall in front of you, the voice is one going to your two ears. When you use two speakers to reproduce the center with two speakers, you have two Tracy chapman singing simultaneously. Can that be correct and natural? 

 

 

[audiventory]

37 minutes ago, STC said:

So you are talking about a system that you have not heard? How can you even be sure that it will sound like live performance?

 

It is not matter how it sound. It is matter acoustic wave field reproduction.

 

With ideal system we have identical "live" and "reproduced" physical impacts to ears (sound field/sound hologram close to ears).

[STC]

Are you referring to this?

 

https://en.m.wikipedia.org/wiki/Wave_field_synthesis

 

and the difficulties..

 

“For reproduction, the entire surface of the volume would have to be covered with closely spaced monopole and dipole loudspeakers, each individually driven with its own signal. Moreover, the listening area would have to be anechoic, in order to comply with the source-free volume assumption. In practice, this is hardly feasible.”

 

[audiventory]

21 minutes ago, STC said:

Are you referring to this?

 

Wave field (sound hologram) is complex oscillation of acoustic waves in each point of 3D-space.

 

Theoretically and practically (in some appoach to ideal) we can manage the oscillations in single or several points (about each ear or ears of several listeners). It is synthesis of sound field (sound hologram).

 

Currently only headphones close to implementation of playback part of such system.

 

Speakers generate many bounced acoustic rays, that distort synthesed hologram.

 

 

[fas42]

9 hours ago, STC said:

Simple! Ask why dialogues sound better with center speaker than the side speakers. 

 

If Tracy chapman were to sing Behind the wall in front of you, the voice is one going to your two ears. When you use two speakers to reproduce the center with two speakers, you have two Tracy chapman singing simultaneously. Can that be correct and natural? 

 

 

 

Of course it's not "correct and natural" - but consider whether the our hearing systems are clever enough to reconstitute the sound when we don't receive it in a straight line of fire: if Tracy happened to have a solid barrier in front of her, say 6 feet wide, and the sound of her voice had to spill around the left and right edge of that barrier to reach us, would we reject the voice as not being realistic?

[gmgraves]

13 hours ago, STC said:

 

Thank you. You just confirmed that without the hiss it became harder to distinguish. Will they be doing another demo with DSD recording?

 

I doubt it. the real Acoustic Research (the one started by Edgar Villchur and Henry Kloss) has been out of business since the mid 1970's. And I think you missed my point which was that it is possible to design a live-vs-recorded demonstration with a carefully proscribed program of live music which cannot be distinguished from the same music which has been recorded and played back over a stereo system. This doesn't mean that you can do this demonstration with just any music. AR3axs were great speakers in their day, but they were really power hungry and required at least a pair of Dynaco MarkIII amps (60 Watts each) or a MacIntosh 275 (75 Wpc) to power them. Today, a pair of refurbished AR3ax speakers would be mediocre at best.  

[fas42]

9 hours ago, audiventory said:

 

It is not matter how it sound. It is matter acoustic wave field reproduction.

 

With ideal system we have identical "live" and "reproduced" physical impacts to ears (sound field/sound hologram close to ears).

 

What all this thinking fails to take into account, is that the ear.brain is capable of compensating for a "non-ideal" wave field reproduction to a very great degree. Something way less than "perfect" is good enough, for the brain to accept the illusion - I have experienced the difference between an illusion being convincing, and failing miserably, constantly over the years; what the current thinking in audio circles says is necessary, is lacking a major chunk of understanding.

[gmgraves]

10 hours ago, STC said:

 

In short, 2 speakers are not enough for realistic live performance. 

 

Bell Labs, in the 1930's, found just the opposite to be true. But it only works with a true stereo microphone arrangement, and few recordings are made like that. A true stereo recording can be palpable and uncanny in it's image specificity; pinpointing instruments in space, providing front-to-rear depth, and image height. It's too bad that so few recordings get it right. 

[fas42]

10 minutes ago, gmgraves said:

 

Bell Labs, in the 1930's, found just the opposite to be true. But it only works with a true stereo microphone arrangement, and few recordings are made like that. A true stereo recording can be palpable and uncanny in it's image specificity; pinpointing instruments in space, providing front-to-rear depth, and image height. It's too bad that so few recordings get it right. 

 

That effect can be heard on all recordings, but an 'ideal' recording technique just makes the cues really, really obvious - and means that such a presentation will be heard on less than optimal playback. What happens as a system is improved is that more recordings yield a "3D presentation" - until there is no recording which doesn't "have space".

 

The improvements allow the more subtle cues, the ones buried in the low level detritus of conventional playback, to register properly to the ear/brain - and the sound field "unfolds", fully.

 

IOW, there are no "special recordings" - it's a continuum of quality, starting from where acoustic clues hit one loud and clear, to those where such are buried deeply - the better the system, the better the unraveling.

[gmgraves]

1 minute ago, fas42 said:

That effect can be heard on all recordings, but an 'ideal' recording technique just makes the cues really, really obvious - and means that such a presentation will be heard on less than optimal playback. What happens as a system is improved is that more recordings yield a "3D presentation" - until there is no recording which doesn't "have space".

 

 

I swore I wasn't going to respond to any more of your nonsense, but this is probably one of the most asinine assertions that you've made yet.

 

A multi-miked recording using a forest of microphones, and indeed any recording where there is a microphone and/or a tape track for each instrument, there is no imaging effect. There is no front-to-back layering nor any image height because the result of pan-potting individual instruments into place puts them all on a single plane. The effect turns out to be that of a single line of instruments stretched across the stage from right-to-left. It results in a group of solo violinists all playing together, but it doesn't sound like a string section because the violins aren't mingling together in the air, they are mingling together electronically in the mixing console, and that does NOT and can NOT sound like a string section.

Only true stereo recordings made with two coincident microphones give real, phase coherent stereo. everything else is multi-channel mono and has no real soundstage. You are wrong, again. Imaging has to be contained on the recording, and playback equipment cannot correct for the lack of it!

[fas42]

18 minutes ago, gmgraves said:

A multi-miked recording using a forest of microphones, and indeed any recording where there is a microphone and/or a tape track for each instrument, there is no imaging effect. There is no front-to-back layering nor any image height because the result of pan-potting individual instruments into place puts them all on a single plane. The effect turns out to be that of a single line of instruments stretched across the stage from right-to-left. It results in a group of solo violinists all playing together, but it doesn't sound like a string section because the violins aren't mingling together in the air, they are mingling together electronically in the mixing console, and that does NOT and can NOT sound like a string section.

Only true stereo recordings made with two coincident microphones give real, phase coherent stereo. everything else is multi-channel mono and has no real soundstage. You are wrong, again. Imaging has to be contained on the recording, and playback equipment cannot correct for the lack of it!

 

Remember, I'm always talking about an illusion - the ear/brain is taking the raw data as heard coming from the speakers, and reconstitutes a believable sound field. And it turns out that there is enough depth information caught, or added via sound manipulation for the listening mind to make sense of it - with a rig that I'm happy with I never hear a single plane from left to right - there is a stage behind the speakers, always behind; which can be small, or gigantic; in pop productions, or heavy manipulation of the instruments, there are multiple stages, of various sizes and positions, overlaid, one on top of the other - you can move one's focus to each of these layers, and see it as having full integrity, in its own right.

 

The imaging is on the recording, but there are often many layers of such, all still having their own identity. Optimised playback allows all this information to be unraveled, in one's mind - and it is a delight to see clearly the details of the "montage".

[esldude]

31 minutes ago, gmgraves said:

 

I swore I wasn't going to respond to any more of your nonsense, but this is probably one of the most asinine assertions that you've made yet.

 

A multi-miked recording using a forest of microphones, and indeed any recording where there is a microphone and/or a tape track for each instrument, there is no imaging effect. There is no front-to-back layering nor any image height because the result of pan-potting individual instruments into place puts them all on a single plane. The effect turns out to be that of a single line of instruments stretched across the stage from right-to-left. It results in a group of solo violinists all playing together, but it doesn't sound like a string section because the violins aren't mingling together in the air, they are mingling together electronically in the mixing console, and that does NOT and can NOT sound like a string section.

Only true stereo recordings made with two coincident microphones give real, phase coherent stereo. everything else is multi-channel mono and has no real soundstage. You are wrong, again. Imaging has to be contained on the recording, and playback equipment cannot correct for the lack of it!

You saved me from typing the same thing.  Thanks.