I think that the brightness of most modern speakers is a deliberate design decision. This seems to be what most listeners want and what most audio reviewers praise.
From the time I got into this hobby back in the 1960s, I've thought of the "New England" sound epitomized by companies like AR, KLH, Advent, EPI, Rectilinear, Dynaco and some others as "natural" sounding in terms of what one hears from live unamplified music in a good hall. When measured on axis, these speakers showed good solid bass with as much low-frequency extension as possible given the box size, a bit of warmth in the midbass, neutral mids and rolled highs above 4 kHz. That's pretty much the frequency response one hears in good halls from the audience, by the way. See the "classic" article on what you hear in a concert hall here, for example. Since most recordings are made with the mikes up close and high up (where there is generally less bass and more highs) rather than from an audience position, the speakers or some electronics in the home listening system needs to correct for the way the recordings were made, assuming the recording engineer did not fully correct for the mike placement. Decades of listening to classical music recordings on a variety of systems strongly suggests to me that most recordings are still inherently too bright compared to the live experience and that something like the old "New England" speaker sound is a benefit to both listening pleasure and musical naturalness.
At the same time as the "New England" sound was typical of East Coast speaker manufacturers, there were designs like those from JBL and Altec which epitomized a much more present, brighter and superficially clearer "West Coast" sound. I think this sound came from the roots of such companies in providing sound reinforcement for movie theaters where the speakers were intended to be used behind a movie screen, a screen which was not totally transparent to high frequencies. When so used and heard from a large distance, these sounded fairly neutral, at least with movie soundtracks and whatever equipment and EQ was typically used in theaters. But up close in a small room, those who know the sound of live music know that these are, unlike real music, ear-bleed inducers. You haven't heard knives stuck in your ears until you hear the sound of unequalized Altec Lansing Voice of the Theater horns in a dorm room. One of my dorm neighbors in college was in a rock band and when not used on stage his VOTTs were his stereo speakers. He seemed to like them; he must have had some hearing loss even at that age.
Speaker colorations are much more subtle these days, to be sure. But the problem these days is that there are very few speaker companies making anything like a "New England" sounding speaker. Even the most respected brands tend to sound at least a bit bright on most commercial material compared to the real thing.
Some designs, like the Harbeths, other BBC-inspired speakers, get a fairly natural sound by relaxing the presence range of 2 - 5 kHz so that the return to flat above that range is not off-putting. Such designs may well sound airier in the top octave than the traditional New England sound and thus may better convey the amount of high frequency transients one hears live in concert (see the linked article above). But just comparing the midrange balance of AR speakers to Harbeths, it is immediately apparent that something like my 1970 vintage AR-5 or AR-3a speakers sound flatter through the midband than the Harbeths. The Harbeths may well be at least as pleasantly balanced overall and they are indeed at least an order of magnitude clearer in terms of revealed detail and lack of perceived distortion, but just in terms of midrange neutrality, a direct comparison shows that the Harbeth sound is "backed off" by the presence dip in on-axis response easily measurable with such speakers.
Maybe I'm just getting more nostalgic as I get older, but I think that if one has to make a choice between getting right the level of highs one hears in sustained notes versus what one hears on transients, I'd go for getting the sustained balance--the hall sound--correct every time. So, yes, while I think the classic New England sound does have less exciting/less lifelike high frequency transients than in real life, with much commercial material such speakers sound more naturally balanced overall than most modern speakers.
The one to five dB push up you typically see in on-axis frequency response measurements over part or much of the 2 to 20 kHz range in most modern speakers is very audible and not at all natural sounding. If anything, response should be relaxed in this range, not enhanced. I'd take a boost in the top octave over a boost in the 5 to 10 kHz region every time, though, since the former is most often heard as just more "air" or "sheen" than one hears live, and that is not a knives-in-ears problem. Look at the degree of excess measured in the top two octaves of B&W 800s for example, and you will realize quickly why cymbals are so much more prominent through such speakers than they are in life. See http://www.stereophile.com/content/bampw-800-diamond-loudspeaker-measurements Sure, the warm ported bass adds some balance, but to me it just sounds like a subtle boom and tizz effect. And yet these speakers have been highly praised.
There may be hope. For example, a few reviewers are now acknowledging that the newest Magnepan 3.7s and 20.7s for the first time integrate the ribbon tweeter with the lower range drivers in a seamless fashion. Yet, a few years ago, that ribbon was being touted as responsible for the best highs around. In my own auditions of ribboned Maggies over the years, it was not even a subtle effect. Yes, the highs were clean and airy and in this way state of the art. But it has also always been apparent to me that the highs stuck out like a sore thumb and were too high in level. In my own auditioning, I'd say that the 3.7 design moved in the right direction, but is still not there yet in terms of a natural amount of the top two octaves.
From the time I got into this hobby back in the 1960s, I've thought of the "New England" sound epitomized by companies like AR, KLH, Advent, EPI, Rectilinear, Dynaco and some others as "natural" sounding in terms of what one hears from live unamplified music in a good hall. When measured on axis, these speakers showed good solid bass with as much low-frequency extension as possible given the box size, a bit of warmth in the midbass, neutral mids and rolled highs above 4 kHz. That's pretty much the frequency response one hears in good halls from the audience, by the way. See the "classic" article on what you hear in a concert hall here, for example. Since most recordings are made with the mikes up close and high up (where there is generally less bass and more highs) rather than from an audience position, the speakers or some electronics in the home listening system needs to correct for the way the recordings were made, assuming the recording engineer did not fully correct for the mike placement. Decades of listening to classical music recordings on a variety of systems strongly suggests to me that most recordings are still inherently too bright compared to the live experience and that something like the old "New England" speaker sound is a benefit to both listening pleasure and musical naturalness.
At the same time as the "New England" sound was typical of East Coast speaker manufacturers, there were designs like those from JBL and Altec which epitomized a much more present, brighter and superficially clearer "West Coast" sound. I think this sound came from the roots of such companies in providing sound reinforcement for movie theaters where the speakers were intended to be used behind a movie screen, a screen which was not totally transparent to high frequencies. When so used and heard from a large distance, these sounded fairly neutral, at least with movie soundtracks and whatever equipment and EQ was typically used in theaters. But up close in a small room, those who know the sound of live music know that these are, unlike real music, ear-bleed inducers. You haven't heard knives stuck in your ears until you hear the sound of unequalized Altec Lansing Voice of the Theater horns in a dorm room. One of my dorm neighbors in college was in a rock band and when not used on stage his VOTTs were his stereo speakers. He seemed to like them; he must have had some hearing loss even at that age.
Speaker colorations are much more subtle these days, to be sure. But the problem these days is that there are very few speaker companies making anything like a "New England" sounding speaker. Even the most respected brands tend to sound at least a bit bright on most commercial material compared to the real thing.
Some designs, like the Harbeths, other BBC-inspired speakers, get a fairly natural sound by relaxing the presence range of 2 - 5 kHz so that the return to flat above that range is not off-putting. Such designs may well sound airier in the top octave than the traditional New England sound and thus may better convey the amount of high frequency transients one hears live in concert (see the linked article above). But just comparing the midrange balance of AR speakers to Harbeths, it is immediately apparent that something like my 1970 vintage AR-5 or AR-3a speakers sound flatter through the midband than the Harbeths. The Harbeths may well be at least as pleasantly balanced overall and they are indeed at least an order of magnitude clearer in terms of revealed detail and lack of perceived distortion, but just in terms of midrange neutrality, a direct comparison shows that the Harbeth sound is "backed off" by the presence dip in on-axis response easily measurable with such speakers.
Maybe I'm just getting more nostalgic as I get older, but I think that if one has to make a choice between getting right the level of highs one hears in sustained notes versus what one hears on transients, I'd go for getting the sustained balance--the hall sound--correct every time. So, yes, while I think the classic New England sound does have less exciting/less lifelike high frequency transients than in real life, with much commercial material such speakers sound more naturally balanced overall than most modern speakers.
The one to five dB push up you typically see in on-axis frequency response measurements over part or much of the 2 to 20 kHz range in most modern speakers is very audible and not at all natural sounding. If anything, response should be relaxed in this range, not enhanced. I'd take a boost in the top octave over a boost in the 5 to 10 kHz region every time, though, since the former is most often heard as just more "air" or "sheen" than one hears live, and that is not a knives-in-ears problem. Look at the degree of excess measured in the top two octaves of B&W 800s for example, and you will realize quickly why cymbals are so much more prominent through such speakers than they are in life. See http://www.stereophile.com/content/bampw-800-diamond-loudspeaker-measurements Sure, the warm ported bass adds some balance, but to me it just sounds like a subtle boom and tizz effect. And yet these speakers have been highly praised.
There may be hope. For example, a few reviewers are now acknowledging that the newest Magnepan 3.7s and 20.7s for the first time integrate the ribbon tweeter with the lower range drivers in a seamless fashion. Yet, a few years ago, that ribbon was being touted as responsible for the best highs around. In my own auditions of ribboned Maggies over the years, it was not even a subtle effect. Yes, the highs were clean and airy and in this way state of the art. But it has also always been apparent to me that the highs stuck out like a sore thumb and were too high in level. In my own auditioning, I'd say that the 3.7 design moved in the right direction, but is still not there yet in terms of a natural amount of the top two octaves.
