I have a question, has anyone else noticed this?

[Inner Space]

Here is a study outline on what goes into the production of a phantom center. It may hold some insight into the workings behind soundstage production as well.

https://www.sfxmachine.com/docs/FixingThePhantomCenter.pdf

The Vickers paper is immaculate in theory, but mentions several times that the deleterious effects of phantom center image comb filtering aren't perceived by listeners as negatively as the theory suggests they could be.

I did some (fairly amateur) experiments on this in the 1980s, by studying listeners' head movements as music was playing. Some was filmed. Some was obviously just nodding along with the beat, or rocking out sometimes, but there was also evidence of clear and regular "sampling" motions, side to side, fore and aft, and rotational. This seemed uncorrelated to rhythm, and most often extended to what would be at least quarter wavelengths of affected frequencies.

My hypothesis was that the brain was thereby "averaging out" the spectral imbalances and performing real-time compensation. Hence the lack of perceived negative effects. Unfortunately it would take a better scientist than me to design a more definitive experiment.

 

[Axo1989]

Can you identify elements that create an illusion of movement?

I can not.

Low visual acuity or poor analysis?

I haven't had an answer to my question yet. You say you can't see something. I can see it. Do you know why you can't see something? Can you talk about it?

I imagine funny memes will help. I apologise for not realising this sooner.

 

[Axo1989]

The Vickers paper is immaculate in theory, but mentions several times that the deleterious effects of phantom center image comb filtering aren't perceived by listeners as negatively as the theory suggests they could be.

I did some (fairly amateur) experiments on this in the 1980s, by studying listeners' head movements as music was playing. Some was filmed. Some was obviously just nodding along with the beat, or rocking out sometimes, but there was also evidence of clear and regular "sampling" motions, side to side, fore and aft, and rotational. This seemed uncorrelated to rhythm, and most often extended to what would be at least quarter wavelengths of affected frequencies.

My hypothesis was that the brain was thereby "averaging out" the spectral imbalances and performing real-time compensation. Hence the lack of perceived negative effects. Unfortunately it would take a better scientist than me to design a more definitive experiment.

I've read that one a while back, it was good to read again.

You've likely experienced taking microphone readings of tones (played via stereo speakers). The comb filtering is obvious in the level fluctuations as you move the mic, quite small distances as pitch goes up and the wavelength shortens. Anyway I think the hypothesis that deep sharp cancellations are relatively inaudible has merit. Also, we can average across two ears. The idea that we augment this with head movement is also interesting. We use eye movement in a somewhat similar way to scan our environment with the higher-resolution macula.

I'm not aware of doing this for less-danceable music, but I haven't checked, movement could easily be unconscious. Video might be revelatory.

In any case, the flaws or stereo reproduction in this respect (cited by multi-channel enthusiasts like Toole and Rubinson) aren't especially egregious, for me at least.

 

[Axo1989]

How do you know if you can not measure?

Can you tell the actual measure (size in LY or something) of the universe?

Certainly, it needs SPL and FR. But you can not measure, out of this, the illusion of soundstage.

We are talking about measurements, not beliefs.

How do you measure "better soundstage" if you can not measure soundstage?

Returning to the moving circles (at post #112) as an both analogy for stereo image and an argument about the hapless credulity of "subjectivists" we can analyse the graphic in a straightforward manner. We have an animation with 330 individual frames. Blue and yellow segmented circles on a grey background. The circles appear to move left/right/up/down/inward/outward corresponding to a series of black arrows. The movement is largely illusory, the circles don't move position nearly as much as they appear to. So far so good.

The inner and outer outlines of the circles use blue/yellow/grey/black segments out of phase with the main circle segments to create the illusion of movement. The black arrows serve to reinforce the illusion and misdirect the eye away from the border treatment. You can cover the circles (using normal graphics tools or makeshift methods) to confirm that the arrows aren't otherwise functional. You can blink rapidly to interrupt the timing of the image manipulation and break the illusion to confirm that timing is critical to the illusion. And so on.

Now apply the analogy to stereo. Assume initially that we value the stereo image (no point to this discussion otherwise). Then say the image manipulation is how we create the stereo image. The parameters are quantifiable and non-mysterious (pixels, location, RGB values, time sequence, etc). So far so good. Here we have created the stereo image:

Now say you have playback systems with hardware and setup that reproduce the image well, and not so well. We can check the known parameters (see above) and evaluate any degradation. Are the borders too thick or too thin? Is the phase/timing off? Are the colours wrong? And so on. With sufficient degradation, we no longer have the illusion of movement (which is the stereo image in our analogy):

You've begged the question for a while now, asking for "a metric" and positing lack of same as a gotcha: it doesn't exist because we can't measure it. The logic is circular. And false. Of course we can measure it (in theory) but it's a complex problem (that needs to be solved in practice). You then fall back on a separate gotcha, asserting that controlled testing using human listeners is a priori untenable/inadmissible. It's a convenient argument for sure (it tends to make your assertions unfalsifiable) but not scientific.

 

[Axo1989]

It's nice to see they form a community in their painful cognitive dissonance, and decide i'm the root of their problem. ... I would guess they would welcome this guy in their club. He would surely hate me with the energy of a thousand suns ...

Finally, and I'll deal with this separately, there is a pattern of insults and ad hominems that occlude and somewhat weaken your argument. I can't speak for Darko obviously, or people in this thread, but assigning or projecting unlikely emotions isn't relevant to a logical argument. I can of course speak for myself. We disagree on some things. That doesn't preclude enjoyment of discussion. It's good to remember what's important:

 

[clearnfc]

But there are things that do exist that we cannot measure. The size of the universe, for example. Maybe measuring soundstage quality presents a similar difficulty, the brain is a key component in that it receives certain cues from speakers that help create the illusion. The brain however cannot fully "imagine" a soundstage, it needs certain pieces of the puzzle before it can fill in the blanks.

It is not beyond belief that some speakers will reproduce those cues in a way that helps the brain perceive a better soundstage (closer to what occurred at stereo pair mics) than others.

Talking about this, just to point out (this is off-topic), it goes beyond soundstage and even audio. There is no way to measure which singer is better. Which genre of music is better.

I posted this a while ago as an example of what cannot be measured. There are no formulas to calculate nor equipment to measure which is the best design for the xbox controller.

Behind the Design: Xbox Controller

How these control freaks leveled up our new wireless controller

medium.com

 

[JWAmerica]

OP should give serious consideration to the fact that of the 10% that actually own the speaker or other device, 50% are wrong-headed idiots with poor taste, abnormal preferences or other psychosomatic issues that mar their opinions.

 

[clearnfc]

OP should give serious consideration to the fact that of the 10% that actually own the speaker or other device, 50% are wrong-headed idiots with poor taste, abnormal preferences or other psychosomatic issues that mar their opinions.

I have to also point out that for every device out there, vast majority of the audio folks will not own it. And only people who actually like it enough will buy it. The rest may not like it, or may feel its good but better options or personal preferences.

Every product has its own pros and cons as well, nothing is perfect. So owners will probably say the pros outweigh the cons, not an issue etc.. to those who didnt buy, it could they feel cons affects them more than pros.

And then, we have fans too. Literally every brand has its own loyal fans, followers. Sometimes, pple buy them becuase they support the brand.

 

[JWAmerica]

Talking about this, just to point out (this is off-topic), it goes beyond soundstage and even audio. There is no way to measure which singer is better. Which genre of music is better.

I posted this a while ago as an example of what cannot be measured. There are no formulas to calculate nor equipment to measure which is the best design for the xbox controller.

Behind the Design: Xbox Controller

How these control freaks leveled up our new wireless controller

medium.com

It seems to me that Dolby Atmos and multichannel audio in general is an attempt to move beyond the limitations of stereo audio. Speaker manufacturers put a lot of effort into a pair of speakers rather than developing a larger multichannel system. I believe the future of audio is in having at least six or seven channels in the front alone. Getting consumers to adopt this sort of design is the biggest hurdle. Microsoft is similarly constrained to what consumers are willing to accept. The N64 controller was the biggest departure from conventional design until the Wii motion controllers were introduced. They gained some mainstream acceptance and set the standard for Sony and others to recreate, but still consumers prefer the traditional controller. We could have better audio right now, but would the market support it? The failure of SACD and other formats would indicate otherwise.

 

[raif71]

AdamG247 wants to create a key (let's call it owners thread's) upon which he can distinguish review measurements relevant and related data and other ones with other relevant and subjective staff so that it's possible to re direct and filter them (put content that doesn't belong to one to other). And it would be additional job for him and other admins (more in start until folks get used where something should go but leaving the option to move something always). I hope you understand and appreciate that.
If anyone has a better proposition let's hear it.

 

[TheBatsEar]

First, i appreciate you taking serious energy to formulate a reply. Sadly, you didn't push through.

The main event:

You've begged the question for a while now, asking for "a metric" and positing lack of same as a gotcha: it doesn't exist because we can't measure it. [...] Of course we can measure it (in theory) but it's a complex problem (that needs to be solved in practice).

You have reduced your previous "we can measure it" to "in theory we can measure it". You are almost there.
Ask yourself this:

1. If i give you two speaker measurements, can you tell me which has more "meters of soundstage"?
2. Is the unit of soundstage meters per instrument or just meters?

Back to the optical illusion:

The parameters are quantifiable and non-mysterious (pixels, location, RGB values, time sequence, etc).

Those parameters don't give you how fast the rings spin or that they spin at all. Maybe 15% don't see it spinning at all. If you ask them, how would they measure the illusion?

• You can not measure the optical illusion in a picture or display. You can only measure physical properties, but an illusion has none.
• All you can hope to ever achieve is statistical analysis asking lots of people, a "optical illusion intensity score". You will agree, that is not a measurement like volts, amps or ohms.

Back to soundstage:

Now say you have playback systems with hardware and setup that reproduce the image well, and not so well. We can check the known parameters (see above) and evaluate any degradation. Are the borders too thick or too thin? Is the phase/timing off? Are the colours wrong? And so on. With sufficient degradation, we no longer have the illusion of movement (which is the stereo image in our analogy):

When has that sufficient degradation occurred? Is it the same for everyone? Only a person can tell you, but that is not a measurement.

• You can not measure the soundstage in a CD or speaker. You can only measure physical properties, but an illusion has none.
• All you can hope to ever achieve is a statistical analysis asking lots of people, a "soundstage illusion intensity score". You will agree, that is not a measurement like volts, amps or ohms.

You continue to think that illusions in your head are measurable by physical properties in reality, but that is never the case. You will never be able to measure beauty, soundstage, best guitar, taste, attraction or believability of an actor in a movie.

We disagree on some things. That doesn't preclude enjoyment of discussion.

Agreed. Shouldn't need saying really, but i hope nobody has a bad day because of this.

Behind the Design: Xbox Controller

How these control freaks leveled up our new wireless controller

medium.com

It sure is interesting how they approach making something better, that is already considered very good, and that there are limits to making things better too.

BTW, I tried to replace my One X with a Series X. Came late, got nothing. One year later, still no Series X to buy.

 

[Shazb0t]

The statement about our understanding of what we perceive has to be qualified by the fact that we currently use about 10% of our brain capacity ... meaning we understand only 10% of how our brain function works. Your blanket statements about what exists and what can be measured has to take that into consideration.

This is a popular misconception that isn't true. We don't only use ~10% of our brain. We use all of our brain. Haven't you ever seen a PET scan or fMRI? Why do you think we would have evolved brains with 90% unused capacity? They take quite a bit of calories to run. This claim is easily dispelled with even the smallest amount of research.

 

[puppet]

Why do you think we would have evolved brains with 90% unused capacity?

Need room to put all da stuff I be learnin'.

 

[Spkrdctr]

I haven't had an answer to my question yet. You say you can't see something. I can see it. Do you know why you can't see something? Can you talk about it?

I imagine funny memes will help. I apologise for not realising this sooner.

That applies to all of the questions for Amir after a hot review! I think he would like it.

 

[Digby]

Please, no more Poppy posting.

 

[TheBatsEar]

Need room to put all da stuff I be learnin'.

Jokes on you, i never learn.

 

[TheBatsEar]

That applies to all of the questions for Amir after a hot review! I think he would like it.

I will now answer this most pressing question. BTW, that Lady is creepy ...

I haven't had an answer to my question yet. You say you can't see something. I can see it. Do you know why you can't see something? Can you talk about it?

I get the illusion very strongly, but i don't know the specifics of why, that is all. I can see fine.

 

[Axo1989]

Please, no more Poppy posting.

You're no fun.

 

[Axo1989]

First, i appreciate you taking serious energy to formulate a reply. Sadly, you didn't push through.

The main event:

You have reduced your previous "we can measure it" to "in theory we can measure it". You are almost there.

No, that was always my position. There was no previous different position. Way back at post #52 I said "in theory".

Ask yourself this:

1. If i give you two speaker measurements, can you tell me which has more "meters of soundstage"?
2. Is the unit of soundstage meters per instrument or just meters?

Begging the question again. Not a single metric, a complex function.

Back to the optical illusion:

Those parameters don't give you how fast the rings spin or that they spin at all. Maybe 15% don't see it spinning at all.

Are you lost here, or is this just word choice? The illusion (and its popularity as a meme) is about the circles apparently (but not actually) moving in the direction indicated by the arrows. Not about them spinning, or how fast they spin. The latter is just basic animation (yes, basic animation is an illusion too, but we know revolutions/frame and frame rate so if you are stuck back there you'll need to catch up). I'll assume you understand this and continue.

If you ask them, how would they measure the illusion?

• You can not measure the optical illusion in a picture or display. You can only measure physical properties, but an illusion has none.
• All you can hope to ever achieve is statistical analysis asking lots of people, a "optical illusion intensity score". You will agree, that is not a measurement like volts, amps or ohms.

The point was that we can precisely measure the parameters of the graphic that set up the illusion. We don't need to measure the brain activity directly (although it would be fun and likely useful if we did). The graphic works this way:

Here's frame 180, roughly half-way through the sequence. You can see the main rings obviously, with 90º segments of yellow and blue. You should also be able to see the inner and outer rings, just a few pixels wide. The circles rotate clockwise in the animation, and the inner and outer rings are shifted 45º "behind" the main segments. The inner and outer rings are opposite phase to each other, 180º. This creates the illusion of movement (notice they correspond to the arrows). The left and right rings are also opposite phase (so one "moves" up and the other down). I'll assume you can see this, and understand it.

Now imagine you are reproducing the animation but your system isn't so good. Pixel resolution isn't sufficient, frame timing is off, frames are skipped, colours are rendered wrong, and so on. Those factors can be objectively assessed, no problem (the source of illusion has measurable properties). Not one metric though, several. You may have to check them all to see where/why your system is f*cking up. Now consider this as analogous to the sonic manipulations we use to create stereo image in a recording (amplitude and timing between channels, phase manipulation to expand width and depth, frequency shaping and echo/reverb to cue depth, and so on, all the things that mimic natural sound behaviour and the way we hear it). We employ these deliberately because we know they work.

Back to soundstage:

When has that sufficient degradation occurred? Is it the same for everyone? Only a person can tell you, but that is not a measurement.

• You can not measure the soundstage in a CD or speaker. You can only measure physical properties, but an illusion has none.
• All you can hope to ever achieve is a statistical analysis asking lots of people, a "soundstage illusion intensity score". You will agree, that is not a measurement like volts, amps or ohms.

You continue to think that illusions in your head are measurable by physical properties in reality, but that is never the case.

Yes, we have to correlate the objective manipulation of the image and performance of the reproduction system to the subjective perception of the illusion (we don't care that it's exactly the same for everyone, it probably isn't). The recording-electronics-speaker-room-listener system is complex, and each element has to be characterised and measured appropriately. You appear to mix this up, assuming that there is no systematic manipulation (which we can measure precisely and directly) or that it is unrelated to the subjective human response (which we can survey). Or that we measure rooms and people the same way we measure DACs and amplifiers. Or both. All incorrect.

You will never be able to measure beauty, soundstage, best guitar, taste, attraction or believability of an actor in a movie.

This is both obvious and irrelevant to the discussion.

 

[TheBatsEar]

1. If i give you two speaker measurements, can you tell me which has more "meters of soundstage"?
2. Is the unit of soundstage meters per instrument or just meters?

Begging the question again. Not a single metric, a complex function.

What is this complex function and what is the unit of soundstage?