TIDAL is NOT Worth it! Listening Test

[Ken Newton]

That has several problems:

1. What is 'heard' in live music depends upon location in the venue and the venue. There is no singular definition of the sound of a piano or a guitar.
2. What the microphones 'hear' is not what your ears hear. They're transducers, they change the sound
3. What the microphones 'hear' almost never survives unmolested except in rare minimalist recordings. Your "live" acoustic recording has already been manipulated by mixing, EQ, compression, etc.
4. Multi-track "studio booth" recordings don't have a live performance reference point because the musicians are often not even in the same place at the same time
5. What's the reference standard for music genres that have no natural component? That are entirely electronic?

Yes, I agree, that all of that is correct. Which is exactly why I suggest including simply sounding live/real as among the selection criteria. If an reproduction is both emotionally engaging as well as sounding live/real, then, man, I've arrived. That may seem a fairly heretical notion. However, ponder how much does it really matter that a given reproduction sounds a bit different in acoustic character from the original event - which is something we cannot know anyhow, unless we were present at the session - so long as that reproduction is both emotionally engaging and live/real sounding?

Conversely, does an faithful (whatever that means, exactly) reproduction which, almost certainly, does not sound exactly like the original acoustic event either, and which carries little to no emotional message, while also not sounding real/live, somehow, better honor the intent of the artists? It doesn't better honor my human enjoyment of their music.

 

[watchnerd]

Yes, I agree that all of that is correct. Which is exactly why I suggest including simply sounding live/real as among the selection criteria. If an reproduction is both emotionally engaging as well as sounding live/real, then, man, I've arrived. That may seem a fairly heretical notion.

No, I don't think it's heretical at all.

What I don't understand is the objection to using measurements and data as part of that journey.

I work in live recording on a regular basis. The first 3/4th of the setup is done using measurements and calibration. The last 1/4th is done via listening. It's much quicker than trying to dial it all in just by ear.

I don't see how the use of tools somehow poisons the outcome.

 

[Ken Newton]

What Ken is seeking is "equipment provides the most convincing illusion of a live perfoemance made with real instruments."...

Yes, that is correct.

 

[fas42]

What changes are you referring to?

Points would have been a much better word to use, yes, than "changes" - the 5 you made.

The best audio engineers in history have tried to make things closer to high fidelity using large amounts of research, tests, budgets, etc.

Do you realistically think you can do better than the combined knowledge that is already out there? Or by ignoring it?

The audio recording engineers have done their job, well. And most audio reproduction gear is also good enough - what the "combined knowledge" has missed, so far, is taking the ability of human hearing to make up for deficiencies in that chain more seriously. The latter is why there is so much "weird stuff" out there, because people has discovered, in various ways, that you can help the process on the way by "fiddling" with things. The classic one, used by just about everyone, is "tuning" the listening room to "optimise it" - one approach, but there are others.

Of course ignoring it won't work, but improving the quality in key areas will.

 

[Ken Newton]

No, I don't think it's heretical at all.

What I don't understand is the objection to using measurements and data as part of that journey.

I work in live recording on a regular basis. The first 3/4th of the setup is done using measurements and calibration. The last 1/4th is done via listening. It's much quicker than trying to dial it all in just by ear.

I don't see how the use of tools somehow poisons the outcome.

I've never objected to that. In fact, it's what I've been consistently suggesting. Meaurements and tools are essential to undertanding how to control the technology in order to get it to deliver the desired subjective experience. My objection is to any notion that measurements exist for their own sake, somehow, validly independent of the subjective experience which they are supposed to be in service of.

The use of tools don't poison the outcome. Their unintentional misuse does. I'm an engineer by training and past experience, so I understand the desire and the need for metrics. Unfortunately, it seems to me, that too many audio products are designed more in the service of the metrics than in the service of the end user experience, which is itself inherently subjective. The problem is that the metrics are used as a proxy for the human experience, which is fully understandable from the engineer's perspective. Design outcomes have to be quantified as parameter metrics. It just seems to me that the ear-brain system has largely, but not yet fully, been translated in to audio circuit design metrics which accurately reflect the user's subjective experience.

 

[watchnerd]

Points would have been a much better word to use, yes, than "changes" - the 5 you made.

The audio recording engineers have done their job, well. And most audio reproduction gear is also good enough - what the "combined knowledge" has missed, so far, is taking the ability of human hearing to make up for deficiencies in that chain more seriously. The latter is why there is so much "weird stuff" out there, because people has discovered, in various ways, that you can help the process on the way by "fiddling" with things. The classic one, used by just about everyone, is "tuning" the listening room to "optimise it" - one approach, but there are others.

Of course ignoring it won't work, but improving the quality in key areas will.

Room tuning is part of "how to have good sound 101." That puts you in line with the objectivists.

As far as human hearing is concerned, there is an entire branch of study, psychoacoustics, dedicated to this. Modern lossy codecs depend heavily on these studies.

Again, I'm not sure what you think you've discovered that isn't already known.

 

[watchnerd]

The use of tools don't poison the outcome. Their unintentional misuse does.

Do you have a recent example of tools misuse that comes to mind?

 

[Ken Newton]

Do you have a recent example of tools misuse that comes to mind?

Yes, there's a couple of easy ones. The focus on ever reducing the THD, or on dropping noise even further beneath the threshold of audibility. Because a tool is capable of a certain fine measurement levels doesn't mean further time and money should be spent on improving those metrics. Misuse was probably a poor choice of terms. I only used it as the inverse of the term, use, which I had utilized in the prior sentence.

 

[watchnerd]

Yes, there's a couple of easy ones. The focus on ever reducing the THD, or on dropping noise even further beneath the threshold of audibility.

Is anyone actually making a big deal about THD? As you point out, it's well below audibility and I think most see it as a solved problem. The fact that it's getting lower is mostly driven by the IC makers having better miniaturization.

 

[RayDunzl]

Unfortunately, it seems to me, that too many audio products are designed more in the service of the metrics than in the service of the end user experience, which is itself inherently subjective.

An example of that is... what?

It just seems to me that the ear-brain system has largely, but not yet fully, been translated in to audio circuit design metrics which accurately reflect the user's subjective experience.

An example of something that has been (anything fully translated to accurately reflect the user's subjective experience) is... what?

 

[fas42]

Room tuning is part of "how to have good sound 101." That puts you in line with the objectivists.

As far as human hearing is concerned, there is an entire branch of study, psychoacoustics, dedicated to this. Modern lossy codecs depend heavily on these studies.

Again, I'm not sure what you think you've discovered that isn't already known.

Room tuning is a solution for some issues, in the same way concert halls are tuned - but won't compensate for more fundamental problems.

Psychoacoustics has moved on, a long way, from where most audio engineers are "stuck" - the field of Auditory Scene Analysis (ASA) is where the action is now; in recent books there are chapters about the "old stuff", and then a chapter or two explaining where ASA is taking us. Which is into the world of the human brain being cleverer than just registering FR, echos, delays, etc. But the latter thinking has been sideswiped by a Black Hole and is invisible, to people in audio land ...

 

[RayDunzl]

where ASA is taking us

Got a link to some of the advanced ASA ?

All I found (when I looked) was some old text and toot-toot examples that left me thinking "Well, that's interesting, now what?"

 

[fas42]

Ummm, I take it you've read all the links in the ASA thread here ...

 

[fas42]

BTW, here is one of the heros of the "speakers are everything" movement, discovering some things ... http://www.linkwitzlab.com/LXmini/BAF2016.htm#PowerBox

Guess he might have to be sent to the naughty boy corner now ...

 

[watchnerd]

Room tuning is a solution for some issues, in the same way concert halls are tuned - but won't compensate for more fundamental problems.

Psychoacoustics has moved on, a long way, from where most audio engineers are "stuck" - the field of Auditory Scene Analysis (ASA) is where the action is now; in recent books there are chapters about the "old stuff", and then a chapter or two explaining where ASA is taking us. Which is into the world of the human brain being cleverer than just registering FR, echos, delays, etc. But the latter thinking has been sideswiped by a Black Hole and is invisible, to people in audio land ...

Is it invisible or just too early in the research to be actionable?

What do you want audio engineers to do differently based upon what we've found in ASA studies?

 

[watchnerd]

BTW, here is one of the heros of the "speakers are everything" movement, discovering some things ... http://www.linkwitzlab.com/LXmini/BAF2016.htm#PowerBox

Guess he might have to be sent to the naughty boy corner now ...

How so? Seems like pretty normal stuff...what struck you as naughty?

 

[oivavoi]

I think Ken Newton makes some good points, though. I don't think psychoacoustics is a settled science, not even close. Here's a paper by S. Linkwitz about directivity patterns in loudspeakers: https://pdfs.semanticscholar.org/9b80/782df06556d3782b8c11968f7aa7c5ebeba4.pdf

It's pretty clear from that paper, and other papers I've read, that it isn't clear what kind of loudspeaker directivity and room acoustics are best for creating auditory illusions in the listener. Maybe there isn't even a single answer to that question. It might be that the directivity pattern of a loudspeaker is just as important as the frequency response for our musical enjoyment. Do we know whether a flat requency response matters beyond a certain point - can a frequency response be flat enough? And do we really know everything there is to know about the very fine distortion in transducers, and resonances/distortions from cabinets etc? I don't think so. In short, I think that focusing only on a flat frequency response is not close to being enough for measuring what we actually hear. If you add time domain, you're getting somewhere. If you add time domain AND directivity patterns, we're really talking. And if on top of that you add really precise measurements of driver distortion, we would probably have a almost complete picture.

(but when it comes to electronics, the science is mostly settled, methinks)

 

[Ken Newton]

Is anyone actually making a big deal about THD? As you point out, it's well below audibility and I think most see it as a solved problem. The fact that it's getting lower is mostly driven by the IC makers having better miniaturization.

Making a "big deal" is a relative notion. Actually, that's not the characterization which I had applied. Only that time and money resources were being needlessly spent in those areas. But, certainly not by everyone.

 

[fas42]

Is it invisible or just too early in the research to be actionable?

What do you want audio engineers to do differently based upon what we've found in ASA studies?

Certainly actionable - a key part of the findings is that the brain analyses what it hears, and works out what the best explanation for what it's hearing is. Once it does that it "fills in the gaps" to construct a fuller picture, inside your head; it automatically compensates for variations in the actual sound waves, because it knows, from a lifetime of experience how what it "believes is the situation" will vary as you move around, the sound making object moves around, outside noise interferes, etc. This works brilliantly until the number of "defects" in the actual sound heard which contradict the mental construction are too great - and the internal representation is rejected: you're not listening to live sound, you're listening to a hifi - is the 'new' mental picture.

What audio engineers should do is work out which anomalies in the sound matter most to this brain mechanism, and throw their effort into making those "distortions" be as low in level as possible.

 

[Blumlein 88]

I've never objected to that. In fact, it's what I've been consistently suggesting. Meaurements and tools are essential to undertanding how to control the technology in order to get it to deliver the desired subjective experience. My objection is to any notion that measurements exist for their own sake, somehow, validly independent of the subjective experience which they are supposed to be in service of.

Again you're making this up. As asked by others give some examples.

The use of tools don't poison the outcome. Their unintentional misuse does. I'm an engineer by training and past experience, so I understand the desire and the need for metrics. Unfortunately, it seems to me, that too many audio products are designed more in the service of the metrics than in the service of the end user experience, which is itself inherently subjective. The problem is that the metrics are used as a proxy for the human experience, which is fully understandable from the engineer's perspective. Design outcomes have to be quantified as parameter metrics. It just seems to me that the ear-brain system has largely, but not yet fully, been translated in to audio circuit design metrics which accurately reflect the user's subjective experience.

I think other than transducers all the problems are below audibility. So that part of the chain is a clear channel without humanly observable differences. This isn't making the experience subservient to the metrics. It means the metrics have become a non-issue with what you hear. Amplifiers are close to this though they can interact with the speaker loading in ways mostly understood.