Preference vs. audibility - please keep them separate.

[j_j]

Audibility has to come before preference. My advice go out and build a preamp kit,voice the unit and then talk about audibility and preference.

Why?

We are taking a soundfield that consists of a huge number of points of information, reducing it to 2 variables (for standard stereo) and then reproducing it, using characteristics of the auditory system to help us imagine it's something better than a system missing 99.9% of the actual information in the soundfield (most of which we can not hear, that's a different issue).

Consider, even, the meaning of those two points. Were they pressure, volume velocity, part one and part the other? Even the 2 points, miked by 2 mikes in a minimalist setup, are only capturing 2/8 of the information at even those two points.

So, if something "fills in", how is this bad? Even if it's the color of the grillcloth. Seriously.

Yes, my own preference is for proper multichannel, but we're no nearer to PROPER multichannel for music than we were when the lame-brained Quadriphonics was introduced back when. Until we see a proper understanding that, as shown by Steinburg and Snow in 1933 ?34?, the center speaker is the key, we're still SOL.

 

[Gregadd]

Originally Posted by tomelex
Well, what kind of a test would you set up Greg to really establish audible differences (or not) in say two power amps?



Greg, I dont need to nor want to attack anyone. I am chuffed o hear that manufactuers listen to their own unique systems to decide how accurate or musical (take your pick) their gear will sound in my system....but thats preference, whos I have no idea..ahahahah

I was being a smartass.
You agree in an AB test the only issue should be sound. Test taking issues like memory should not be an issue, Being consistent with guessing is not guessing exclusively. If i get 6 out of ten there is no way to know whether I guessed.So we wind up with meaningless results. maybe i can't design a better test. I think somebody can.

 

[microstrip]

Been there done that no, they don't. Next?

The "Next?" was written in the original post.

 

[Phelonious Ponk]

I was being a smartass.
You agree in an AB test the only issue should be sound. Test taking issues like memory should not be an issue, Being consistent with guessing is not guessing exclusively. If i get 6 out of ten there is no way to know whether I guessed.So we wind up with meaningless results. maybe i can't design a better test. I think somebody can.

If you just guessed because you couldn't actually hear the difference, the test results will show that. There is nothing meaningless about that result. If there was an audible difference you could have heard, but you yelled nanny nanny boo boo through the playback so you couldn't hear it, then guessed randomly just to invalidate the results, you would be thrown out or not chosen in the first place, because you're four years old. You can't "guess" if you can clearly hear a difference; you can lie. In scientific testing, there are control to limit such possibilities and protocols to render them insignificant. In casual blind listening, if you want to ruin any possibility of learning anything from isolating bias from evaluation, have fun with that. An odd way to amuse yourself, but enjoy.

Tim

 

[RogerD]

Why?

We are taking a soundfield that consists of a huge number of points of information, reducing it to 2 variables (for standard stereo) and then reproducing it, using characteristics of the auditory system to help us imagine it's something better than a system missing 99.9% of the actual information in the soundfield (most of which we can not hear, that's a different issue).

Consider, even, the meaning of those two points. Were they pressure, volume velocity, part one and part the other? Even the 2 points, miked by 2 mikes in a minimalist setup, are only capturing 2/8 of the information at even those two points.

So, if something "fills in", how is this bad? Even if it's the color of the grillcloth. Seriously.

Yes, my own preference is for proper multichannel, but we're no nearer to PROPER multichannel for music than we were when the lame-brained Quadriphonics was introduced back when. Until we see a proper understanding that, as shown by Steinburg and Snow in 1933 ?34?, the center speaker is the key, we're still SOL.

What flows through the signal path....electrons,and those electrons are effected by how the circuit is designed and built,what components are used,and those components have a dramatic effect on what we hear. All this other stuff is really such a small percentage of what you hear,it really doesn't matter.

And as far as proper multi channel or even enhanced stereo,the technology is here for a long time,just not in the home environment.

There is way too much bias,too much money to be made and the status quo will continue,not to mention that most people hate change,and can't get out of the box.

 

[Atmasphere]

What do you reckon? Am I the same as every self-deluding amateur speaker builder who believes their latest chipboard monstrosity is the best thing ever, or might it just be possible that I've built something good? The individual components are nothing special, but the configuration of them is a 'paradigm shift' compared to most systems. Can you be totally sure it'll sound crap compared to your favourite turntable - valve amp - passive speakers?

There is a tremendus temptation to think that one's baby is the best. To answer the question though, no. But there are many that would think that the answer is 'yes'. Put another way, can you be just as totally sure?

Why? Just look at vinyl distortions. L+R is symmetric except for skating error. L-R is not symmetric. How will an active crossover do anything beyond send the appropriate signals to the appropriate places?

FWIW skating error is a function of certain tone arms, not vinyl. I would argue that barring a valid test you may not have an argument for L-R symmetry too. There is a huge difference between the capabilities of the media and individual experience.

 

[Gregadd]

"If you just guessed because you couldn't actually hear the difference, the test results will show that"

Let's suppose you get 6 out of 10. A score consistent with chance. How do you differentiate that from someone who was just right six times and wrong four times maybe they got tired or bored. Maybe the test was really difficult.

 

[Groucho]

Why? Just look at vinyl distortions. L+R is symmetric except for skating error. L-R is not symmetric. How will an active crossover do anything beyond send the appropriate signals to the appropriate places?

Hello j_j

I was talking about the amp distortions themselves. With an active crossover, there is a separate amplifier transfer function for each of the drivers meaning that even if an amp distorts (and they will do less because their job is now much easier than with the passive case), the result is restricted to a narrower bandwidth. In an extreme case you could have the bass distorting heavily while the mid and treble are perfect, for example. In contrast, with its heavier loading the passive crossover causes the amp to distort more, and because it's amplifying the full spectrum signal, the IMD contaminates everything.

But if the idea of euphonic distortion in amplifiers is that louder signals = increased full spectrum IMD, then this simple relationship is "scrambled" by the active crossover.

 

[Phelonious Ponk]

"If you just guessed because you couldn't actually hear the difference, the test results will show that"

Let's suppose you get 6 out of 10. A score consistent with chance. How do you differentiate that from someone who was just right six times and wrong four times maybe they got tired or bored. Maybe the test was really difficult.

Control, repetition, statistical modeling. As I believe has been said here recently, more eloquently than this, it's not about proof, it's about probability. And if you listen sighted, and conclude that the big metal boxes full of components from the respected high end manufacturer are "better" than the slim little class D amp you've never heard of, how do you know that you're not the victim of expectation bias? You don't. But the probability is very high.

Tim

 

[Phelonious Ponk]

Hello j_j

I was talking about the amp distortions themselves. With an active crossover, there is a separate amplifier transfer function for each of the drivers meaning that even if an amp distorts (and they will do less because their job is now much easier than with the passive case), the result is restricted to a narrower bandwidth. In an extreme case you could have the bass distorting heavily while the mid and treble are perfect, for example. In contrast, with its heavier loading the passive crossover causes the amp to distort more, and because it's amplifying the full spectrum signal, the IMD contaminates everything.

But if the idea of euphonic distortion in amplifiers is that louder signals = increased full spectrum IMD, then this simple relationship is "scrambled" by the active crossover.

I really enjoy being scrambled.

Tim

 

[NorthStar]

Are you JJ Johnston from Secrets?

I don't even know what that means.

Forgive me; I mistakenly thought that you could have been John E. Johnson, Jr. from Secrets of Home Theater and High Fidelity,
an online audio/video website magazine with reviews and articles. ... John E. Johnson, Jr is a very good/serious/experienced audio writer and reviewer.
And you too are a serious professional/scientific audio writer and connoisseur.

* Johnston and Johnson are two totally different words, names, and persons. ...I should have checked further first.
Again, my mistake.

 

[NorthStar]

I was thinking; can we audibly prefer inaccuracy over accurate music sound reproduction?

Also, the way electrical current carries the music flow through its channels (cables) of metal compositions (copper, silver, gold, etc.),
and the physical/chemical reactions of the electrons, neurons, & protons inside them electrical audio cables; can they transport audio signals in a better and cleaner state if the electrical source (grid) would be from a closer, cleaner and more effective source of energy, like batteries charged from solar panels or wind turbines? ...And how impervious the level of protection to external vibrations, sources of distortion, electrical pollution, atmospheric debris, etc., can it be effectively and audibly productive?

Would that have an effect on our audibility discernment's ability, and also, ultimately, on our preference's perception? ..Two very different things indeed. ...Related, or not?

 

[Atmasphere]

Hello j_j

I was talking about the amp distortions themselves. With an active crossover, there is a separate amplifier transfer function for each of the drivers meaning that even if an amp distorts (and they will do less because their job is now much easier than with the passive case), the result is restricted to a narrower bandwidth. In an extreme case you could have the bass distorting heavily while the mid and treble are perfect, for example. In contrast, with its heavier loading the passive crossover causes the amp to distort more, and because it's amplifying the full spectrum signal, the IMD contaminates everything.

The flip side of that coin is that the electronic crossover is one more block in the signal path, and one that introduces group delay/phase shift components. As a general rule of thumb a 6db slope can be expected to cause issues to 10x or 1/10th the cutoff frequency, which is why its a good rule of thumb to have full power bandwidth in an amplifier down to 2 Hz or lower to avoid phase shift in the audio passband.

So far all the actively crossed-over systems I have heard have suffered two problems- that of getting the whole system to behave as if it really is one stereo (solvable, IMO) and the other of loosing information due to the additional complexity (not solvable, IMO). As a result I prefer speakers truly capable of full-range operation, so a single amp can manage it. This seems to be a lot more transparent, and with greater impact. At the same time with such a speaker you don't want the amp working hard (distortion kept low = increased detail with smoother sound) so the match between the amp and speaker is paramount.

 

[NorthStar]

So Ralph, are you saying that an amplifier with a negative 3 dB point at 5 Hz is less than perfect in the way it can create phase shifts higher in the human/audible audio spectrum of the full frequency range?

 

[Groucho]

I was thinking; can we audibly prefer inaccuracy over accurate music sound reproduction?

Also, the way electrical current carries the music flow through its channels (cables) of metal compositions (copper, silver, gold, etc.),
and the physical/chemical reactions of the electrons, neurons, & protons inside them electrical audio cables; can they transport audio signals in a better and cleaner state if the electrical source (grid) is from a closer, cleaner and more effective source of energy, like batteries charged from solar panels or wind turbines?

Would that have an effect on our audibility's ability, and on our preference's perception? ..Two very different things indeed.

Hi Northstar

If you think about it, it's remarkable that the incidental distortions that happen to sounds that have been been converted into electrical currents, should have certain specific meanings for humans. So the insertion of a high end cable "opens up the sound stage" (= makes the recording space bigger which would normally require an architect and team of builders), "tightens the bass" (= improves the bass player's technique which would normally require payment to a qualified teacher and/or years of practice), "adds mellowness" (= transports the performance from the city to a small cottage in the country, changes the season to Autumn and introduces some gentle mist). And all these are accidental by-products of small amounts of resistance, inductance and capacitance. Why these apparently unconnected effects and not "makes the singer sound slightly impatient as though he has to rush off to an important appointment" or something equally random? Even when audio is completely scrambled into digital bits and back, remarkably similar things happen to the sound while in the alternative dimension, apparently.

 

[NorthStar]

Bob, you're a bit late to the party but indeed. As I have said before, plain old stereo is so limited, that, for some folks, additional "flesh" or what some mistake as details (more odd order harmonics) etc, can "enriched" the sound. But, so can proper mastering in the first place. IMO if one breaks away from our mental chains of thinking that somehow two channel stereo is going to replicate a live event then their set free to manipulate those two channels to their preference (or just enjoy the variety of recordings). But we all call that a preference, and sometimes, with less complex music, I prefer my SET amp. Its a very definite tone control and dynamics enhancer. But it muddles any complex music too much.

It's never too late to continuously repeat ourselves Tom. :b ...It is human nature and history.

At the beginning there was Mono, then later they added two more speakers, one on each side of that Mono speaker in the center. ...And they called it Stereo (with three speakers).
Then they deleted (banned) the center speaker because of recording issues (can you imagine three tracks on an LP); and they simply kept two channels stereo with imaging between them to recreate the lost channel with an illusion. ...Easier and economically viable that way.

Are we our own slaves somehow of our own conventional wisdom from this hobby of ours? Not me, I want to explore further in the Art of Sound. And Sound is not restricted to only two-channel Stereo, no sir.

 

[NorthStar]

Hi Northstar

If you think about it, it's remarkable that the incidental distortions that happen to sounds that have been been converted into electrical currents, should have certain specific meanings for humans. So the insertion of a high end cable "opens up the sound stage" (= makes the recording space bigger which would normally require an architect and team of builders), "tightens the bass" (= improves the bass player's technique which would normally require payment to a qualified teacher and/or years of practice), "adds mellowness" (= transports the performance from the city to a small cottage in the country, changes the season to Autumn and introduces some gentle mist). And all these are accidental by-products of small amounts of resistance, inductance and capacitance. Why these apparently unconnected effects and not "makes the singer sound slightly impatient as though he has to rush off to an important appointment" or something equally random? Even when audio is completely scrambled into digital bits and back, remarkably similar things happen to the sound while in the alternative dimension, apparently.

Indeed, remarkable.

P.S. I did add two more questions to my original post (edited), where I mentioned various forms of external distortion, in addition to the internal ones;
and as to see/hear/know/question if preference and audibility are related, or not. ...And that, before I could read your post just above.
- What you just did is to go even further than I.

 

[Groucho]

...Sound is not restricted to only two-channel Stereo, no sir.

I'd be curious if anyone else has listened to Amused to Death by Roger Waters. At the start where there is what sounds like a radio interview going on, where does that sound seem to be coming from when you listen to it on your stereo speakers?

 

[Atmasphere]

So Ralph, are you saying that an amplifier with a negative 3 dB point at 5 Hz is less than perfect in the way it can create phase shifts higher in the human/audible audio spectrum of the full frequency range?

Yes- this is well known, which is why the times 10 thing is considered an engineering rule of thumb. You will hear artifacts (loss of bass impact) up to 50Hz if the cutoff is at 5Hz. If you cut off at 20Hz (and a lot of amps do, particularly SETs, which might cut off well above that) you get artifact to 200Hz.

 

[Groucho]

The flip side of that coin is that the electronic crossover is one more block in the signal path, and one that introduces group delay/phase shift components. As a general rule of thumb a 6db slope can be expected to cause issues to 10x or 1/10th the cutoff frequency, which is why its a good rule of thumb to have full power bandwidth in an amplifier down to 2 Hz or lower to avoid phase shift in the audio passband.

So far all the actively crossed-over systems I have heard have suffered two problems- that of getting the whole system to behave as if it really is one stereo (solvable, IMO) and the other of loosing information due to the additional complexity (not solvable, IMO). As a result I prefer speakers truly capable of full-range operation, so a single amp can manage it. This seems to be a lot more transparent, and with greater impact. At the same time with such a speaker you don't want the amp working hard (distortion kept low = increased detail with smoother sound) so the match between the amp and speaker is paramount.

What about systems that use DSP to invert each driver's measured impulse response (including the amp in the measurement) thereby achieving (as close as possible) zero phase shift and a perfect impulse response? e.g. Meridian 7200