John Atkinson's of Stereophile Talks About Measurements

[audiophile]

I would love to hear the definition of “trained” vs “non-trained”.

In Sean Olive's research "trained" means at least level 8 achieved with Harman's training method: https://harmanhowtolisten.blogspot.com/

 

[JoachimStrobel]

I wonder if those untrained listeners would state the same preferences if they were forced to listen for an hour or longer before specifying. I find that I'm often quite enamored of a bass-boosted, sparkly highs presentation for a few minutes... my older Klipsch were this way and my Polks (if used on axis) are as well. After more than a couple songs however... they need some toeing out or the fatigue starts to set in.

My guess is that "trained listeners" aren't just more familiar with frequency differences and how they effect presentation... they've also listened to more and for longer than their untrained counterparts... at least that would be my guess.

I share my best speculative guess:
The bass boost from the untrained listeners is almost identical to the difference of the Iso226 loudness curve between 60-90 dB. The untrained people may simply have dialed in a loudness correction. Whereas the trained guys know the loudness thing and stay away from it. The high frequency flatness can not be explained that simply by differences in loudness, but the trend is similar, keeping in mind that the various flavors of loudness curves differ substantially for that range. Alternatively, Toole stated often that professional engineers might have a hearing problem. I could guess that they are trained compensating for it, unlike the untrained group, and dial the frequencies down. I am still searching for free version of paper...
Anyhow, if had to manufacture audio equipment I knew which curve I picked...

 

[Koeitje]

I don’t like that he believes it’s wrong to pair great $500 speakers with $15,000 upstream equipment - he believes in price matched equipment.

I'm not sure he does anymore, I remember a review of some cheaper bookshelves that he ran through his main system to see what he could pull out of it. There is nothing wrong with that, and shows that a better amplifier might help. I've heard the same, I had a cheaper Cambridge Audio 540A (2x50w?) that ran out of steam when playing loud while my PM-16 (2x90w) does so effortlessly.

 

[Koeitje]

In Sean Olive's research "trained" means at least level 8 achieved with Harman's training method: https://harmanhowtolisten.blogspot.com/

This is cool stuff, I tried it real quick and am sure I'll toy around with it some more. Need to test it with headphones or my main system, because I couldn't hear the 50hz equalization on my desktop speakers .

 

[audiophile]

Reason is that he has no method to determine what is or is not good.

He uses his critical listening skills.

Everything can sound "good" if you play well recorded music on it.

That's true, but good recordings can also reveal differences in equipment. Steve often uses Chesky Records' releases, which were recorded with two microphones in a natural environment with no dynamic compression applied. He was present at many Chesky sessions when those same records where made.

There is no critical listening ability or objective measurements are involved in what he does.

If by "critical listening ability" you mean getting to at least level 8 in Harman's quick-switching ABX tests with the goal to identify various distortions, then what you say may be true. I doubt Steve ever used that test. However, the subjective evaluation approach to "critical listening" is quite different. According to my knowledge, it is best described in Robert Harley's book in the chapter called "Becoming a Better Listener", which I myself used to learn (and still keep improving) critical listening skills. I recently tried Harman's software too and got to level 7 quickly. I'd say there are some commonalities in Harman's and Harley's approaches as both teach how to isolate and analyze various aspects of sound reproduction, but with different goals: one is focused on detecting distortions, another on finding out which of the components subjectively perceived as more enjoyable to one's ears.

I have sat with a large group of high-end audio retailers in the same blind test and find them they can't tell whatsoever what is colored about the sound they are hearing. Research shows that such people don't have good listening abilities.

Very low sample size, doesn't prove what Steve G hears or doesn't hear.

Why would you put any trust in what he has to say where he can't even critically evaluate plainly audible differences in speaker sound?

No professional reviewer asks their audience to trust them blindly, quite the opposite they say to listen with your own ears. Any sane person would not buy a piece of gear based on a single reviewer's opinion. Those are subjective reviews, nothing more.

It is just that it is plainly obvious that there is no real skill involved in much of what he talks about.

This is your subjective opinion. What's seem obvious to you doesn't seem that obvious to me.

A tube power supply in a DAC powered by an old and obsolete DAC chip. Tell me why you would believe such a DAC is "crazy good".

Is all old equipment bad? I don't think so. Can it be "crazy good" to someone? Certainly.

Back to your statement, this is what he said in his written review:

What is that comparison for heaven's sake? One has a sweet sound but the other is pure? One has good soundstage depth but the other has clarity?

And oh, veils were removed as if we have not heard that before.

All of this is word salad made to make sense to reader/viewer but in reality has no substance behind it. It is fantasy description of audio based on faulty evaluations.

Steve's words make sense to me. I understand what he is talking about while describing his subjective impressions.
"Sweet sound" - that's about having soft nice treble, that sounds just right, not harsh or fatiguing
"Veils removed" - "veiling" is lack of transparency, which means lifelike immediacy that makes every detail clearly audible
"Soundstage depth" - doesn't need explanation. Everyone, whose speakers positioned properly, know this aspect of sound reproduction and understand that perception of depth can vary between different components
"Sluggish" - this is about transients being less articulated

Having said all that and being an owner of the Ares DAC I don't quite agree with everything Steve said about it. But it's totally OK for the subjective impressions to vary between people, rooms, systems, tastes, etc.

 

[xr100]

I have sat with a large group of high-end audio retailers in the same blind test and find them they can't tell whatsoever what is colored about the sound they are hearing. Research shows that such people don't have good listening abilities. From Sean Olive's research:

Why would you put any trust in what he has to say where he can't even critically evaluate plainly audible differences in speaker sound?

This is a self-selecting industry. Someone can't call themselves a doctor and dispense medicine. Yet we think it is perfectly OK for someone to dispense the same advice about audio with no proven ability to have any kind of good listening ability, or judgement about the design of a product.

On those odd occasions when I've met hi-fi reviewers in person, not all of them seemed terribly enthusiastic about audio. At the end of the day, if you're called upon to review yet another amplifier or DAC, unless there is something strange about the design or it breaks, what is there to say (in terms of how it "sounds")? More or less nothing terribly interesting, certainly nothing to gush over, so out gets trotted the same old journalistic tropes of "there was a palpable sense of" and so on.

It's like the rest of the "press" which churns out the same lines and stories over and over, with all too often an extremely lazy approach to understanding and researching the subject, or cribbing the bulk of the material straight out of press releases or "think tank" copy.

Given that (presumably--I haven't systematically checked!) there is absolutely no consistency between "subjective" reviews in one publication and another, one might as well just write the review using a "random review generator," modified up or down based upon who bought the drinks at the last trade show... even then very similar products within a manufacturer's line have ended up with widely different reviews in the same publication!

One reviewer, who did have an engineering background, categorically told me that "electronics are cheap" and basically cautioned against spending too much. They also pointed out that speaker driver from manufacturer Y was not particularly better than one from manufacturer X, even though it cost several times the price, because manufacturer X produced a much higher volume of product in a far more efficient way.

What they wrote in "print" (and it definitely was "print" in those days) was absolutely not the above.

 

[audiophile]

This is cool stuff, I tried it real quick and am sure I'll toy around with it some more. Need to test it with headphones or my main system, because I couldn't hear the 50hz equalization on my desktop speakers .

Don't get overexcited. It quickly gets boring as hell

 

[xr100]

That is so true. I play flute in a chamber group and the violin sitting next to me has a lot of "zing" to the sound, nothing like what it sounds from the audience or on a stereo, unless you crank up the treble a lot (which some recordings do!). Same with the cello and other instruments. Especially so with my own flute since the sound I hear is resonating in my own throat and skull. It's like what your own voice sounds like to you when speaking, compared a high quality recording of it.

Several years ago when I got a new flute, when trying different ones, I picked one that had more edge and zing to its tone than I like because I realized that the flutes having a softer, rounder tone actually sound muted and dead from a few feet away. What sounds to me while playing as too edgy or zingy, sounds softer and rounded just a few feet away. Or, as a friend who was with me said, "If you want be heard, get that one!".

I hear this timbral difference in recordings all the time. You can use it to tell how closely it was miced. And recordings aren't always consistent. Often, some instruments are miced closer than others. Concertos are one of the worst offenders here.

Interesting story!

My favourite example is the underdamped upper strings which resonant "sympathically." It's like an acoustic "harmonic exciter" of sorts, brightening the sound.

What, then, in principle, is the difference between using this and shelving up the HF with EQ? How does one know whether a recording is neutral or not? What does that even mean?

(EDIT: Even allowing for Floyd Toole's "circle of confusion.")

Just questions to ponder over as, of course, I'm familiar with recordings that are obviously grossly spectrally imbalanced.

Maybe we should apply an "auto EQ match" process to recordings, with a suitable "flat" reference curve (not actually flat but suitably distributed in some way) to ensure that the source material is actually "neutral!"

 

[PierreV]

"Veils removed" - "veiling" is lack of transparency, which means lifelike immediacy that makes every detail clearly audible

Here's the problem with "Veils removed"... We are regularly told power cables, dacs, preamps, interconnect, amps, speaker cables and speakers have the ability to lift veils. That's a lot of veil stacking if you ask me.

Now, I listen to 4 systems on a regular basis, two in the main room, one in my home office and one in my home theater. Obviously, there's a bunch of power cables, speaker cables, DACs, interconnects, amps and speakers involved which I fairly often rotate (experimenting is part of the hobby, isn't it?)

But, for some reason, I have never experienced the mystical veil lifting revelation that seems to one of the defining features of this hobby.

Sometimes I feel really unlucky, for all I know the veils remain at every step and I listen to the auditory equivalent of a thick fog. Sometimes I feel really lucky and believe I am blessed with a totally unveiled experience. Hard to tell, isn't it?

On the other hand, moving stuff between rooms now and then, I hear huge differences. Go figure...

PS: OK, to be 100% honest, CCA Analog out does sound veiled to me. But that's about the only thing that does.

 

[xr100]

Alternatively, Toole stated often that professional engineers might have a hearing problem.

Apparently, producer Mike Chapman (Blondie--"Parallel Lines") monitored at such high volumes that he kept on blowing up drivers. UREI's with 15" coaxials/bass drivers! It's possibly apocryphal, but many have said that he cranked it up to absurd SPL's. Heck, there was (or is?) even a YouTube video of him doing so...

 

[xr100]

Here's the problem with "Veils removed"... [snip] That's a lot of veil stacking if you ask me.

Nothing wrong with the idea of "stacking," the last thing desirable is distortion upon distortion. The only problem is that in the world of competently designed electronics there is no "veil" to lift in the first place; if there is an audible difference, it's far more subtle than that implies.

 

[audiophile]

But, for some reason, I have never experienced the mystical veil lifting revelation that seems to one of the defining features of this hobby.

In his interview on measurements, John Atkinson said he suspects that "veils" are a combination of distortion riding along with the music, resonances and noise (including modulation noise). So it's hard to say what's going on in one's particular system/room.
And actually some people may prefer a more "veiled" presentation to the more "transparent" on some recordings with seemingly overemphasized high-frequency details.

 

[Wombat]

Here's the problem with "Veils removed"... We are regularly told power cables, dacs, preamps, interconnect, amps, speaker cables and speakers have the ability to lift veils. That's a lot of veil stacking if you ask me.

Now, I listen to 4 systems on a regular basis, two in the main room, one in my home office and one in my home theater. Obviously, there's a bunch of power cables, speaker cables, DACs, interconnects, amps and speakers involved which I fairly often rotate (experimenting is part of the hobby, isn't it?)

But, for some reason, I have never experienced the mystical veil lifting revelation that seems to one of the defining features of this hobby.

Sometimes I feel really unlucky, for all I know the veils remain at every step and I listen to the auditory equivalent of a thick fog. Sometimes I feel really lucky and believe I am blessed with a totally unveiled experience. Hard to tell, isn't it?

On the other hand, moving stuff between rooms now and then, I hear huge differences. Go figure...

PS: OK, to be 100% honest, CCA Analog out does sound veiled to me. But that's about the only thing that does.

Never hear about jock-straps present or removed.

 

[vavan]

Maybe we should apply an "auto EQ match" process to recordings, with a suitable "flat" reference curve

fwiw there are tools exist allowing to do that, at least to some extent
for example stereo tool plugin offers built-in autoeq allowing to set target level for different frequencies and even can follow ITU-BS.1770 recommendations

 

[xr100]

fwiw there are tools exist allowing to do that, at least to some extent

Indeed, they are what I was referring to. IME they don't always work very well, though (particularly if the problem isn't the overall balance of the mix, but within individual tracks or instruments. For one you can't really use them with too high frequency resolution but rather an overall "smoothed" curve, for obvious reasons.)

for example stereo tool plugin offers built-in autoeq allowing to set target level for different frequencies and even can follow ITU-BS.1770 recommendations

I'm familiar with plug-ins for conformity to BS.1770 in terms of amplitude/level but not for frequency?

 

[vavan]

I'm familiar with plug-ins for conformity to BS.1770 in terms of amplitude/level but not for frequency?

here's the quote from online doc:
ITU-BS.1770
Measure levels based on ITU-BS1770.

This reduces the effect of deep bass, and increases the effect of highs above about 1000 Hz

 

[Koeitje]

I've never heard a veil being lifted unless I made huge jumps in speaker quality (especially if you start out at budget speakers). There have been some slight improvements at higher SPL levels using better amplifiers, but I wouldn't call that lifting a veil. I don't think people using the term lifting the veil have any idea how that really sounds. May I suggest putting your speakers behind your curtains and play with the curtains open and closed. Now that's a veil. If you get that effect by switching out anything other than your speakers the new or old piece of equipement was definitely broken. So we can conclude these reviewers have piles of broken electronics.

 

[Sgt. Ear Ache]

Have you measured your ears' frequency response and adjusted EQ to compensate, while also taking Fletcher-Munson equal loudness contours into account? If you didn't then what you hear is far from neutral sound. Actually, even if you did you will not hear exactly what the mastering engineer heard in his studio and considered "neutral", since his ears will measure differently from yours.

I'm 53. my ears are not super great. but it doesn't matter. I'm easily able to hear the distinct differences in sound when I switch EQ on and off, and often the differences are in mid range areas. As I've said before, there's no perfect. But I don't believe for a second that most of the people who have a preference for boosted bass or scooped mids or whatever have it because they are compensating for some problem with their hearing. I think people should really give it a try. Get the system set up so that it's providing an objectively neutral (via pink noise EQ to the degree possible as well as speaker positioning and so forth) sound and just live with it for a bit, then switch the EQ off and see how the old way sounds.

 

[Sgt. Ear Ache]

I've never heard a veil being lifted unless I made huge jumps in speaker quality (especially if you start out at budget speakers). There have been some slight improvements at higher SPL levels using better amplifiers, but I wouldn't call that lifting a veil. I don't think people using the term lifting the veil have any idea how that really sounds. May I suggest putting your speakers behind your curtains and play with the curtains open and closed. Now that's a veil. If you get that effect by switching out anything other than your speakers the new or old piece of equipement was definitely broken. So we can conclude these reviewers have piles of broken electronics.

if you actually get that effect even by switching out speakers, you had some pretty ****** speakers before.

 

[audiophile]

I'm 53. my ears are not super great. but it doesn't matter. I'm easily able to hear the distinct differences in sound when I switch EQ on and off, and often the differences are in mid range areas.

Even people who can hear full frequency range may be less or more sensitive to certain frequencies so that their hearing curve shows peaks and dips of few dBs here and there. Audiogram from Wikipedia: "Ideally the audiogram would show a straight line, but in practice everyone is slightly different, and small variations are considered normal." Example: