WAJ on AUDIO - for truth in hifi / stereo / high-end audio




by W.A.J.

What would audio-reproduction be like without the treble frequencies?


Absolutely dull - and certainly boring too!

Naturally (and not just because of this) audiophiles tend to believe that treble is a good thing - which it is, of course. Many also believe that more's better. And some even believe that obviously-apparent 'excess' is best.

Manufacturers and the mainstream audio-press encourage even the latter by stressing the desirability of the detail and 'air' attainable up 'there' at the highest reaches of treble reproduction. But, the fact is; for the highest treble frequencies to be realistically beneficial, to the audio-reproduction, there are certain 'infra-structural' features which must first be implemented, in regard to the reproduced audio-spectrum, to ensure optimal performance. 

The question is whether these features are present, in most modern speakers, or whether  most manufacturers are simply cutting corners to 'give it' to the people (once again) as they're always happy to do - 'giving it' to the people is what many of them thrive on, of course.

To expand on a point made in a previous article about aural balance, the trouble is that in many cases a flat response to the highest audible frequencies, in speaker-reproduction, is apt to cause problems, for those seeking realism, if things are amiss elsewhere in the spectrum. Prolonged exposure to apparent 'excesses' in this region  under such circumstances, therefore, would likely cause some to hastily exit a room in a bout of screaming hysteria. In other cases, some may be agitated without knowing the reason why. In extreme cases, exposure to apparent treble-excess could, arguably, be deemed a 'pleasant' form of torture.

Scientific-Studies Recommend; Bass & Trebles in Equal Levels - The True Balance: It is apparent that a treble-prominent type of audio reproduction suits the tastes of many. But invariably after the novelty has passed, many others, especially those seeking realism, find such audio-reproduction to be grossly objectionable, patently unrealistic, and obviously 'hifi', in the most unflattering sense of the term.

And though the audio-science is certainly not fully understood, even today, there is enough scientific evidence to suggest the proper proportions of bass and treble, relative to the middle-range, in audio-reproduction - if not, exactly, why this is so:

Around 3 years into my audiophile-life, I acquired a book on loudspeaker-design (and built my first DIY scaled-down slot-loaded corner-horn from it too, at 13 years of age, by the way). It was written in the 1950s, or some other time before I was even born. I can't remember the title but, apart from the actual designs, it also featured the results of several studies on speaker-design, and many  of the fundamental principles - which still hold true today. One of these studies, and the resulting principle, is what the arguments in this piece are based upon. However, in the absence of a viable reference to that book, I'll link to a newly-found article in which the writer refers to 'The 400,000 Law' cited by famous publisher/speaker-designer, Jean Hiraga. (The relevant section is only one paragraph atop the linked page, so go right ahead and check on it). I believe this is the same Law* I encountered in that book I mentioned - though I'm almost sure that's not what it was called in the book. *(Refer to the footnote).

First, let's recognize the fact that perhaps 99% of speaker-systems lack the capability of flat response down to 20hz - a well-known fact. But this is bass. So what does bass have to do with treble, or its effect on believable and realistic reproduction? Well, studies have shown that for believable reproduction, or even tolerable reproduction, treble and bass MUST be in balance - relative to the all-important mids. (This is based on the premise that the midrange is THE most important region in the scheme of sonic realism, and that bass and treble play a secondary role in augmenting the all-important midrange - a concept foreign to conventional thinking, thanks to both 'un-informed' and corrupt elements of the industry and the press). Too much at one end or the other, in regard to bass or treble, tilts the overall balance in that direction, to the detrement of the mids and the rest of the overall audio-spectrum.

Therefore, if the bass-response of a speaker-system stops short of the audible limit, as most do, then the treble-response of that speaker-system should also decline at a similar rate. And there are scientific formulae prescribing the corresponding frequencies, regarding bass and treble, as to the recommended points at which this decline should be effected, conducive to natural, balanced, and believable reproduction - perceived REALISM!

However, this is certainly not what we have today. The majority of today's speakers are bass 'shy' and treble 'forward'. This is GROSSLY out-of-true (significantly flawed reproduction) according to the findings and recommendations of those exhaustive scientific-studies and, perhaps, explains why many audiophiles are 'concerned' with modern speakers' light and 'bright' tonal balance. Obviously their 'concern' is well founded.

What we have today are mini-monitors, for example, with near-flat frequency-responses of, say, 20khz - 70hz, and floor-standers also with flat response from 20khz to 50hz, 40hz, or perhaps a maximum of 35hz, typically. In virtually every case of the 'better' amongst typical modern speakers, it can be seen that their bass to treble ratio is out of balance, as regards the findings of the afore-mentioned studies and conclusions.

This certainly would, in part, explain why significant numbers of audiophiles find the balance of modern speakers, indeed, to be 'too bright' (excluding high-frequency 'fans', of course, who display a 'taste' for a preponderance of treble, and perceived 'air').

But even in top-echelon speakers - or in sub-woofed examples below that level, but with flat response to 20hz - many find that the reproduction is far short of truly realistic.

Some time ago, The Absolute Sound magazine (TAS; issue # 162) carried an article pondering the obstacles to sonic realism. In a round-table discussion, amongst their top writers, one consensus arrived at was that; in modern speakers, treble and bass 'stick-out' and, therefore, detract from a perception of realism by virtue of the un-natural prominence of said treble and bass.

On the surface, this opinion expressed by TAS seems to be contrary to the findings of the afore-mentioned scientific study. (20khz treble should ideally be viable with a response 'flat' to 20hz - according to the studies). But, in reality, TAS' opinion does not actually contradict the findings of that study. It supports these findings in a profound manner:

This study, and the resulting law, evidently assumes that (as they should be) the midrange frequencies would be full and complete, as was the case among the best speakers at the time this study was done, in a previous era decades ago. A speaker-system with such a naturally full and complete midrange, as existed in this previous era, would definitely benefit in realism from a flat response to both frequency extremes. Their full, natural, and complete midrange would be (and are) absolutely able to support these frequency extremes in such a manner as to enhance the perception of realism.

However, modern speakers erroneously display a much leaner midrange than what obtained in the best speakers then (or even in nature). This is a fact which modern manufacturers actually boast about - supported, of course, by elements of the audio-press - and favored by most audiophiles influenced by the standards set by both.

But the facts, in this case, are impossible to ignore: The 400,000 Law is as constant and relevant as the Law of Gravity and other similar laws established ages ago. They do not go out of date, style, or shed their relevance with the passing of time. Breaking any of these laws (as modern speakers do, with the advent of the small mid-woofer's slighting of the midrange) upsets the natural balance. All laws of physics (nature) relate to other physical laws in some way - some closer than others. (If anyone needed another iota of proof that the lean midrange of modern speakers is wrong, then this is one). The overwhelming evidence is that this law works with large-coned mid-woofers, such as those of olde. This is because those large drivers reproduce midrange sounds similar to the way they exist in nature, which is what the 400,000 Law is based upon - ALL the midrange, as supplied by large mid-woofers. The introduction of the small mid-woofer (which lacks a similar ability) is not a step forward in the direction of realism, it's ten paces backward, compared to the competence of the large mid-woofer it replaced. And this is manifest in the discrepancies exposed when the full extent of the 400,000 Law is applied to the limits of the frequency-extremes. None of physics' (or nature's) laws corrects flaws, or laws broken in other areas (in the midrange, for instance) they expose them. And TAS' surprisingly pertinent observation is evidence of some of the consequences:

In modern speakers, response to both the frequency-extremes causes a deleterious effect, in terms of the perception of realism, as TAS proclaims. And this is because the lean midrange of modern speakers cannot support the frequency extremes in a manner that enhances realism. As we've articulated in that article about balance, and several others, the small-coned mid-woofers of modern speakers lack the ability to realistically reproduce the lower-midrange body and substance of musical notes. Despite the measurements, there's effectively a 'real-world' void between 500hz and 200hz. This causes a disconnect between lower-mids and bass - causing even natural levels of full and complete bass notes to assume un-natural  levels of prominence or, as TAS describes it; bass 'sticks-out'.

A similar scenario applies to the treble region:

Effectively, modern speakers only reproduce a part of the whole midrange (to the degree that we can actually perceive it, despite what the measurements say) as we've articulated in the case with bass. (And, as we'll see later, TAS will also tell you that modern speakers lack the realistic midrange-presence of horn speakers. Note that the midrange-presence region approximately spans from the middle-mids to the upper-midrange, just next-door to treble. A deficiency here equates to the same deficiency at lower-midrange where the effect causes bass to 'stick out'. Therefore, lack of midrange presence would also contribute to the fact that 20khz-treble also 'sticks out'. (But even if TAS were wrong about the lack of 'presence', the afore-mentioned deficiency at low-mids by itself, by virtue of its weakening effect on the overall midrange, would also cause 20khz treble to 'stick-out'). So then, effectively, there's a definite 'suck-out' at lower-mids, next-door to bass and, according to TAS, there's also a 'suck-out' at the upper-mids' presence-region, next-door to treble. Leaving only the middle-mids 'intact'. This partial,  weakened, and lean midrange is now at a disadvantage in the sense that; consequent to the mids' weakened state, bass and treble now assume un-natural levels of prominence. Therefore, this lean and weakened midrange cannot compete with a treble that extends to the extreme. And so, as TAS would put it, an extended treble-response also 'sticks-out' from the un-naturally lean, weakened, and relatively recessed midrange which prevails in speakers today.

The immediately preceding example relates, however, to the 1% of speakers with response that extends, flat, to the lower limit at 20hz. So what about the other 99% of speakers with bass response which stops short of that limit while, at the same time, closely approaching broaching or breaching the high-frequency limit?

The logical answer would indicate to us that; if the 400,000 Law is correct (and, similar to the Law of Gravity, the overwhelming evidence supports the fact that it is) then it means that, apart from the relevant points above that also apply, the vast majority of high-end speakers are also flawed with an IMBALANCE tilted toward the treble (just another in a list of flaws, unfortunately). This is because of their treble-response extending to the limit while their bass-response stops short.  

[Incidentally, and surprisingly, TAS' argument asserting that bass and treble 'stick-out', in modern conventional top-echelon speakers' reproduction, corroborates the points made in our preceeding article on aural balance. This article refutes the claims of modern manufacturers, and elements of the mainstream audio-press, that assert that bass, mids, and treble, should be evident in relatively equal levels - as is the case with the majority of speakers today (less so with bass, in lesser models). TAS' observation of the characteristics of the very 'best' of these, and their consequent lack of realism, poignantly illustrates the fallacy of that premise. I reiterate; for realistic audio-reproduction, the midrange, ALL of the midrange, MUST dominate - as this is the way it is in life, and in live un-amplified music. Anything less, as exists in most modern speakers, results in a falsification of the reproduction, as TAS so rightly observes - again; surprisingly.

Those who actually believe that modern small-coned speakers are realistic, as generally implied by many mags including TAS (though rarely directly stated) should perhaps take note of TAS' own observation of one of the very significant flaws nullifying realism in even the very best of these. (Several other factors are featured in other articles here). Note below, also, TAS' hinting at the older large theatre-horns, and a modern horn-system, as systems possessing greater realism. Audiophiles should realize that they've been led astray by TAS, and other mags, which routinely tout and rave about these small-coned speakers that TAS itself admits (on this one occassion) are incapable of the realism supplied by another speaker-type.]

Summarizing this segment; according to The Absolute Sound, modern top-echelon speakers, obviously with 'flat' response from 20hz to 20khz, are flawed with a bass and treble which stick-out (because of a lean midrange, I might add). And according to the '400,000 Law', the vast majority of others, with response broaching the treble limit, while not equalling the same feat at bass, are imbalanced (in addition to their lean mids, I might also add).

Whether actual sonic realism can be achieved with such  flaws or imbalances, I'll leave to you to ponder.

Whither the Despicable? ...They're In Evidence & In Living Presence: But the-above is only another manifestation of the dishonesty of the manufacturers who led us in this direction, motivated by their own greed (other manufacturers are innocently following the trends).

And this is also manifestation of the dishonesty of elements of the mainstream audio-press that’ve long opted to assist their allies in such nefarious deeds, motivated by the same greed.

For instance, that discussion by TAS 'pondering' the reasons for the lack of realism in today's speakers is thoroughly hypocritical since, just as I know, they also know the answers to the questions they're asking.

They even hint at it at one stage where they cite the 'midrange presence' of previous theatre-type horn speakers as one element missing from today's speaker-systems. (At another stage they also hint at the efficiency/dynamism, of those past speakers, as the greatest contributor to their realism - I urge anyone interested to purchase a copy of issue # 162, on-line, and read their article on Sonic-Realism).

Trust me, these mainstream mags KNOW that the modern speakers they routinely rave about are severely flawed, in terms of their ineptitude at realism. They also know why. And they also know the types of speakers which are intrinsically more realistic. Yet they relentlessy persist in promoting those that are not - while they generally malign and disregard those that are (other articles here explain why). Then they hypocritically indulge in round-table discussions 'pondering' the reasons for the lack of realism in the types of speakers  they've long opted to promote, raving about aspects (soundstaging, high-frequency detail, and 'air', etc.) which are secondary to those required for overall realism.

As most such mags tend to do, they dance and skirt around the issues, tease, tantalize, and frustrate the reader, sometimes coming close but never coming out to tell people the WHOLE truth. They insult, and make a mockery of, poeple's intelligence. I'm sure they get a sadistic kick out of doing the things they do. They are self-serving, greed-driven, and monumental hypocrites, of the lowest order. Despicable is too nice a word to describe this sort.

The whole TAS charade was only staged, in my opinion, to further highlight the attributes of a recently reviewed 'magical and 'ultimate' horn speaker manufactured by one of their obvious favorites, which they also mentioned in the 'discussion'. But this was done in such a way as not to 'rock the boat' and make ordinary audiophiles too aware of the inadequacies of the rest of that particular line or, indeed, those lines of other favorite manufacturers of the currently popular speaker-type. They succeeded immensely.

The 'discussion', also, was perhaps partly a measure to correct the glaring omissions of that review, but separate from said review to lessen the likelihood of readers combining the facts (now ensconced in the two separate articles of disparate issues of the mag). The review itself was skillfully done in such a way to deter readers from thinking that, though they cannot afford those 1/4 mil horns (now 400-grand) they could achieve similar performance from horns, or similarly competent systems, that are actually affordable.

The real virtues of the horns were also downplayed, in the review itself. Very little mention was made of the horn's midrange-presence or its dynamism, in the actual review. These are some of the features which make a horn more realistic than others, but then these features are not shared by the conventional small-coned speakers in the rest of that line, or those in the lines of other favored manufacturers of the currently popular speaker-type. So why stress those points? What was stressed, and over-stressed, was the horn's low-level detail resolution. Virtually the whole body of the actual review, regarding this highly DYNAMIC horn, was about low-level DETAIL - it's bizzare - I've never seen the likes of it. But then, this is a feature the others can lay claim to, so this is what the mag wants readers to aspire to - low-level detail - not midrange presence and dynamism, which the popular inefficient, undynamic, and midrange-lean, small-coned speakers generally pushed by the mag cannot provide. We can't expect them to kill their golden-egg'd goose, now can we? (Never you mind how lame the goose really is - that's not important).

And now they tell us, in that 'discussion', that one factor in modern small-coned speakers' lack of realism is that bass and treble 'stick out'. Yet, they neglect to tell us why. I assure you they do know why. As I said, they mentioned one of the reasons in the 'discussion'; modern speakers' lack of the horns' 'midrange-presence' - yet, they skillfully omitted to make the obvious connection. Another part of the reason, which they also omitted, I've also already highlighted; the lack of lower-midrange heft. These are only some of the major reasons modern small-coned speakers lack realism. 

And these are the reasons why bass and treble 'stick-out' - there's the connection TAS refused to make.

V8 Bicycles: We've dealt with similar scenarios before so now, in regard to the manufacturers, we'll only take a quick look at why this particular situation prevails, regarding this treble imbalance:

Imagine that the whole and complete midrange and its competent large mid-woofer (as previously existed) is the foundation, the relatively expensive chassis, the frame, the body of a race-car, to which all accessories are appended (bass and treble are those accessories).

In seeking to cut costs and expand profit-margins, they discarded the original sturdy and robust race-car chassis/body (the large mid-woofer of olde) which was able to support the accessories (bass and treble) in a truly realistic manner.

Pursuant to expanding profit-margins, they've replaced the race-car chassis/body with the much cheaper flimsy frame of a bicycle (i.e. small mid-woofers) to which they now append powerful super-charged V8 engines (i.e. 20khz-capable tweeters). Yet, in the majority (most speakers) these V8 bicycles still ride on the flimsy frame (small mid-woofer) of  a bicycle , and on the flimsy tyres (bass) of a bicycle.

There's no need to explain the consequences of such actions, in race-car context, except to say that we're reaping the sonic equivalent.

Oh, and as for the earth-shaking 20hz bass of the top-echelon; all they do is to add wide racing-tyres (potent bass) to the flimsy bicycle-frame (small mid-woofers) and you're good to go. 

Good luck in that race toward sonic realism (you'll need it).

If you're now thinking that a big V8 engine (20khz treble) and extra-wide slick tyres (20hz bass) would overwhelm, or 'stick out' from, the flimsy foundation of a bicycle-frame (the lean mids of the small mid-woofer) then you're probably right. If you're also thinking that a V8 bicycle would not fare too well in the stakes for the Indy 500 car-race, then you could be right on that point too. And if now you're also thinking that the V8 bicycle's equivalent, modern small-coned speakers, would also not fare too well in the stakes for sonic realism, then perhaps you're bang on the money. 

Corrective Measures: How would one go about correcting the problem(s) dishonestly created by many modern speaker manufacturers and disingenuously endorsed by the mainstream audio-press as they cheerfully assist in shafting the audiophile to satiate their own greed?

I my view, there's only one ultimate solution.

Any other alternative offered should be viewed as purely hypothetical, unless one feels sure it would work in one's own particular scenario.

So then what does one do if one already owns a speaker-system with the characteristics here described?

Well, if one is a high-frequency 'fan', then the apparent preponderance of treble should not pose a problem. But then, if one is not then, perhaps, any of a number of 'solutions' may apply:

If one's source is a turntable only, then a simple change of cartridge to one of the four lines of wood-body cartridges at the top of the Grado line will likely alleviate the treble trouble. They range in price from $3oo to somewhere above $3000. But the most remarkable feature about them is that, I believe, they were deliberately designed to compensate for the vagaries of bright recordings and bright speakers and, also, the lack of lower-midrange prowess in modern designs. In these regards, they tend to offer a sweet though slightly restrained top-end compensating for the brightness. (Models from the Sonata upwards display a treble that apparently extends to the highest reaches, but at a level perhaps 2db below that of the mids). And, also, they present a fuller and naturally mellifluous midrange, partially compensating for the lean mids modern speakers are plagued with.

If one also uses a digital source, then the problem is more difficult to address. Perhaps the addition of a tubed buffer-stage, specifically one with the tubes' typical characteristics of fuller mids and slightly restrained highs, would be a better prospect than seeking to replace an existing player. This is because one is unlikely to find such a digital source which would have the effect of a Grado cartridge, for example (a few, with tubed output-stages, may be exceptions). But this is also because there are more ideal ways to solve the problem - by addressing the source of the problem, not the source component(s) or even the amps and/or pre-amps.

If one is dissatisfied with the reproduction of one's speakers, along the lines here-outlined, then perhaps it'd be better to address the speakers directly, instead of tampering with other components to compensate.

If one is even remotely technically inclined, and also, if, and only if, one does not have warranty concerns, then the brightness problem may be solved by simply padding down the tweeter. If there are no user-adjustable x-over level controls for this purpose, then a resistor in series with the tweeter may ease the problem somewhat. Various values of resistors may be tried until the balance seems amenable. Perhaps one could start with a 10 ohm, 10watt resistor and continue from there (a final check to verify a 'safe' overall system impedance is also imperative). Unfortunately, this 'solution' stands the very best chance of completely succeeding only in cases where super-tweeters are in play above, say 12khz. The lower the x-over-point determines the less likelihood of great success. Sorry. And don't forget that the lean mids will remain un-addressed by this means, though perhaps less obviously bothersome, if the brightness is alleviated. Electrostatics, obviously, do not qualify for such experimentation, even those with separate tweeter elements, unless one knows exactly what one is doing.

Perhaps it'd be best, if circumstances permit, to just replace a 'bright' speaker-system. If one opts to continue shopping amongst typical modern small-coned speakers (not recommended, but old habits die hard, I know) then there are several that cater to those who're averse to bright tweeters, some are among the best of modern speaker-types. The Sonus-Faber line springs immediately to mind, and even some that are easily affordable are, also, outstanding - the Concerto Grand, is one example. An excellent past great, the Celestion SL 600 also had a deliberately designed-in down-tilted high-frequency response (it's one of the features which makes it great, in the first place). But in general, one should pay attention to the specs, and avoid speakers which claim flat response to the high-frequency extreme in all cases where the same speaker betrays the imbalance of a limited low-frequency response. All that glitters (in the treble) is not necessarily gold. Even if one opts not to strictly adhere to the prescribed scientific guidelines, one should generally be safe if one seeks a flat response to, say, around 13khz, with a gentle shelving of around 3db per octave to 20khz. Then listen thoroughly to confirm the choice.

However, even with the treble-brightness addressed, almost any of the typical modern small-coned speakers will still be guilty of lean mids, relative to the real-world standards of nature and natural live-music. For those with separate mini-monitor/sub-woofers, a seemingly far-fetched experiment may actually present the 'perfect' solution - ultimately better than even most of the few current large-coned examples that were purpose-built for the ultimate in truly lifelike reproduction, in the sense that it offers the benefits of both the small driver and those of the large. Refer to the article; 'The Ultimate Speakers...', with particular regard to the section which highlights DIY speakers, and this experiment.

Others seeking to be rid of modern speakers' lean mids and bright trebles, along with benefiting from all the other features of truly realistic reproduction, may find it easier just to choose  a speaker-system similar to those cited in that linked article. Problem(s) solved!

This is the ultimate solution alluded to at the beginning of this segment - the only one unreservedly recommended, actually.

Conclusion: It's unfortunate that manufacturers seek to take advantage of people's tendency to believe that more equals better. And so they gild a flawed concept/design with more high-frequency (and/or bass) capability, which only exacerbates the problems exhibited by a fundamentally flawed design.

Among consumers, the only 'winners' in this scenario, are those (perhaps many) whose tastes and preferences are satisfied by un-natural enhancements and omissions, regardless of the deleterious effects on overall realism.

But those who seek sonic realism should perhaps take a dim view of the disingenuous shenanigans of most speaker manufacturers. In light of the midrange limitations of modern speakers - with specific regard to these limitations' relationship with the treble-region, and also bass – more is not necessarily better.

If lifelike realism is the goal, it's better to fix the foundation (mids) before implementing elaborate extensions and additions.

Down with the trouble with treble - and bass too!


*Note: The title of that Stereophile article is; 'God Is In the Nuances' - an interesting read, in parts - go ahead and indulge. Be advised, though, that  it's quite long, 11 pages (and here I am - trying to cut back on the lengths of my own pieces). Some of the writer's views seem to coincide with mine (Stereophile - really?) including those on this subject. However, the writer seems to be under the impression that Jean Hiraga was the originator of the tenets of the so-called Law of 400,000. Hiraga San most assuredly was not. The book I had (with those tenets outlined) was ancient, dating from the 1950s or beyond. And I'd guess it was only quoting from the source of the original studies which resulted in that 'Law', as many others would have done during that era, and should continue to do today - the accomplished speaker-designer, Jean Hiraga, obviously does. He is indeed a great thinker and designer, but Hiraga San had nothing to do with the origins of the tenets of that 'Law'. A poster on this thread (discussing the linked article) asserts that these tenets may have originated from a book, by Badeimaff and Davis, in which they were cited as the Law of 500,000. He could be right, but unless that book also includes speaker-design plans, it wasn't the one in which I first encountered this Law either. But even if it's not the book I had (or especially if it's not) all this does hint at the wide usage, acceptance, and credibility of this Law - sadly discarded, along with other well-established laws and principles, by all but a few modern manufacturers more motivated by hype and profit than sonic realism. I'd bet many manufacturers aren't even aware of this 'Law', or others, relevant to their 'vocation'. Little wonder modern speakers are the way they are. WAJ.


Copyright 2012