True to life recording? - We are fooling ourselves!

[STC]

 

Meaningless.wav

 

What do you hear? You would probably hear or imagine one or two words.  I can make this meaningless noise to have a clear meaning and no matter what you do you cannot undo once I assign words to it. This is what is happening in stereo recording and how recordists believe that their recording seemed to be true to life quality when in reality that is hardly so.

 

In this forum and many other forums, it appears that people really believe that stereo recordings are true representation of life performance when recorded and playback by state of the art equipment. It is not, it just you your imagination equating them to be real but it seems that I and a few are the only odd ones to reiterate that. I cannot perceive a fictitious extra large hall in my room nor I can hear sound extending beyond the boundary of the walls. It cannot because no matter how hard I create the virtual hall ambiance or eliminate the physical cues of the speakers location, the ear/brain still knows the reality. It still detects the original point where the sounds originates from. However, setting aside the reality and once I begun to enjoy the music, it transports me to another zone to associate with the music. Some system does this better than the other but as long as you have never heard another better system what you hear is capable of giving you intended pleasure.

 

We are programmed to decode whatever sound we hear to have a meaning. Once, you associate and decode the sound it is hard to undo it from your memory bank and our brain will try its best to associate whatever sound we hear to the one we previously heard. Of course it can harder when vision is involved. In McGurk effect is a good example. It can be "Baba or FaFa" depending which word is hear last before closing your eyes.

 

 

Now listen to this and listen again to the meaningless clip. Now no matter how hard you try to forget, every word in the Meaningless clip will be heard. I guess this the problem where we really believe what we are hearing to be true.

 

meaningful.wav

 

p.s. @The Computer Audiophile, kindly allow me to moderate this thread. Thanks

[STC]

3 hours ago, firedog said:

Yep, just like the McGurk effect, once you've heard it, you can't unhear it. It's one of the reasons I think much of audiophilia is based on expectation or other types of bias. What you think you will hear, you hear, even if you aren't aware of it and even if you are sure you are being objective. Even if you consciously  think your expectations are "A", they might actually be "B". 

 

I also think this applies to high res. I don't think the difference between hi-res and the same master properly made into Redbook is large. It may be non existent. I've several times experienced hearing better detail retrieval etc on a hi-res version  and  said to myself , "wow, I've heard this music many times and never heard that before" Then I go back to the Redbook version and can hear that same thing I never heard before. It may not be quite as easy to hear, but it's there. It 's sort of the same effect as your clips. 

I think this may partially be dependent on having a system that does a really good job squeezing the last bit of info/detail out of Redbook. Not all systems do, and some may give better reproduction with hi-res, which results in the listener thinking his hi-res version is superior, when it actually is system dependent. 

 

I have previously written and given examples why A?B blind test may not be reliable for some types of differences. There are research which shows that we may fill in missing information based on our auditory bank to make the sound some sense to us.

 

AFAIK, no one could correctly identify a single sufficient with sufficient fidelity sound to be whether they are hirez, or coming from multi million dollars system. To the ears it is either it is real enough or bad. Only when you make side by side comparisons then other difference becomes evident. Once, I have randomly picked 10 different songs and played them true a SS Classe and Tube Supratek preamp. Without repletion of the song through the amp no one could score a perfect score in distinguishing whether the sound was from tube or solid state preamp despite when hearing the same song side by side showed a lot of difference. No preference was shown when heard in isolation which point what really mattered in sound reproduction is how close is to the minimum level where we perceive the sound to be real/good enough.

[STC]

11 hours ago, GregWormald said:

I don't want 'true to life'.

 

Even if the recording is 'true to life' that's going to mean either the musicians in my room or me in the venue. So I get to choose between 'too loud in order to get the right tonal balance' or sitting with venue/audience noise.

 

No thanks.

 

It need not be too loud all the time. Church choirs or a jazz performance from a comfortable distance can sound pleasant without being loud.

[STC]

5 minutes ago, fas42 said:

a good, everyday example of this is when two radio stations in the car fight for ascendancy in an area of poorer reception, they blend in with each other, but you can still hear the content of one as being distinct from the other, back and forth.

 

That's known as Cocktail Party Effect.

[STC]

8 minutes ago, fas42 said:

 

Did I say, AM, radio? The term was used by STC, I recall - and last I checked there was just endless gabbling in that area of the broadcast spectrum ... .

 

Frank, are you posting for the sake of posting? I mentioned AM radio to point out your reference audio track was mastered to be played over AM radio. 

 

And as usual, you now jumped to another subject without addressing the cocktail party effect. 

[STC]

7 hours ago, fas42 said:

 

Yes ... I'm just pointing out that such a listening 'technique' also works for highly manipulated recordings ...

 

Listening technique? Wow...you have some far fetched mind control or believe that you do.  

[STC]

20 minutes ago, fas42 said:

 

Yes, I call it the Cocktail Party Effect ... it means that I'm able to switch my attention from one source of sound in the environment, to another, when there's a mix of sounds occurring ... I'm not sure whether I'm the only one who can do this, though …

 

You didn't say "cocktail party effect" but gave an impression like some sort mind control technique where in reality it is a common occurrence and nothing to do with stereo reproduction in the context of the discussion.

[STC]

1 hour ago, fas42 said:

IOW, the soundstage is fully consistent, no matter how little or how much attention I pay - just like having real musicians in the room, you see ... .

 

1)How soundstage and real musicians in the room are related?  

 

2)Looking at some of your examples, it looks like you (fas42) could still hear soundstage ( as happened in in actual recording) with one speaker place in front and the other places at the back. True?  

 

Please give direct answers and dont hit don’t hit around the bush. 

 

 

[STC]

19 minutes ago, fas42 said:

If real musicians were in some space beyond the speakers, their music making would have a certain 'presence', create a 'vibe' in the room you were listening, irrespective of where you were sitting, etc. There wouldn't be pinpoint imaging, as beloved by audiophiles, but you would still have no trouble pointing to where each player was, have a strong sense of the location from where their contribution was coming - that's the sense of soundstaging I'm looking for, from an audio system.

 

A stereo creates soundstage by making use of two speakers’ sound level/intensity difference. You identify using the level difference between the two speakers to reconstruct the image. This image is known as phantom image because it isn't real. Technically, if you could also get one single timing difference between the two speakers then you get a more precise and sharper image; i.e. location of the performers.  

 

 In live performance, the sound from each performer is only from a single spot. Your ears only receives one level, timing, phase cues for each ear for each source. These cues will be consistent irrespective of where you are in the room. This is what distinguishes as natural and unnatural to us when we hear a sound. 

 

Unlike what you alleges, in life performance we would always able to to pinpoint accurately where the performers are because all the cues relating spatial hearing is received correctly to the ears. 

 

In a stereo playback, the position of instruments across the soundstage which is a sound field between the most extreme side where sound emerges. This is limited to the width of the two loudspeakers in a true stereo recording. 

 

It is simply impossible to claim that you could recreate the stage at any position outside the mid point of the triangle. Outside the mid point you would perceive the sound of the speaker closer to you more louder than the other which destroys the ability of stereo to fully utilize the difference of level between the two speakers to create the phantom image. 

 

It is not possible. But if you insist your brain could reconstruct the positioning information then it is definitely impossible because me and the rest of the people I know could not. Having said that, there can be some occasions where it is possible where the recordings have a very small number of instruments and the instruments are split between extreme left and right channels. For an example, the Sax could be strictly confined closer to the left speakers and a drum to the right. In such circumstances, it is generally possible to perceive a stage because you are can hear two distinct source. So in a way, what you are saying is possible for certain type of recording. 

[STC]

17 minutes ago, fas42 said:

This depends upon the intensity of the sound in the space - when the room is very strongly energised with sound, from live instruments and players, then the echos bouncing around like crazy make precision locating rather difficult,

 

Didn’t I say don’t bring what fas42 brain interprets into this discussion?  Unless, you live a room where the reflecting surface is less than 1 foot away what you said is nonsense. Sound delayed more than 1ms ( it could be lower depending on your pinna distance) does not affect localization. 

 

 

[STC]

59 minutes ago, Blackmorec said:

Its really very simple.

In real life an instrument launches a sound wave that creates a circular, expanding set of soundwaves in air. The soundwaves travel at a fixed speed, losing energy on the way as they expand and compress the air ahead.  As the soundwave reaches your ears, depending on your head’s orientation it will either reach both ears simultaneously (when the origin of the sound is at 90 degrees to both ears i.e straight ahead or straight behind or it will take longer to reach one ear than the other, in which case it will have slightly different timing, slightly different phase and slightly different amplitude. Your brain utilises the differential between the 2 signals to provide location. If there’s no difference, the sound originated from straight ahead...if it arrives first at the right ear, it’s origin is some distance off centre to the right and the same for left. 

 

That’s what Blauert says too. 

 

59 minutes ago, Blackmorec said:

What stereo does is to duplicate those same 2 signals that reach the ears from a natural sound and give them the same differential amplitude and delay as the original, so your ears can’t tell that the sounds are coming from 2 locations...rather it hears the same signal in both ears, with the same differential amplitude and delay as the signal from the single, natural source. When the 2 sets of signals (natural and stereo) reaching each ear are more or less identical, how can the brain tell if they come from a single source (as in nature) or a double, coordinated source (as in stereo). The short answer is it can’t, because all the qualities the brain requires to define location are present in both sets of signals...namely the differences caused by hearing a sound with 2 ears, placed at different locations around the head.    

 

This can be debunked easily. Take a stereo recording. Delete one channel and copy the other channel to it. Adjust the amplitude. You still hear the phantom image. 

 

Secondly, get agood microphone and record a true stereo recording of a small live ensemble. At the same time, also record the same performance using a a binaural microphone. Now play the recording and record them with the binaural microphone. Listen to both and you would realize how fake and blur the stereo playback is. Interaural crosstalk is a well researched area and you can find many research papers on them. Anyway, it is hard to describe this to someone who is yet to hear what a binaural sound over loudspeaker is. 

[STC]

5 hours ago, Blackmorec said:

What you’ll get is a centre image and NOTHING else. In other words a mono sound.

 

😂 so going by your logic, if I invert one channel then you get absolute silence because you said stereo creates one sound to the ears. 😂 😂 😂 

[STC]

10 hours ago, Blackmorec said:

Nope. Take one channel of a stereo signal, duplicate it and adjust the amplitude and play it back over a stereo system. What you’ll get is a centre image and NOTHING else. In other words a mono sound. Why? Because the entire content has the same delay and the same amplitude difference.

 

Just to prove you're wrong. Stereo can work with just level difference although timing can sound more accurate. There is no such thing as stereo creating one image. Number (5) sample shows you how the original right channel can now sound like the left channel of the original recording.

 

Here is the original 10 seconds of Money track from DSOTM album.

 

1) Listen to the "Shhhhhss" sound at the extreme right.

 

Moneyori.wav

 

 

2) Same track. But the left channel is deleted and replaced by a copy of the right channel. You will hear all sound in the middle like Mono. 

 

Moneyequal.wav

 

3) Now the same track in (2) is modified by reducing the level in the right channel. In the original track the "Shhhhhss" was heard on the right side. The same right track is now producing "Shhhhhss"  from the left side.

 

MoneyRightlevel.wav

 

 

4) Same file in (3) now added 250 microseconds delay to the right channel. Listen to this with headphones and loudspeakers. Which one is clearer and why?

 

MoneyrightD.wav

 

 

 

5) This is the most important one. Here you can see why stereo replay over loudspeakers' cannot retrieve all the information due to crosstalk. Listen with headphones and through loudspeakers. This the mono track (2) with identical copies of original right channel in both. The only difference in this there was a 250microseconds delay added to the right channel which will shift the position of the "Shhhhhss"  to the left and can be heard clearly with headphones but almost centred with loudspeakers playback.

 

Moneyequalbutdelayed.wav

[STC]

@Blackmorec  I just showed you that there need not be phase shift or timing difference for phantom image in stereo. The original Blumlein stereo microphone technique was known as intensity stereo. The only thing that mattered in stereo reproduction to create the soundstage is intensity. 

 

The only consistent cue in stereo reproduction via loudspeakers is intensity or level difference. The brain is still confused and knows a stereo production is fake because the timing and pinnae cues do not tally with the level different cues. 

 

The (5) sample clearly shows that a loudspeakers cannot produce the image shift based on timing because it is receiving two timing difference which cause the phantom image to be smeared. 

[STC]

In the case of identical signal from each speakers but one is inverted, you do not hear silence because the ears receives two different sound. However, if you were to listen to one speaker only at a time both will sound exact the same unless you the one of the rare person with sect who could hear absolute phase. 

 

When the the wind reaches the ear at the same time, what was heard as two do not merge into one image because the ears are still hearing two separate sound which provides their own ITD and ILD. However, since the sound from one speaker reaches the other by a delay of 250 microseconds, the sound which hit the ear at the exact moment from each speaker cause confusion as the two phases were different. The brain becomes confused and loses the ability to locate them. What you hear is like sound coming from every direction. 

[STC]

46 minutes ago, Blackmorec said:

BTW, a 250ms delay is massive in terms of distance, so a bit of an out of context example.

 

Who said 250ms?  

[STC]

10 minutes ago, Blackmorec said:

You know I think were are saying more or less  the same thing but mis-interpreting each other’s writing. Its rather a complicated subject to write clearly about. Let’s agree to errr agree 😁 

 

I am afraid we are not. You are equating how human ears work to stereo production. That is a myth perpetuated by audiophiles and also some audio manufacturers. Stereo is nothing more than two sound and got no relation to our hearing. If it is true that stereo contains all the information of phase, amplitude and timing that can be accurately reproduced to the ears than you will hear natural 3D sound which will sound like a recording made by you wearing a binaural microphones. Even that too wouldn’t be inaccurate because you have to place the microphone exactly where your ear drums are. 

 

 

[STC]

16 minutes ago, Blackmorec said:

The difference is that in nature every signal originates from a single points source, whereas in stereo, the signal originates from 2 closely matched sources

 

There you go again. A sound from a speaker is a sound like any other sound you hear. Whether they are from a bird tweet, drum, car, horn, siren, guitar and everything else. Even an electric guitar played live is a sound like the same coming out from a speaker. The ears function to detect sound. Any sound and the HRTF works similarly whenever any sound reaches the ears. 

 

Whenever you keep on saying “in nature” you are giving the impression that the sound from the speakers are not. They are. There is no distinction in sound perceived by us irrespective where the originates. It is absurd to think just because you hear a single sound from stereo speakers, stereo somehow manages to defy human’s hearing.  A simple 17cm AB microphones acoustic transmission of a stereo playback would provide the evidence the evidence. I have provided mine. 

 

 

[STC]

Test. @The Computer Audiophile the edit button is missing. Is it a new feature?

[STC]

3 minutes ago, Blackmorec said:

In stereo, with 2 loudspeakers we set out to fool the brain. Instead of a single sourced signal reaching both ears L & R,  with a difference in amplitude that corresponds to the geometry between loudspeaker and head, we instead send a signal to the left ear and another separate signal to the right. Our ears hear these 2 signals and see that they match each other perfectly, other than some subtle  changes in amplitude and phase.  Because WE have generated those 2 signals we can manipulate the relative amplitude of all the Signals’ elements,  such that the brain assigns different locations to each of the elements, the so called sound stage. 

 

Ah finally we are on the same wavelength. But two questions. 

 

1) what do you mean we send one signal to the left ear and another separate signal to right ear?

 

2) how the sound emerging from one speaker imposes self discipline to itself so that it doesn’t travel beyond the intended ear? Unless headphones that’s not possible. The ears will always perceive two sets of signal for one phantom image. 

[STC]

5 hours ago, The Computer Audiophile said:

Can you describe this more?

 

And this the shot of my previous post with edit button being replaced with option where you can only delete the message without the option for editing. 

 

 

[STC]

1 minute ago, The Computer Audiophile said:

You can remove posts, not edit them.

 

I mean I cannot edit my own posts. I understand it would be unethical to edit others posts but maybe you could configure that own posts could be allowed to be edited like other threads with non moderation rights. 

 

Thanks. 

[STC]

7 hours ago, Blackmorec said:

The ears will always receive (not perceive!!!) two sets of signal for one phantom image. But the psycho-acoustic phenomenon known as the precedence effect or law of the first wavefront works as follows:

 

Yes. I use presence effect all the time. From 50 microseconds to 100ms. Ambiophonics is all about understanding the precedence effect and clear understanding of it required for effective crosstalk cancellation and the exact values to create the virtual concert hall reverberation. 

 

But it is a fallacy to think you don’t perceive the second sound (5) sample clearly shows that the image is no longer shifts in stereo playback because our ears receive two sets of ITD cues unlike listening it with headphones. Either you can explain (5) or you do not. 

 

 

[STC]

1 hour ago, Blackmorec said:

Hey STC, this discussion could still finish up with us both agreeing.  

 

Regarding example 5, here’s what I’m GUESSING is going on, based on physics and logic. 

 

With headphones the actual sound sources are clamped to each ear, so the distance involved from sound source to eardrum is a couple of centimetres, and the distance is fixed,  identical L&R and with no external reflections or diffraction.

 

So; 

sound travels at approximately 322 metres a second

250 microseconds (us) is 0.000250 seconds 

In 250us sound would therefore travel 322,000 X 0,00025 cms = 8cm,  approximately the width of the back of your hand 

 

Loudspeakers are anything from 2000 to 4000 cm away from each ear and the head moves quite freely between them. 

 

With headphones, the delay in the signal in terms of distance more or less equals the distance between sound source and eardrum, so essentially with the delay, when the sound wave hits the left eardrum, the right signal is only just being generated at the headphone membrane, so the ear/brain has no difficulty in sensing the difference between the 2 channels and in assigning precedence to the first arriving signal. The ratio between delay and distance is ca 1:1 

 

With speakers at say 4000cm distant,  the ratio between distance and delay is 8/4000 is 0.002: 1  (1:500) and head movement i.e error can be larger than the delay. Further, any minute differences in speaker position vs ears will add to the error.  As soon as the error gets even close to the delay, the effect disappears (I noted elsewhere how sensitive the precedence effect is). 

 

So in essence, headphones will provide a highly controlled stable environment capable of resolving a 250us delay, whereas with speakers, the far greater latitudes of movement and therefore experimental error mean that 250us is to all intents and purposes undetectable. 

 

Is that such a problem?  Not really, because the precedence effect is only considered to work between 1 - 40ms, so minimum 4 times longer than 250us. It probably doesn’t work below 1ms for exactly the above reason.   

 

So to answer your question, due to their fixed nature and the much smaller distances involved between sound pressure wave source and eardrum, headphones have the ability to resolve far smaller (shorter) times delays between L&R channels than loudspeakers. 

 

As you said - guessing and that is pretty much incorrect. This is what crosstalk is all about. Theoretically the sound from left speaker is intended only for the left ear and the same for the right speaker was meant for the right ear. 

 

If you place a divider between the speakers right up to your head ( where you sufficiently attenuate the cross speakers sound from reaching the ear) , you will even hear a resolution of 40microseconds. I have no problem of reproducing the tearing of the paper sound from 30 degrees to the left from my loudspeakers playback.

 

 

[STC]

18 minutes ago, Blackmorec said:

I should summarise the above by asking if this difference in resolution (which is about 1 order of magnitude (2:1 vs 1:500)  between headphones and loudspeakers is important. Experimentally sure, however in the real World, the delays we are interested in i.e the ones that bestow location on musicians all fall well within AN ACCURATE and WELL SET UP loudspeaker’s ability to resolve and are reinforced by amplitude differences.  The smaller delays, ie the delays caused by the head when the L channel reaches the R ear we anyway want to ignore in terms of assigning location. 

 

 

 

 

Now, you are slowly agreeing that accurate phase is no longer relevant. The point is the stereo "ACCURATE and WELL SET UP" loudspeakers could not produce the accurate position due to crosstalk. I have another track where the male and female voice would appears to be coming from the centre but with headphones the male will be on the left  and the female on the right. No stereo system could reproduce that accurately. There goes the reality and accuracy.