Wilson Audio Speakers: Why do people like them?

[Beave]

He explains about the measurements in this video @4:40
Link:

- YouTube

Enjoy the videos and music you love, upload original content, and share it all with friends, family, and the world on YouTube.

tinyurl.com

OK, I just wasted my time looking at the video.

1. It's an Andrew Robinson video.

2. You said that he gives measurement details at 4:40, but I watched and saw no details whatsoever, just a claim that it's remarkably flat other than the bass hump.

So his measurement doesn't tell us anything, but he thinks it does because, hey, it's a measurement!!

 

[Pearljam5000]

OK, I just wasted my time looking at the video.

1. It's an Andrew Robinson video.

2. You said that he gives measurement details at 4:40, but I watched and saw no details whatsoever, just a claim that it's remarkably flat other than the bass hump.

So his measurement doesn't tell us anything, but he thinks it does because, hey, it's a measurement!!

I didn't make the video

 

[Buckchester]

1. It's an Andrew Robinson video.

 

[Arindal]

My MBL Omnis, which obviously did not put out a downward tilt off axis, never sounded bright, or too bright, to me in my 13 x 15 room - and that was the case whether I was cutting down room reflections or leaving the walls more reflective.

No audiophile pals found them bright either.

If the reverb in a room is in tonal balance to the direct sound, even if overly dominant, and both are more or less linear, it does neither appear to be bright nor annoying, but rather smooth, diffuse, silky, balanced, distant, however you want to call it. Me thinks that Dr. Amar Bose has discovered this phenomenon in the mid-1960s already and constructed his (in)famous 901 speaker based on this idea.

A flat in-room response will sound thin and bright. That is well known.

It is not well-known, it is simply not the case. There is no indication why a flat response should in general sound thin and bright, if not being the result of a thin and bright anechoic response. In the latter case, well, it might sound exactly like that.

It should have a gentle downward slope as you progress through the frequency range.

The gently tilted downwards slope will for sure lead to a dull, lower midrange/presence-laden tonality and cause all sorts of unwanted effects deviating from a neutral response. With the exception of the highest treble bands (7K+ Hz), which might be subject to a gentle decrease in level due to dissipation and absorption of natural materials without being perceived as overly dull.

Human hearing evolved in environments where high frequency sound was almost always being damped by air, vegetation, terrain, and other organic materials. Because of that, we're not used to hearing strong upper treble energy relative to the rest of the spectrum unless the sound source is very close .Over time, this shaped what our brains interpret as “natural” sound. As a result, a slightly recessed treble often comes across as more realistic, while too much upper treble energy can easily sound artificial or fatiguing.

With all due respect, but this does not sound plausible to me.

If assumed that there is such general ´preference for dull timbre, allergy to treble´in humans, would it not be inherently integrated in all existing recordings? Let us not forget, studio monitoring is usually done under more linear conditions, with tonality being closer to flat in-room response, due to better broad-band absorption in studio control rooms, nearfield monitors being used, boosted bass/lower midrange being avoided.

So if too much of treble is in general annoying and perceived as ´not natural´ (which I doubt listening to the majority of modern pop music productions), recordings would be mixed and mastered in a more dull, bass/lower midrange heavy fashion compared to how they will sound in any home listening environment. So why should the latter add another layer of treble attenuation/dullness? It is counterintuitive.

I don´t mean to say that any method to achieve flat in-room response will lead to a balanced tonality. Boosting treble under anechoic conditions, particularly for some narrow-banded peaks only, might lead to annoying or overly bright sound.

 

[PGansz]

A flat in-room response will sound thin and bright. That is well known. It should have a gentle downward slope as you progress through the frequency range.

The measurement of that watt puppy really doesn’t give us enough info. But it does look like it will sound a little bright. It also looks smoothed. We would need a little more info to see directivity behaviour and to see if there are any resonances. But this can’t be done in room for frequencies above the room transition frequency. Those would need to be anechoic (or properly gated). Because a microphone will sum the direct and reflected sounds, whereas two ears and a brain can differentiate between the two above this range.

Isn't this only the case of the speaker is pointed directly at you? If you point a bright speaker straight into a room and you sit between them at a 20-45 angle it is no longer bright (for most rooms) but actually neutral.

 

[Sokel]

Isn't this only the case of the speaker is pointed directly at you? If you point a bright speaker straight into a room and you sit between them at a 20-45 angle it is no longer bright (for most rooms) but actually neutral.

Vertical response will not change much though, or it depends also.
Unless you have a treated ceiling with cloud and thick carpet (but even then! ) one has to take it into account too.

 

[Buckchester]

If the reverb in a room is in tonal balance to the direct sound, even if overly dominant, and both are more or less linear, it does neither appear to be bright nor annoying, but rather smooth, diffuse, silky, balanced, distant, however you want to call it. Me thinks that Dr. Amar Bose has discovered this phenomenon in the mid-1960s already and constructed his (in)famous 901 speaker based on this idea.


It is not well-known, it is simply not the case. There is no indication why a flat response should in general sound thin and bright, if not being the result of a thing and bright anechoic response. In the latter case, well, it might sound exactly like that.



The gently tilted downwards slope will for sure lead to a dull, lower midrange/presence-laden tonality and cause all sorts of unwanted effects deviating from a neutral response. With the exception of the highest treble bands (7K+ Hz), which might be subject to a gentle decrease in level due to dissipation and absorption of natural materials without being perceived as overly dull.



With all due respect, but this does not sound plausible to me.

If assumed that there is such general ´preference for dull timbre, allergy to treble´in humans, would it not be inherently integrated in all existing recordings? Let us not forget, studio monitoring is usually done under more linear conditions, with tonality being closer to flat in-room response, due to better broad-band absorption in studio control rooms, nearfield monitors being used, boosted bass/lower midrange being avoided.

So if too much of treble is in general annoying and perceived as ´not natural´ (which I doubt listening to the majority of modern pop music productions), recordings would be mixed and mastered in a more dull, bass/lower midrange heavy fashion compared to how they will sound in any home listening environment. So why should the latter add another layer of treble attenuation/dullness? It is counterintuitive.

I don´t mean to say that any method to achieve flat in-room response will lead to a balanced tonality. Boosting treble under anechoic conditions, particularly for some narrow-banded peaks only, might lead to annoying or overly bright

Some comments from Floyd Toole about room curve targets, room EQ and more

Room curve targets Every so often it is good to review what we know about room curves, targets, etc. Almost 50 years of double-blind listening tests have shown persuasively that listeners like loudspeakers with flat, smooth, anechoic on-axis and listening-window frequency responses. Those with...

www.audiosciencereview.com

 

[Arindal]

Some comments from Floyd Toole about room curve targets, room EQ and more

Room curve targets Every so often it is good to review what we know about room curves, targets, etc. Almost 50 years of double-blind listening tests have shown persuasively that listeners like loudspeakers with flat, smooth, anechoic on-axis and listening-window frequency responses. Those with...

www.audiosciencereview.com

I have discussed the matter with Dr. Toole, and he pointed out that the Harman curve is what he called a ´result curve´ of existing, subjectively most preferred loudspeakers models they used for comparison at the time, all of them linear on axis and flawed this or that way in terms of directivity. He even mentioned a common directivity error in the transitional band between midrange and tweeter, which indicates these speakers were not even meeting the basic standards for constant directivity when the term was popularized some 30 years ago. A test involving constant directivity speakers without flaws, resulting in a flat in-room curve (at least below 7K), has not been included in his or Dr. Olive´s tests, as such speakers are only commercially available only for about 15 years, after the last tests had already ended.

If you point a bright speaker straight into a room and you sit between them at a 20-45 angle it is no longer bright (for most rooms) but actually neutral.

There are indeed such speakers on the market, if I am not mistaken, Dali is favoring such concept. Such behavior is to be expected when narrowing treble directivity (e.g. caused by huge tweeters, planar tweeters, non-CD horns, fullrange drivers or alike) and linear response under angles (such as a desired 30deg horizontal listening window when placing speakers parallely to the wall) come together.

It is not a general rule, though, and applicable only to a limited number of speakers.

 

[Subob]

I heard some Watt Puppy's ( original 80's model, I think) in a secondhand vintage hifi shop last year. They were being powered by some rather large mono blocks and an amp that I don't remember. Yes, the shop owner was playing vinyl.

The shop is of reasonable size and has pretty good acoustics, to my untrained ear. I think the sight of 'vintage' hifi all over the shop, records for sale, nice coffee, relaxed atmosphere and letting you play your own tunes all helps to rule out any objective listening and turn most experiences into subjectively positive ones. He's not a salesman either, more of a collector with money who never really tries to sell you anything, lol.

Anyway, scene set, what did I think of the Puppys? Well, I was suitably impressed by the rigidity and solidity of the cabinets. The finish was nice, too. From memory they seemed to have a lot of bass 'punch' as their strong point, but that was at volumes higher than I generally listen to at home. Did I think they were worth the money when I was told the price?, not especially, no. I didn't dislike them, though.

Out of the speakers I've heard in that shop, I think the ones that I enjoyed the most were the Kef Model 105/2. Those funny ones on wheels with a separate mid range and tweeter housing that you can move around on top of the sealed unit woofer. The bass response was among the 'tightest' I've heard in a speaker ( I haven't heard loads of speakers, tbf). For sheer bang for your vintage buck enjoyment, those old Kefs sounded great to me.

 

[Buckchester]

I have discussed the matter with Dr. Toole, and he pointed out that the Harman curve is what he called a ´result curve´ of existing, subjectively most preferred loudspeakers models they used for comparison at the time, all of them linear on axis and flawed this or that way in terms of directivity. He even mentioned a common directivity error in the transitional band between midrange and tweeter, which indicates these speakers were not even meeting the basic standards for constant directivity when the term was popularized some 30 years ago. A test involving constant directivity speakers without flaws, resulting in a flat in-room curve (at least below 7K), has not been included in his or Dr. Olive´s tests, as such speakers are only commercially available only for about 15 years, after the last tests had already ended.



There are indeed such speakers on the market, if I am not mistaken, Dali is favoring such concept. Such behavior is to be expected when narrowing treble directivity (e.g. caused by huge tweeters, planar tweeters, non-CD horns, fullrange drivers or alike) and linear response under angles (such as a desired 30deg horizontal listening window when placing speakers parallely to the wall) come together.

It is not a general rule, though, and applicable only to a limited number of speakers.

I’m trying to understand your point. They tested many speakers, I don’t understand why they would have to test speakers that have perfect or constant directivity to arrive at their conclusions.

Speakers with flat anechoic response and smoother directivity were preferred over those that were not. These types of speakers will measure flat with a downward slope in room.

A speaker that measures flat in room would have an upward tilting anechoic response.

Wouldn’t a speaker with a flat anechoic response and perfect directivity also measure with a downward slope in room because the high frequency energy gets absorbed thought the room?

I don’t understand your logic.

 

[NTK]

I’m trying to understand your point.
...
I don’t understand your logic.

I also don't understand what points Arindal was trying to make. I think the comments were about the flat in-room response of the Watt/Puppy in post #364. I will bet good money that the Watt/Puppy is far from constant directivity, and Dr. Toole's observations on the downward sloping in-room response applies.

 

[Arindal]

I don’t understand why they would have to test speakers that have perfect or constant directivity to arrive at their conclusions.

They would have had to test neutral, constant directivity speakers as well (which did not exist at the time) and find them to be less preferred, to back the claim that a downward-tilted in-room response is preferred over a flat one (400...7,000Hz) and their result curve is in general correlated with preference.

Speakers with flat anechoic response and smoother directivity were preferred over those that were not.

No doubt about that. Anechoically neutral speakers with increasing, i.e. flawed, directivity, were preferred over other speakers with presumably more dramatic flaws. Not really a solid justification for declaring the tilted curve a result curve of general preference, in my understanding.

Wouldn’t a speaker with a flat anechoic response and perfect directivity also measure with a downward slope in room because the high frequency energy gets absorbed thought the room?

No, it does not. The in-room curve of such a speaker is surprisingly linear, at least up to the region of 7K. In the highest treble bands, a minimum of downward slope is expected, but not due to dissipation but rather increasing absorption grade of materials in the room (and practically increasing directivity of existing speakers).

 

[Buckchester]

They would have had to test neutral, constant directivity speakers as well (which did not exist at the time) and find them to be less preferred, to back the claim that a downward-tilted in-room response is preferred over a flat one (400...7,000Hz) and their result curve is in general correlated with preference.



No doubt about that. Anechoically neutral speakers with increasing, i.e. flawed, directivity, were preferred over other speakers with presumably more dramatic flaws. Not really a solid justification for declaring the tilted curve a result curve of general preference, in my understanding.



No, it does not. The in-room curve of such a speaker is surprisingly linear, at least up to the region of 7K. In the highest treble bands, a minimum of downward slope is expected, but not due to dissipation but rather increasing absorption grade of materials in the room (and practically increasing directivity of existing speakers).

Can you give us examples of such speakers with their measurements?

 

[Blockader]

I have W371 + 8361s, which have more or less constant DI above 60 hz. They measure close to flat in my room, yet sound horrible(very very bright) unless I apply a high shelf filter.
During Toole's research era, there weren't many constant DI speakers available for comparison against designs with gradually decreasing DI. That said, the constant DI speakers that did exist performed quite well in fact, some scored among the best: the JBL M2(half constant) and Revel Ultima Salon 1(constant thanks to tweeters in the back).
Where I agree with Arindal: constant DI can sound really good. Maybe even better! Where we differ: I believe constant DI only sounds good when paired with a consistently rolling-off on-axis response to accommodate it. Dr.Geddes was also defending this speaker design idea.

I unignored him today to see what he's been on about. Turns out he's become even more deranged to the point where he's now defending flat in-room response. Well, back to ignore then.

 

[PGansz]

I have W371 + 8361s, which have more or less constant DI above 60 hz. They measure close to flat in my room, yet sound horrible(very very bright) unless I apply a high shelf filter.
During Toole's research era, there weren't many constant DI speakers available for comparison against designs with gradually decreasing DI. That said, the constant DI speakers that did exist performed quite well in fact, some scored among the best: the JBL M2(half constant) and Revel Ultima Salon 1(constant thanks to tweeters in the back).
Where I agree with Arindal: constant DI can sound really good. Maybe even better! Where we differ: I believe constant DI only sounds good when paired with a consistently rolling-off on-axis response to accommodate it. Dr.Geddes was also defending this speaker design idea.

I unignored him today to see what he's been on about. Turns out he's become even more deranged to the point where he's now defending flat in-room response. Well, back to ignore then.

What happens when you point the speaker straight ahead and sit off-axis?

 

[Arindal]

Can you give us examples of such speakers with their measurements?

Unfortunately, for many of those which I am familiar with, there is no data publicly available. A very good example for which this is the case is GGNTKT M3. Another prominent example of such concept is D&D 8c, although it comes with a certain degree of compromise in terms of directivity, but certainly good enough to make the idea clear.

Turns out he's become even more deranged to the point where he's now defending flat in-room response.

Just to be clear here: You call the very idea of linear in-room response as a result of linear anechoic response plus flat indirect soundfield leading to good tonal balance, ´deranged´?

I do not mean to defend the WA speakers in question or say all this applied to them. When I heard them (briefly), some tonal changes under different angles were apparent in the lower band of the tweeter, which could be taken as a hint that some lobing and edge diffraction issues are at play. So I assume a lot of effort has been put into positioning and toeing-in to achieve both the in-room response and tonality described in the video, but I do not doubt that this is possible.

I have W371 + 8361s, which have more or less constant DI above 60 hz. They measure close to flat in my room, yet sound horrible(very very bright) unless I apply a high shelf filter.

Did you ever try to figure out what makes them sound horrible and very very bright while being linear on axis? Sounds to me like there is some serious issue that does not have much to do with response curve or the speakers.

Whenever I compared them to more or less constant directivity speakers (w/o sub), they sounded rather on the dull, soft, non-brilliant side, at times even bass-laden or lower midrange heavy under typical home listening conditions.

During Toole's research era, there weren't many constant DI speakers available for comparison against designs with gradually decreasing DI.

In my understanding, before 1991, there was no such design on the market, but i might be wrong here.

Where I agree with Arindal: constant DI can sound really good. Maybe even better! Where we differ: I believe constant DI only sounds good when paired with a consistently rolling-off on-axis response to accommodate it.

Could you elaborate please on the question why is this so and who has come to this conclusion under controlled listening test conditions?

From both theoretical point of view as well as practical experience in studio monitoring, everything speaks against such theory. The majority of existing recordings professionally mixed and mastered were created under studio conditions which do not show such roll-off. Even if there is a certain amount of tonal variety when it comes to studio monitoring conditions, flat anechoic response and significantly underrepresented indirect sound in the lower frequency bands (due to nearfield monitoring, room treatment or alike) or anything close to that is a widespread standard.

So why should the same recordings in studios be affirmed under comparably linear, ´very very bright-sounding´ (your words!) conditions, while in homes they sound best with boosted bass/lower midrange and attenuated treble? Does not make sense to me.

 

[RexrothPigeon]

@Arindal - You argue passionately, and you are clearly possessed with strong conviction. For all I know you may be right about all of your bold claims (I know vanishingly little). But despite exhortations you have provided no data, despite claiming to have tested thousands of speakers. What's the deal? And lacking any empirical support, why should anyone here countenance such heterodox ideas?

 

[Mort]

@Arindal - You argue passionately, and you are clearly possessed with strong conviction. For all I know you may be right about all of your bold claims (I know vanishingly little). But despite exhortations you have provided no data, despite claiming to have tested thousands of speakers. What's the deal? And lacking any empirical support, why should anyone here countenance such heterodox ideas?

Ignore is the best option

 

[RexrothPigeon]

Ignore is the best option

If I had any real conviction that he's wrong, I wouldn't hesitate. But as I said, I have no idea.

 

[Vacceo]

He´s the friendly version of "trust me bro", and essentially, not different from Darko, Gutemberg and so on.