"Subjective" v. "Objective" in Audio

[EricW]

Here's a take on the debate from a well-known amplifier designer (Nelson Pass) that I found amusing and sane:

"That is a matter of opinion. I have seen pieces in Stereo Review and elsewhere stating outright that measurements have already adequately defined performance and that the subjectivists are fooling themselves. The opposite viewpoint is stated just as dogmatically. Both sides are emotional, and neither side is particularly reasonable.

Obviously, reality sits somewhere in between. Some measurements do tell you something about the sound, but not very reliably, and there are clearly some phenomena going on that are not being measured. On the other hand, I have witnessed blind tests where the participants could not hear a difference, or heard wild differences that could not have existed.

Me, I don't care that much; in fact, I find the subject kind of boring. We build amplifiers that sound good and measure reasonably well and don't break. If you want to get a machine to listen to them for you, be my guest!"

[kittykat]

Me, I don't care that much; in fact, I find the subject kind of boring. We build amplifiers that sound good and measure reasonably well and don't break.[/INDENT]

That’s his escape clause? Perhaps his standards for “reasonable” is someone else's unacceptable (especially at the prices he is asking)?

Have a look here

http://www.stereophile.com/solidpowe...ss/index4.html

Umm. Think a 5 month old baby can draw a straighter line than his freq response curves. Umm, also looks like a Harbeth supertweeter would be a bit under utilised using this amp. And those harmonics, umm, warm and fuzzy. Nice….

Ok, then have a look here

http://www.stereophile.com/solidpowe...74/index9.html


another pass lab amp. Look at how flat it is. I find it a bit bizarre coming from the same designer. Consistencies? Philosophy? Zero. No wonder so many of you think there are differences in the sound of amps. That is because the amp designer is perversed and not really an engineer. He is a DSP machine with a soldering iron in his hand and trays of components in from of him.

What’s consistent about Pass amps? Go to the bottom and you’ll see… harmonics . Nice. Tubes in a solid state. Supercool.

Bring out the MRI machine!

Ok, you wanna see real amps? Go here…

http://www.stereophile.com/solidpowe...s1/index4.html

this is a working tool, a real amplifier. People talk about a straightwire with gain but they dont want to work hard enough to look for it. Want to see another working tool? Search Bryston, any model and look at how it measures.

MRI machine time!

[EricW]

kittykat:

What I liked was Pass's philosophy - I'm not knowledgable at all about his amps. I assume there are people who like them, as he's been in business a long time. But I agree, for a bulletproof, well-engineered, high performance amplifier at a non-ridiculous cost, I think a Bryston is pretty much unbeatable.

Oh, and I've been in an MRI machine. Have you? Personally, I didn't think the acoustics were all that good.

Eric

[yeecn]

Obviously, reality sits somewhere in between. Some measurements do tell you something about the sound, but not very reliably, and there are clearly some phenomena going on that are not being measured. On the other hand, I have witnessed blind tests where the participants could not hear a difference, or heard wild differences that could not have existed.

There is no 'in between' in a blind test. It is either you choose A or B as being your preference. It is just that simple. It is to establish whether there is any basis to claim that A is better than B (or vise versa). It is not about the what differences you hear or why you made your choice.

Nelson Pass is either incompetently ignorant - or irresponsibly trying to confuse the issue.

[kittykat]

Hi Eric

Sorry, the MRI wasnt meant for you. (no i havent been in one before. touch wood). The MRI was rolled out in this forum yesterday (by Yeecn) and will be used from now on to read the brains of those who need some serious realignment. eg If anyone buys a Pass after this*, electroshock therapy from the outputs of their Pass Labs amp (playing Def Leppard) would be in order, followed by MRI scans to check if all the gunk has fallen out from their brains.

* I know we live in a democracy and are free to choose, but this is also the process which has resulted in some people voting for governments which have invaded other countries to kill innocent civilians. Its off topic but relevant. The commonality is smokescreens, circus sideshow type mirrors, confusion and as yeecn has said irresponsibly trying to confuse. Actually NP might himself be very confused.

[Gan CK]

Perhaps Nelson Pass is a legendary amp designer but i never quite got into the sound of Passlabs & its earlier incarnation, Threshold from the 80s. I recall one shootout where a very expensive passlab preamp didn't even manage to sound more musically convincing that the pre amp section of the cheap but venerable Nad 3020. That comparison was quite an eye opener for those who only believe in American exotica. Many Audiophiles here with deep pockets look down on brit amps, always placing them below the big names but i prefer the musicality of a simple brit amp anytime of the day.

[kittykat]

Sound and measurements are important. cool.

Philosophy and Consistency of the company (which indicates discipline, and equipment measurements will tell you this) are also just as important. Remember, it is an article, an actual physical item we are doshing out our hard earned cash for, not some subjective (although that is the end result) thing.

How are we going to practically do a listening comparison of hundreds of different amps under so many different conditions and variables. If you had listened to pass labs amp x above and then pass labs amp y above, that alone would have made you confused enough. So everytime someone asks how does amp x sound with Harbeth M30's, its again time to bring out the MRI machine. The answer doesnt mean anything very much really. i think most of the time we recommend something its self appeasement (and i am guilty of this) for the choices we've made. some dumb, some even dumber i admit.

[honmanm]

Eric, thanks for starting this thread... the previous discussion was getting interesting but I felt a bit bad that we'd invaded Alan's Harbamp announcement thread!

I'd pretty much agree with the opening quote from Nelson Pass, especially

Obviously, reality sits somewhere in between. Some measurements do tell you something about the sound, but not very reliably, and there are clearly some phenomena going on that are not being measured.

The objective/subjective divide is really about what do with these "phenomena" - the extreme subjectivist approach is to treat them as "magic" imparted by religious artifacts like Mystic Meg capacitors, and extreme-objectivists will deny the effect because there is no known way of measuring it.

We all have philosophies somewhere on this continuum... and then into the mix we throw taste, so that even two objectivits may not agree on whether one objectively well-designed component sounds "better" than another (because really we should be saying "I prefer that one" instead of "that one is better").

And it also seems that designers have two approaches to managing the "phenomena" - objective designers tend to prioritise stable, repeatable perfomance despite the effect of these not-yet-measurable effects (even at the cost of complexity), subjectivists often produce designs that are intentionally less complex - with fewer components to contibute an effect.

[honmanm]

And kittykay's liks to the amp graphs was very useful... the first "wiggly" one is a classic - 0.5dB is not much of a wiggle but if you add it to wiggles in speaker, room, and weird cables the amp probably sounds sublime in some settings and pretty horrible in others.

My take on this is, if buying costly equipment a subjective comparison of objectively designed products is the way to go. For cheap stuff subjective designs are fine, but one just has to keep in mind that the product is affected by everything around it and wil always be prime suspect if the system is sounding "wrong".

[kittykat]

For cheap stuff subjective designs are fine,

Hi honmanm, the beauty of it is that you can find incredible consistency in some brands which dont need to cost a lot eg. Marantz. The SM11 example is extreme. i personally cant justify spending that much. if you have time, do try and look for some of their other models, they measure just as consistently, good and decent, even down to their humble 5003's. Please note however that the 5003 was faulty (damaged in transit?) when it was being measured at stereophile and golden ears couldnt pick it up! The print version (think it was one issue after) has a re-measurement (not on the web unfortunately) of another 5003 supplied by Marantz USA. The measurements were pretty much typical Marantz, good.

BTW i dont have a vested interest or own any shares in DNM. dont want any misunderstandings about going on about this brand. As such we shall from now on just know this brand as Big M.

[honmanm]

As such we shall from now on just know this brand as Big M.

No harm in praising the good-value consistent brands, sort of like the inverse of naming and shaming! (although that said, 10 years ago when I last bought a CD player it was demonstrated on a nasty sounding Marantz AV amp). Not sure what you mean by DNM, do you mean Mr Morecroft (www.dnm.co.uk)?

It'll probably be several years before there is any serious amplification in this system but in the meantime it has been fun discovering the "British-but-not-Quad" type of electronics that were made in a garden shed and can be picked up quite cheaply secondhand. But Quad (well, Peter Walker's Quad) is still my most admired brand and one day I'd like to get my hands on one of their 500 series "pro" amps.

I guess my vintage/DIY perspective is a bit unusual here, my only defence is that the Harbeth journey started with hearing a pair of old HL-P3s, and how they portrayed the kind of music I love. And the discovery that even in the UK they are like the proverbial hen's teeth!

[tozen]

The objective/subjective divide is really about what do with these "phenomena" - the extreme subjectivist approach is to treat them as "magic" imparted by religious artifacts like Mystic Meg capacitors, and extreme-objectivists will deny the effect because there is no known way of measuring it.
.

I am not so sure. The divide is really about on what basis we make evaluations and distinctions between pieces of audio equipment. Do we distinguish by measuring or do we distinguish by listening? The extreme subjectivist might be happy to be clueless about what gives rise to the sonic differences. He doesn't need to resort to magic, though of course he might. The subjectivist feels no urge to explain the phenomena - he is just enjoying the music and how it makes him feel.

[honmanm]

he is just enjoying the music and how it makes him feel

Well hopefully we are all aiming to enjoy music!

As a customer I don't distinguish by measurement (how could any customer outside of the broadcasting industry?) But in a system that consists of source, amplification, speakers, and room (at least!) that is not performing as it should, how do you pin down the source of the problem when the components have not been designed for consistent performance with a wide variety of partnering equipment? You end up swapping components trying to find that magical mix where their failings cancel each other out.

And when one tries to upgrade a component in that kind of system, it becomes so hard to find a replacement that gels with the other equipment - and I find I end up listening so hard to the system, trying to evaluate the component match, that the joy of the music is lost.

As someone (Daved Lovel?) posted a while back, IMO the sensible thing is to use objective criteria to narrow the field, then use one's taste to make the final choice. Upgrades then become comparatively painless.

[EricW]

There is no 'in between' in a blind test. It is either you choose A or B as being your preference. It is just that simple. It is to establish whether there is any basis to claim that A is better than B (or vise versa). It is not about the what differences you hear or why you made your choice.

Nelson Pass is either incompetently ignorant - or irresponsibly trying to confuse the issue.

Mr. Yee, I did not understand Nelson Pass to be arguing against blind tests. In fact, in the quote he makes it quite clear that he's used them and that, under those conditions, people will often either hear no differences, or (same thing) "hear differences which could not possibly exist". Either way, I read that as an affirmation of the methodology, not a repudiation, and I'm not sure why you think otherwise.

[kittykat]

Not sure what you mean by DNM

sorry honmanm, serious typo. meant D&M holdings. denon marantz group, think its under one unbrella now. but feel their philosophies (and marketing) are perceptually still distinctive enough.

DNM? lets not go there. dont get me started on them. MRI brain scans wont be enough for people who want them. Lobotomies as well sending owners my bank account number would be more befitting.

[Euler]

IMHO the purely objective approach fails to capture the essence:

Grownups love numbers. When you tell them that you have made a new friend [Amplifier A], they never ask you any questions about essential matters. They never say to you, “What does his voice sound like? What games does he love best? Does he collect butterflies?” Instead they demand: “How old is he? How many brothers has he? How much does he weigh? How much money does his father make?” Only from these numbers do they think they have learned anything about him. [The Little Prince, Saint-Exupéry]

But the purely subjective approach cannot yield credible statements, since, as Alan has pointed out,

My point (again) is that you are not able to draw any valid conclusions that are worthwhile (i.e. as a piece or rational science) unless you equalise the levels and frequency responses exactly and make instantaneous switchovers between the amps. And when you do take the trouble to set up and construct a rational scientific test, those differences that you're sure that you hear and you'd swear on your grandmother's grave are tangible, suddenly diminish to virtually nothing or less.

I'm reminded of the poor parents of autistic children, parents who were absolutely convinced that their autistic child, with a parent's hands on the child's, could communicate using a ouija board. Imagine the joy in becoming convinced that your autistic child could now express her intelligent thoughts. Sadly, several blind tests confirmed that the parents, not the children, were, without realizing it, the ones doing the communicating.

The nature and value of real scientific testing, and blind tests in particular, are severely under appreciated in the general population.

Bruce

[EricW]

While we're all on this topic of blind testing, I'm wondering if the distinction between single and double-blind testing is that important, at least in audio.

As I understand it, in a single-blind test, the test subject is not aware which of the two products (let's assume it's an amplifier test) he/she is listening to at any given point. The person administering the test does know, however, and presumably says something like "this is amplifier A" or "this is now amplifier B".

In a double-blind test, neither the tester nor the subject is aware of which of the two amplifiers is being listened to - all anyone knows is that a switch is occurring, and the results are analyzed later.

Presumably the double-blind test is considered preferable because it elimates the possibility of the tester unwittingly influencing the choice of the test subject, even if the influence is unconscious.

So my question is this: do we know for a fact that the influence of the tester's knowledge in a single-blind test is so significant that it will skew the results, as compared to a double-blind test? If so, how do we know this? Is there data confirming the effect, or is it just supposition? And does it make a difference whether the tester has any feeling one way or another about the alternatives on offer? Suppose it's a non-audiophile tester who has no feelings either way about, say, whether the amplifier is a Bryston or a Sony - how then is any influence possible?

Thanks in advance for any input.

[yeecn]

IMHO the purely objective approach fails to capture the essence:

Grownups love numbers. When you tell them that you have made a new friend [Amplifier A], they never ask you any questions about essential matters. They never say to you, “What does his voice sound like? What games does he love best? Does he collect butterflies?” Instead they demand: “How old is he? How many brothers has he? How much does he weigh? How much money does his father make?” Only from these numbers do they think they have learned anything about him. [The Little Prince, Saint-Exupéry]

Not like that at all. Let me describe one particular form of blind test - the ABX test. There are two channels connected to equipment A and B respectively. And there is a third channel X, which is connected to either to A or B. The participants are asked to determine whether X is A or B. The participants are allowed to switch freely between the channels for repeated listening. There is no time limit in ABX test. Theoretically it can take days or months.

This is basically sound matching. No numbers, no calculations and no rational thinking process is involved at all. It you can't get that right - whatever 'essence' you attribute to the equipment is pure imagination.

EricW - I hope this answer your question as well.

[EricW]

EricW - I hope this answer your question as well.

It doesn't, actually. What I was curious about is whether there was any evidence that a double-blind protocol would lead to a significantly different results to a single-blind protocol, in which only the person undergoing the test does not know which of the two products he/she is evaluating at any moment, but the tester does know.

[yeecn]

It doesn't, actually. What I was curious about is whether there was any evidence that a double-blind protocol would lead to a significantly different results to a single-blind protocol, in which only the person undergoing the test does not know which of the two products he/she is evaluating at any moment, but the tester does know.

Sorry - I mean your earlier question regarding Nelson Pass.