Message to golden-eared audiophiles posting at ASR for the first time...

[audio2design]

You raise some interesting points, chiefly that the usual measurements of THD, IM and perhaps a few more esoteric forms such as TIM may not completely explain the story, insofar as it relies on the assumption that systems are mostly linear, and the most significant errors can be captured in frequency and phase deviations occurring in amplification and transduction, the latter being moving energy from electrical to mechanical forms in speakers and phonograph cartridges (IMO transduction could also embrace conversion not just in energy but also in forms of information, that is from digital to analog waveforms).

The non-linear errors that introduce new waveforms are assumed to consist entirely of IM and THD, which can be easily measured. That they may not tell the entire story is the consistent observation that we are not annoyed by levels as high as several percent in the case of THD, and even higher when the music is loud. But there are some irritating exceptions as postulated by Geddes in this paper and tested in this one. A forum discussion can be found here and more papers on his website at this link.

I'm not at all qualified to comment on the accuracy of these claims, but they are interesting and curiously largely ignored. Whether that is because subsequent research has shown them to be of trivial magnitude and therefore inconsequential, or whether they are an inconvenient truth I can't say. I just mention this in passing as your comments stimulated my recollection of the debate, but it may be worth pursuing in your quest to identify the "missing" measurements. So have a look at Geddes papers, in particular Part 2 which discusses the GedLee (the two authors, Geddes and Lee combined) metric.

While I agreed conceptually with Earl when he released the paper in 2003, I had issue with aspects of the hypothesis then and still do, but most of all, because his paper unfortunately didn't say a lot, other than there is a problem, it needs to be addressed, and here is a formula to discuss, but with little in the way of experimental data to back it up.

BUT, let's highlight one critical thing he wrote copied below. What he is saying is if the THD, IM, and multi-tone IM is low enough, then distortion is no longer relevant. What he is approaching is a practical discussion w.r.t. distortion audibility when costs are constrained, when dealing with things like transducers that are highly distorting, etc. and trying to achieve a metric for audibility. This is likely to be much more complex than even a fairly complex equation and may be best achieved through massive experimentation and deep learning, not an easy tasks since you need both human input in a controlled setting and accurate real time distortion measurement, hence a low distortion transducer. Hard to isolate audibility of distortion in an amplifier when you are playing into transducers of often much higher distortion. Even once you have that deep learning result, as pointed out by @tvrgeek (and you thought no one was paying attention), whether that distortion results in an audible difference is source material dependent (including level), because our sensitivity to distortion changes with level, and with masking which varies with the source material.

 

[dshreter]

As I needed another DAC anyway, I wanted to get the Modius but it is out of stock, so I bought the Atom. After all, the conclusion was that is as good as it gets, and it measured better than my Asgard, it should be the same. Tossed it on. Huge difference in the specific sound that bothers me and my wife. I mean walk by the hall different. Sent the Topping back. I really wanted it to work. I mean, it was slick, every feature imaginable, fantastic specs. Loved the remote which surprised me. But the JDS flat out sounds better. So the CORRECT question, is what is different? Can we measure it? Can we quantify it, put it on a scale? Because we measure something as better by belief smaller or bigger must be better, is that true? What does our brain do to sounds based on the pattering it makes when we walk into a room and it evaluated the environment, audibly and visually? What allows us to perceive "Goldilocks"

It is extremely possible that the specific Topping unit you have or the specific JDS unit that you have deviate from the published measurements of those devices. So, there's a scenario that you hear an actual difference that exists, and that this difference is also measurable. When YOUR specific hardware has not been measured at all, I don't see the reason to leap to the conclusion that existing measurements would not be able to discern those differences.

I'm not saying the entire universe of things that are measurable has been discovered! But even with the assumption your hearing is reliable, why dismiss the existing tools that have proven effective before they have even been used?

 

[Weeb Labs]

It is extremely possible that the specific Topping unit you have or the specific JDS unit that you have deviate from the published measurements of those devices. So, there's a scenario that you hear an actual difference that exists, and that this difference is also measurable. When YOUR specific hardware has not been measured at all, I don't see the reason to leap to the conclusion that existing measurements would not be able to discern those differences.

I'm not saying the entire universe of things that are measurable has been discovered! But even with the assumption your hearing is reliable, why dismiss the existing tools that have proven effective before they have even been used?

Or it could simply be the expected result of another uncontrolled, sighted comparison.

 

[JRS]

While I agreed conceptually with Earl when he released the paper in 2003, I had issue with aspects of the hypothesis then and still do, but most of all, because his paper unfortunately didn't say a lot, other than there is a problem, it needs to be addressed, and here is a formula to discuss, but with little in the way of experimental data to back it up.

BUT, let's highlight one critical thing he wrote copied below. What he is saying is if the THD, IM, and multi-tone IM is low enough, then distortion is no longer relevant. What he is approaching is a practical discussion w.r.t. distortion audibility when costs are constrained, when dealing with things like transducers that are highly distorting, etc. and trying to achieve a metric for audibility. This is likely to be much more complex than even a fairly complex equation and may be best achieved through massive experimentation and deep learning, not an easy tasks since you need both human input in a controlled setting and accurate real time distortion measurement, hence a low distortion transducer. Hard to isolate audibility of distortion in an amplifier when you are playing into transducers of often much higher distortion. Even once you have that deep learning result, as pointed out by @tvrgeek (and you thought no one was paying attention), whether that distortion results in an audible difference is source material dependent (including level), because our sensitivity to distortion changes with level, and with masking which varies with the source material.

View attachment 179046

Thanks, that was a helpful reminder that the GedLee metric doesn't arise from thin ear, but I will be damned if I can see the relationship between the three figures in the following table:

Of course, none of the correlations were that strong, just that the combo outperformed either THD or IMD alone. So why wasn't this pursued in greater depth--it sounds as if you were following the flow of ideas closely.

 

[dshreter]

Or it could simply be the expected result of another uncontrolled, sighted comparison.

I know, and I agree. That's been thoroughly discussed.

My point is that even if that isn't the situation, even then IT STILL doesn't logically follow that the existing measurement approaches should be dismissed as insufficient.

 

[pkane]

Are double-blind tests the only way we can proceed to examine this issue? The expense and practicalities make it (almost) too easy to avoid by both sides of the argument.

Can tvrgeek offer any practical suggestions?

There's no reason to question known science, or to postulate that we don't know enough to explain what @tvrgeek is hearing. His reports go against known objective facts, and yet, he himself provides not a shred of objective evidence. His reports are nothing but opinions and anecdotes, and science knows a very simple explanation for what he's hearing: bias and lack of controls. Until @tvrgeek does something more than tell stories to substantiate his findings, there's little need for talking about measurements and the whole discussion about measurements being insufficient is way premature.

 

[Weeb Labs]

There's no reason to question known science, or to postulate that we don't know enough to explain what @tvrgeek is hearing. His reports go against known objective facts, and yet, he himself provides not a shred of objective evidence. His reports are nothing but opinions and anecdotes, and science knows a very simple explanation for what he's hearing: bias and lack of controls. Until @tvrgeek does something more than tell stories to substantiate his findings, there's little need for talking about measurements and the whole discussion about measurements being insufficient is way premature.

Agreed. Such anecdotes are extremely common but ultimately amount to begging the question. The conclusion is foregone and then an explanation is sought but while imposing far greater standards of scientific rigor than was used to reach said conclusion. GIGO.

 

[Kevbaz]

No answer to my question why some folks are here if they believe they all sound the same.

I’m here to learn more, meet like minded people and research information on here to help make my next purchasing decision on a new headphone and later speakers as I know my speakers are not good but learning if any reason to change headphones.
I have no plan to upgrade my RME adi 2 ever unless it breaks outside warranty as I believe at this level they are transparent and I would have nothing to gain.
Kev

 

[PenguinMusic]

Hi,

After reading the latest posts, the conclusion is quite simple. There are a lot of people having time to waste...
110 pages of posts that can be summed up in a few lines that would go like this more or less :

"I like this device but I do not like this other one.
- They both have the same measures.
- I know, but I hear differences.
- That cannot be.
- Maybe, but I hear a difference between those devices.
- Then your ears need to be siringed.
- Yes but I hear a difference.
- Then your testing has not been done properley.
- Yes, but I hear a difference.
- You are not a scientist.
- Do I need to be scientist to hear a difference ?
- You need to give proof that you hear a difference.
- Simple. Here is proof : I hear a difference.
- That's only proof you're stupid because you believe in things that do not exist".

Did I forget something ?

Hi,

To all those who replied this : you have been aware that I was being a little ironical, have you ?
And that it was not to be taken serioulsy (except maybe for the first sentence).

And that in no way tried to take position for one camp (the "subjectivists") or the other (the objectivists").

Regards.

 

[dkinric]

"If you choose not to decide, you still have made a choice" -Some Canadian rock trio

 

[RickSanchez]

-Some Canadian rock trio

Was it Nickelback?

/joking

 

[Spkrdctr]

I have not seen that in your posts before, but that is just me. Since it is obvious on a few recordings and not others, that pretty much tells you that it is the recording causing the problem. UNLESS, you have an intermittent problem somewhere in your equipment. If a piece of equipment is putting out a boosted 3khz signal, it should do it all the time if not broken. I think you have figured out in a rough way that it is the recordings.

Now, why do you have some that do that and others don't? Heck if I know, recording engineers do whatever they want and we are usually none the wiser. But, I have to say good catch! I would steer away from those few recordings as something was intentionally done or accidently done and it went all the way to the final customer with no one catching it, unless it was on purpose.

I have not seen that in your posts before, but that is just me. Since it is obvious on a few recordings and not others, that pretty much tells you that it is the recording causing the problem. UNLESS, you have an intermittent problem somewhere in your equipment. If a piece of equipment is putting out a boosted 3khz signal, it should do it all the time if not broken. I think you have figured out in a rough way that it is the recordings.

Now, why do you have some that do that and others don't? Heck if I know, recording engineers do whatever they want and we are usually none the wiser. But, I have to say good catch! I would steer away from those few recordings as something was intentionally done or accidently done and it went all the way to the final customer with no one catching it, unless it was on purpose.

Guys, it is his recordings. He in essence trouble shot the problem himself. It has nothing to do with DACs or any other hardware or psychoacoustic issues. Thread question is solved by the OP. Continuing to harass him about the DACs and lack of testing and other issues is unfair. We now know the answer. Once again we came through for a new guy. Just took awhile!

 

[Newman]

So have a look at Geddes papers, in particular Part 2 which discusses the GedLee (the two authors, Geddes and Lee combined) metric.

While I agreed conceptually with Earl when he released the paper in 2003,

Let’s also note that Earl has walked away from the issue — not because the paper was wrong in any major way, but because he no longer thinks it has any relevance to audio playback and gear. And that is because:-

• The least preferred harmonic patterns were artificially generated and not found in real audio gear; and
• Even if some hifi electronics has a pattern that is part way along the path to non-preferred, as long as it is even half-decently designed, the distortion levels are so far below any human detection threshold that it is moot; and
• Speakers, which undoubtedly do have higher distortion levels, never exhibit those non-preferred distortion harmonic patterns, being dominated by low order 2HD and 3HD.

So, when people from time to time bring up the Geddes work in this area as an important example of audio science not measuring the right things and needing new measurements, remember that it’s not actually an example of that. In the end, it was an interesting sidebar. For Earl, too.

Cheers

 

[Newman]

Tvrgeek posted about his own experiences (post #2101 and later) and I can understand the approach taken by subsequent replies, that (as I understand it) if you can't measure it, it does not exist outside one's own head.

IMHO that is not the main approach taken by respondents to his claims and accusations. Rather, we have mostly taken the approach that things you hear in sighted listening ‘tests’ are not valid evidence that measurements need to get better, or the wrong things are being measured. Even if it turns out to be true that measurements need to evolve or improve, that is not the way to prove it.

 

[JRS]

Let’s also note that Earl has walked away from the issue — not because the paper was wrong in any major way, but because he no longer thinks it has any relevance to audio playback and gear. And that is because:-

• The least preferred harmonic patterns were artificially generated and not found in real audio gear; and
• Even if some hifi electronics has a pattern that is part way along the path to non-preferred, as long as it is even half-decently designed, the distortion levels are so far below any human detection threshold that it is moot; and
• Speakers, which undoubtedly do have higher distortion levels, never exhibit those non-preferred distortion harmonic patterns, being dominated by low order 2HD and 3HD.

So, when people from time to time bring up the Geddes work in this area as an important example of audio science not measuring the right things and needing new measurements, remember that it’s not actually an example of that. In the end, it was an interesting sidebar. For Earl, too.

Cheers

Thank you, I have had that niggling doubt in the back of my brain for some time, and every so often it gets triggered, Back to terra firma. But the larger point is that there is likely something we are missing, how important remains to be seen. I am comfortable enough with uncertainty in science/nature that I won't be crushed if a better, more complete set of understanding evolves. be no fun otherwise.

 

[Newman]

…and I look forward to further advances in audio science. I always have.

BUT, I have gotten to the point where I can instinctively tell when someone’s carry-on, about the fact that audio science is not finished yet, is because they are waiting for research that confirms their sighted listening impressions, and until that day happens, audio science is useless and the wrong things are being measured.

 

[acbarn]

I'm quite proud that my article in Linear Audio a decade or so ago was (I think) the first to connect the non-audiophile wife in the next room and the mathematical horse. But note what the cue was for my disillusionment- not body language, but choice of music.

Yup. Whenever I start playing Steely Dan’s Aja, my wife asks, “Are you messing with the stereo system again?”

 

[Robin L]

Yup. Whenever I start playing Steely Dan’s Aja, my wife asks, “Are you messing with the stereo system again?”

This didn't happen when the wife was in the kitchen.

19 years ago, I was still convinced that LPs had a little more going on than CDs. When I demonstrated, 19 years ago, to my future wife, CD playback of Astral Weeks vs. playing back an early issue, unworn, LP copy of Astral Weeks, she was convinced that the LP was better. Under the circumstances, she was no doubt right. Of course, I already told her the LP was better. I'm sure the LP replay was louder. However, the first CD issue of Astral Weeks probably was one of those situations where a production copy was used to master the CD. Much that was wrong with the first issue on CD of Astral weeks has been rectified in later remasters. The turntable I was using at the time was quite good---Strathclyde 305m with a SME III. The LP played back via a Shure 97xe would have the treble rolled off, the net effect would be more "mellow" sounding.

In any case, when the subject comes up, she still remembers the LP playback of Astral Weeks as being better.

 

[ahofer]

just because someone claims the experience of running a two minute mile doesn't stop me from knowing that it's BS, even if I have not experienced it myself.

Exactly. And while we can’t “tell @tvrgeek what he hears”, everyone is starting with the assumption that he is human, with the auditory limitations implied by that. There just isn’t evidence of the controlled audibility by humans of differences in electronics with linear, low-distortion performance and adequate power in reasonably impedance-matched setups. So we doubt them. Just as we would all doubt the two-minute mile claim, unless you were a horse. But then we would doubt that a horse was posting on ASR.

As for ”why are we here”, I think

1) the Venn diagram posted earlier is a good start;
2)I also think it is very difficult to shop for (locate, listen, understand all features,compare) so much audio gear available these days, and ASR testing has pointed me to the least expensive electronics that are still well beyond audibly-transparent thresholds (while also sorting through the features); and
3)helps narrow down the loudspeakers/headphones I might want to audition. When I bought my desktop rig, I bought straight from ASR measurements, and I got a cheap set of comfortable headphones that EQ easily to really satisfying sound;
4)All sorts of great tips on doing room measurement and EQ, how to integrate subwoofers;
5)Cool discussion from really knowledgeable people about research in audio engineering and psychoacoustic science
6)Generally interesting group of folks;
7)Fun snarking at audio foo, like expensive cables, power cords, etc.

What’s not to like?

 

[anmpr1]

However, the first CD issue of Astral Weeks probably was one of those situations where a production copy was used to master the CD. Much that was wrong with the first issue on CD of Astral weeks has been rectified in later remasters.

In any case, when the subject comes up, she still remembers the LP playback of Astral Weeks as being better.

1) I remember buying the initial CD release of one of the Carpenter's hits LP. Whether you liked their music or not, Karen had the Bene Gesserit 'voice'. But the poorly mastered CD turned that voice into a screech! LOL

2) I'm always suspect when influencers (I refuse to call them 'reviewers', which implies they are somewhat disciplined) bring their wives into the auditory mix. There was a guy working for Stereophile that used to do it all the time. I think waifu's name was Kathleen (don't call her Kathy...). From my experience, women don't care about hi-fi. For her part, my wife would just as soon listen to her Galaxy Note than the stereo. That said, whenever we watch something 'home theater' oriented, she always checks to see if the sub is turned on. I'm not quite sure what that is about!