Inverted law of diminishing returns?

[George47]

I sort of agree with both guys.

Yes, it is possible that some inexpensive products are made down to a price and sound OK. If more resources are available then the sound quality could increase and spending twice as much money gives you sound quality that is twice as good. Depending on how you value sound quality. However, after that spending more money only results in incremental increases in sound quality. And finally increases in costs only get you a prettier box.

Of course, some hifi has very pretty boxes and basic electronics.

Speakers are a good example. For a few hundred pounds you may get a 5" bass and a basic tweeter. Make it £600-700 and an 8" bass and tweeter become possible and the bass quality (if nothing else) increases significantly.

[Pharos]

Thank you Geoff for your post 5 which is educational.

I do think that diminishing quality improvement with increased expenditure applies to most that we buy, but in the case of Hi-Fi another law, which you allude to may also apply, that of the law of limiting factors.

If we have a very good system, it is reasonable to suppose that much of it is very near impeccable, and hence near perfection. If that is so then it will be more transparent to the tiniest of improvements in any part of the system, and so these are more significant, this improvement being perhaps the removal of a limiting factor.

[Jazid]

Sorry, James - you're a good guy, but that response is just plain rude.
Far too reminiscent of the bully-boy days of a certain other forum.

I'd better explain myself then, I don't want to be accused of being a Trojan Horse from the unspeakable side!

The principle of diminishing returns is not just an economic law but a far more general scientific postulate that is useful because it has such simple explanatory potential. It is not a law, it is fallible, and is based on another principle, that of limiting factors, developed from Liebig's 'Law of the minimum'. The principle is simply that in any complex productive system whatever critical resource is in most limited supply will limit the rate of output of that system. For example in crops where growth is limited by the level of nitrate available to the plant, adding phosphate fertiliser will not increase growth rate as their growth is being 'strangled' by the (lack of) availability of nitrate.

The interest to hifi buffs I humbly suggest is far less in the application of the 'law' but in the reasons behind its failure. It turns out that it is quite common that by 'pushing' a system by adding extra quantities of one resource when its output is limited by a different resource often does lead to unpredicted fractional gains, and the reasons for this often remain obscure. To quote an early research article that tried to separate out various factors affecting one aspect of plant growth:
"...since the whole process consists of a series
of interlinked reactions, the effect which a change in one factor brings
about ... may induce slight modification in others as
well, thus indicating that there cannot be a sharp line of demarcation
between the changes produced by one or the other of the factors."

And to take this explanation into a hifi related context - this is a classic explanation of the unpredictable effect of system synergy. With regard to arms and cartridges the effect is clearly explainable, a different arm has a different mass and resonance, it will react to a cartridge in a different way and the dynamics of this effect will be translated into different output from the coils. As has been said a thousand times on this forum, even with amps and speakers, with sources and pre-amps, with rooms and room treatments the effects of even apparently small changes can be complex and unpredictable, to me this seems to be a 'curved ball' that makes it impossible to attribute any absolute value to a single component as it may be used in so many different systems. If there can be no absolute value attributable to a system component, how can there be a rank of values that such changes can be measured against? Without a rank how can the law of diminishing returns (that this discussion has become about) be employed for explanatory value?

TLDR - The Law of Diminishing Returns, also colloquially expressed as the 80:20 rule. Not a rule but an effective explanatory tool. Eg. 80% of meaning may be communicated in 20% of a sentence.
TLDR - Too long, didn't read. A blunt application of the 80:20 rule, brutal ( maybe to the point of rudeness) because of its economy. In fairness if the acronym meant 'not relevant, didn't read' that would have been more accurate. But Geoff, please don't think that I am not interested in your economic exposition of what is clearly a pervasive principle, I was not even aware of its use in your field and to me it is interesting in its own right.

It's rather just my opinion that this 'law' is the limiting factor in this discussion.

[Sherwood]

I'd better explain myself then, I don't want to be accused of being a Trojan Horse from the unspeakable side!

The principle of diminishing returns is not just an economic law but a far more general scientific postulate that is useful because it has such simple explanatory potential. It is not a law, it is fallible, and is based on another principle, that of limiting factors, developed from Liebig's 'Law of the minimum'. The principle is simply that in any complex productive system whatever critical resource is in most limited supply will limit the rate of output of that system. For example in crops where growth is limited by the level of nitrate available to the plant, adding phosphate fertiliser will not increase growth rate as their growth is being 'strangled' by the (lack of) availability of nitrate.

The interest to hifi buffs I humbly suggest is far less in the application of the 'law' but in the reasons behind its failure. It turns out that it is quite common that by 'pushing' a system by adding extra quantities of one resource when its output is limited by a different resource often does lead to unpredicted fractional gains, and the reasons for this often remain obscure. To quote an early research article that tried to separate out various factors affecting one aspect of plant growth:
"...since the whole process consists of a series
of interlinked reactions, the effect which a change in one factor brings
about ... may induce slight modification in others as
well, thus indicating that there cannot be a sharp line of demarcation
between the changes produced by one or the other of the factors."

And to take this explanation into a hifi related context - this is a classic explanation of the unpredictable effect of system synergy. With regard to arms and cartridges the effect is clearly explainable, a different arm has a different mass and resonance, it will react to a cartridge in a different way and the dynamics of this effect will be translated into different output from the coils. As has been said a thousand times on this forum, even with amps and speakers, with sources and pre-amps, with rooms and room treatments the effects of even apparently small changes can be complex and unpredictable, to me this seems to be a 'curved ball' that makes it impossible to attribute any absolute value to a single component as it may be used in so many different systems. If there can be no absolute value attributable to a system component, how can there be a rank of values that such changes can be measured against? Without a rank how can the law of diminishing returns (that this discussion has become about) be employed for explanatory value?

TLDR - The Law of Diminishing Returns, also colloquially expressed as the 80:20 rule. Not a rule but an effective explanatory tool. Eg. 80% of meaning may be communicated in 20% of a sentence.
TLDR - Too long, didn't read. A blunt application of the 80:20 rule, brutal ( maybe to the point of rudeness) because of its economy. In fairness if the acronym meant 'not relevant, didn't read' that would have been more accurate. But Geoff, please don't think that I am not interested in your economic exposition of what is clearly a pervasive principle, I was not even aware of its use in your field and to me it is interesting in its own right.

It's rather just my opinion that this 'law' is the limiting factor in this discussion.

The concepts you refer to were developed by economists and have a precise meaning to economists. I think you will find that most economists avoid the use of the term "law" and use the term hypothesis. Like all hypotheses, DMR must be tested empirically (i.e. by rigorous measurement). A hypothesis is accepted as "true" only when it hold up to this type of scrutiny and measurement.

The idea of DMR originated amongst economists in 18th century France and was used by the then Minister of Finance to highlight the limitations on agricultural productivity gains. It was taken up in the UK by the economist David Ricardo, and used by his friend Thomas Malthus to warn of the dangers of population growth rates in excess of the carrying capacity of the land. In the 19th century, Francis Edgworth and others developed analytical tools and mathematical methods to explore production processes and the marginal contribution to output of specific inputs. The hypothesis does not relate to the scarcity of a specific input as a constraint on output (i.e. a shortage of fertiliser) but to the physical production process (i.e. the potential for increasing output by using ever increasing quantities of that input). I mention this because the concept, originally developed by economists to describe and explain input/output relationships has been adopted and in many cases mis-applied to other fields on activity, including investment decisions where outputs are not objectively measurable (e.g. hifi).

Perhaps I have not made myself clear: but the hypothesis of DMR is absolutely not appropriate when evaluating individual hifi components. A stronger, but still weak case, can be made for applying the concept to complete systems, and intuitively this is what we all do when we set out to build a hifi system. When I started out I had a friend's system as my reference point for SQ. I then listened to perhaps half a dozen systems at different price points (all far below that of the reference system which was way beyond my means at the time). I ended up spending way more than I had planned to because my preferred system hit the sweet price/performance ratio. In evaluating these systems I did not focus on the marginal contribution of any single component, but on the overall result and the synergy of different arrangements of a number of components, though I did decide early on that a pair of Rogers LS3/5a speakers would feature in the chosen system. The only way in which DMR was relevant to my choice was at the system level and whether I could justify the extra spending over the most basic system in terms of subjective performance gains.

Geoff

BTW: you are confusing DMR with the Pareto Principle (the so called 80:20 rule). Initially, the term was coined by Italian economist Vilfredo Pareto who observed that approximately 80% of the land in Italy was owned by 20% of the population (an observation that was subsequently found to apply to many other countries). The Pareto Principle, like the DMR concept, has subsequently been "borrowed" and modified by management consultants and others, to highlight the most important elements in any process. Again, a case where an economic concept has been misunderstood and often misapplied to other areas of choice

[Jimbo]

Come on guys, less Intellectual willy waving, I think the OP was just using the concept to express his thoughts regarding his arm upgrade - right or wrong.

[mikmas]

Come on guys, less Intellectual willy waving ...

If willies were as fascinating as this discussion there might be a point to waving them ...

[Sherwood]

Come on guys, less Intellectual willy waving, I think the OP was just using the concept to express his thoughts regarding his arm upgrade - right or wrong.

I do not disagree with the underlying premise. I just object to the misuse of terms and misunderstanding of economic concepts used to explain it. "Willy waving" is a bit harsh, as I am an economist and this is my area of expertise. If somebody on this forum incorrectly quoted an electrical formulae or suggested a factual inaccuracy about a circuit, then are you suggesting it should not be corrected by those who are better informed on the subject?

I have made the point, so hopefully the confusion and misrepresentation is over.

Geoff

[magiccarpetride]

I do not disagree with the underlying premise. I just object to the misuse of terms and misunderstanding of economic concepts used to explain it. "Willy waving" is a bit harsh, as I am an economist and this is my area of expertise. If somebody on this forum incorrectly quoted an electrical formulae or suggested a factual inaccuracy about a circuit, then are you suggesting it should not be corrected by those who are better informed on the subject?

I have made the point, so hopefully the confusion and misrepresentation is over.

Geoff

My motivation for bringing up this (mis)used concept of 'diminishing returns' stems from my repeated experiences with hi fi salespeople. I suspect that when you're in sales, your primary motivation is your commission. So if a customer is asking for a piece of advice, and if there are several components that the sales person might recommend, he will most likely recommend the component where he stands to make the largest commission. Irrespective of the actual quality, or appropriateness of that component to the customer's situation. Sorry to sound so cynical, but such are the harsh realities we live in day in, day out.

So what that means is that if I then ask for a component that maybe is more expensive and why wasn't that one suggested to me as an improvement, I usually get the 'diminishing returns' speech. But I think it's a red herring. I don't really believe there is such thing as 'diminishing returns'.

To me, more expensive usually signals better engineered, with much more stringent (and costly) quality control process. QA costs a lot of money, and the cost must be passed to the consumer. I'd gladly pay more if I can be convinced that the extra cost means quality assurance.

[Sherwood]

I think the concept of diminishing returns may have originated in the field of economics but has now become accepted language in the wider world to mean "pay increasing amounts of money for ever diminishing benefits". In that respect it's similar to the term "critical mass" which originated in the field of nuclear physics but is now a widely used phrase meaning "big enough to become viable" or "big enough to become self-sustaining".
Thanks for explaining the origins of the term diminishing returns - I wasn't aware of them - but language evolves.

Andrew,

You are correct in saying, that DMR, like many technical terms, has entered into common usage, critical mass being a good example. In some cases, the migration is better than others in preserving the original meaning. For example, the term "decimation" is now used to describe a massive reduction in capacity or effectiveness (e.g. the decimation of our fish stocks; the team was decimated by injuries; our industry has been decimated by foreign imports). Originally, the term applied to the symbolic punishment of one in ten soldiers in lieu of punishment of the whole group (hence decimate for one in ten). Nowadays, the term is generally taken to refer to a greater reduction of capacity than a mere 10%. I mention this only because I am a big fan of George Orwell and Ernest Gowers in advocating plain writing. Metaphors and unnecessary jargon are best avoided. In this case, the concept was just not relevant to the situation, even in modern usage.

... and on that note I end my sermon!

Geoff

[magiccarpetride]

It's a matter of opinion, but I disagree. Good design doesn't have to cost any more money poor design, and in my opinion it's easy to spend a lot of money on production costs for no real purpose other than because it looks good, feels heavy etc. Yes, but does it perform any better?

You've managed to ignore one of the most important aspect of good engineering -- quality control. Yes, it is possible to come up with good design with not much more money than bad design, but to implement it properly, you need to pony up some serious cash. And that's where we enter the pricey hi fi territory. Most of us have seen some pretty decent gear that sells for cheap and ships from China, but buying such items is a dicey game -- you may or may not luck out and get a well assembled piece. Or you may get a shoddily assembled piece, which breaks down after few hours/days.

So you basically get what you pay for.

Also, performance is a very broad category. Does it perform any better? Depending on perform what? A mini van performs better than a Ferrari when it comes to transporting a family from one place to another. Ferrari, on the other hand, performs better when it comes to rolling down the street turning heads.