Law of Accelerated Returns


I think back over the many decades of pursuing high end audio and I realize some of the most inspirational were listening to state of the art systems. Systems I could never dream of affording. I occasionally would get up early and drive the two hours to Phoenix in hopes of finding no one listening to the state of the art system in “the big room” at one of the four or five high end audio stores there in the early ‘90’s.

One such time I was able to spend over an hour with the most amazing system I have ever heard: Wilson WAAM BAMM (or something like that… all Rowland electronics, Transparent interconnects). The system cost about over $.5 million… now, over a million… although I am sure it is even better (I can’t imagine how)..

 

But listening to that system was so mind blowing… so much better than anything I could conceive of, it just completely changed my expectation of what a system could be. It was orders of magnitude better than anything I had heard.

 

Interestingly, as impressed as I was… I did not want “that” sound, as much as I appreciated it. It still expanded my horizon as to what is possible. That is really important, as it is really easy to make judgments on what you have heard and not realize the possibilities… like never having left the small town in Kansas (no offense).

I keep reading these posts about diminishing returns. That isn’t the way it works. I recently read an article by Robert Harley in The Absolute Sound called the Law of Accelerated Returns that captures the concept perfectly. March 2022 issue. The possibilities in high end audio is incredible. Everyone interested in it in any way deserves to hear what is possible. It is mind expanding. 

 

 

ghdprentice

Showing 9 responses by noske

Harley has been promoting this idea for quite some time. In 2014 he wrote this -

Many audiophiles think of an audio system from the perspective of The Law of Diminishing Returns. That’s the idea that each successive dollar you add to your hi-fi budget realizes progressively less sonic improvement. To use an extreme example, there’s undoubtedly a greater sonic difference between a $1000 loudspeaker and a $2000 loudspeaker than there is between a $100,000 loudspeaker and a $101,000 loudspeaker. The additional thousand dollars buy you much less at the top end of the scale.

.....

My experience with state-of-the-art audio suggests that The Law of Diminishing Returns is a fallacy. In fact, I think that an audio system follows The Law of Accelerating Returns.

Is there any means by which I can read the March 2022 version? I can't seem to find anything from March, possibly only for subscribers.

Perhaps he doesn’t contradict himself after thinking about it for a couple years short of a decade.

 

Harley wants you to maintain interest in increasingly expensive equipment so that you will read the magazine that employs him. He then gets to play with the really expensive stuff while you just read about it.

I have no wish to speculate on his motives for getting something so basic so fundamentally wrong, but anyway, in my experience price often has very little to do with quality (however defined) so its not an issue..

There is a well respected school of thought, that has been around for quite some time now, that what someone pays for something is less than or equal to what they value it at.

What Mr Harley is on about is something rather different.  Although, as I said, I have only read what he wrote in 2014, not 2022.

I’ve heard extremely high fidelity equipment that put high end equipment to shame at a fraction of the cost

Yes indeed. Robert Harley’s many edition book is called "A Complete Guide to High-End Audio".

He promotes such things as MQA technology.  I digress - my pet peeve.

 

@djones51 If you don't think diminishing returns applies to this first world hobby then you haven't the slightest idea what diminishing returns means. 

And to clarify the issue, the principle is actually called the law of diminishing *marginal* returns.   This concept is rather more nuanced than perhaps people realize.

Robert Harley came close to getting it correct in the passage I quoted from 2014.  (This despite him then rejecting it in audio with some magic logic.)

@onhwy61 I just think the concept of accelerated returns in audio reproduction is false. I find it interesting that nobody has offered actual examples of it happening.

Your thoughts are 100% correct.

So, that nobody has (or can, or ever will) offered actual examples of accelerated marginal returns is irrelevant. 

However - should there be honest examples provided they would be similar to what is sometimes loosely known as a Black Swan event

@asctim Yes, what you say has a lot of merit..

Normally, marginal, or incremental, return is easiest to understand if only one thing is changed at a time. Even that can be tricky with audio.

Should something else be changed as well, like knowledge or appreciation through learning, this complicates but does not invalidate the principle. This is why the words "äll other things being equal" are often used.

Actually (and I digress), the more you learn and discern about what matters, you may find yourself spending less than you otherwise would have on that next bit of kit.

The landscape has changed and any meaningful comparison with the previous change becomes problematic.

I suspect that may be the opposite intent of Mr Harley’s editorial.

I hope I’ve understood you correctly.

{edit - having taken my time tapping that out I now see others have contributed in a meaningful manner so anyway...} 2nd edit - to be abundantly clear, I am talking about marginal or incremental changes in sound quality or spending, like little bite size portions - I speak not of levels of, or total, enjoyment or expenditure. They are increasing, even when marginal may be decreasing.

 

@asctim And what I said about changing just one thing at a time in audio being tricky is indeed illustrated by your experience.

The two amps are quite different in many of their characteristics. I know Topping are very committed to being "transparent" - adding nothing to the source material. Like it sounded in the studio. This is actually something that most people probably haven’t experienced first hand.

Other amps (especially tube amps) add 2nd and 3rd harmonics, and I would say that the Hafler amp probably does too from your description. There is much more involved, but that is sufficient for now. There has been research into why the brain interprets these aspects as pleasing.

So, a bit like comparing a photo of something (perhaps one by Ansel Adams?) with a painting of the same subject.

What is important about all this in the context of the thread is that this comparison cannot be captured at all well by the concept of marginal return, because the technologies are rather different.

What must be compared are at least two recent releases by the same manufacturer, just at different price points. One may cost twice as much as the other one, and the sonic difference rather slight. If any. One that only a reviewer may be able to discern.  

But I dunno, the more I think about it the more I think that anything like that (demonstrating declining marginal return) is just giving some warped credibility to whatever it is that Mr Harley was on about, and I no longer want to do that - he made a claim so its his responsibility to prove it.

This is a dismissal of the nature that is known as Hitchen’s Razor.

 

I can’t think of any valid examples of comparisons where marginal returns in audio gears is increasing.

The reason is that the fundamental technology must remain constant. This is mostly an engineering issue, and usually there are more features as you pay more and this is a similar issue. So, I should stop now. But just to illustrate -

A trite example, about as generic as I can think of - balanced connections added to a DAC. Topping, Gustard and others have this kind of range, and they cost more.

The balanced is supposed to be better than just RCA.

Let’s pretend that the percentage "betterness" is more than the percentage more than you paid, and Harley announces that he has discovered increasing marginal returns.

This isn’t correct. He’d be wrong. There are two different technologies on the table.

There are plenty of other examples.

Edit - there maybe obvious engineering thresholds. Below the threshold, blah!! Just add one more tiny component, and it goes from blah to actually OK, a hundred times better. This is hardly relevant to what Harley is talking about - high-end, or at least pretty good stuff to start with.

.