Discussion on loudspeaker measurements as they relate - or don't - to sound

[English210]

I asked a question in another thread about B&W seems to get stronger negative responses from forum members, here and elsewhere, despite having some very listenable speakers.

I am well aware that different strokes for different folks is never going to be more true than when dealing with a subjective hobby such as audio, nevertheless, B&W 'seems' to me to get more than their share of slamming.

The initial responses indicate that it may be partly due to less-than-ideal measurements on their speakers, and two links were posted to illustrate the point. This no doubt leads some to say that B&W can't/won't design and build speakers that measure well, and for the money they demand, that's inexcusable.

Because I am 'that kind', I can't leave it there, so I'm starting this thread to pursue my questions in regards to that point.

I am a hobbyist. I have no ax to grind, beyond learning for myself. I am in the early stages of choosing replacement mains, and so this plays into that partly, but largely it's an intellectual discussion (for which I am not nearly as well armed as many here!)

We know there are laws of physics involved in speaker design and performance. I am not qualified to know what all of those laws are, but believe they exist. Most manufacturers have their 'pet theories' as to what makes a good speaker, and most back it up with science of some sort (again, I can't comment on the veracity of much of this science, but I have read lots of ads, so I know many of the claims). There seems to be few hard and fast rules, besides 'air must move', hence the myriad of design philosophies. The marketplace will, to some degree, sift the wheat on some of these ideas, but many options remain.

We can all stipulate that there is no one right answer, and different designs will work better or worse in different rooms. My interest is in furthering the discussion of measurements as a predictor of performance. To those of you who know how to read the various graphs and numbers, PLEASE DO NOT ATTEMPT TO IMPRESS US WITH YOUR KNOWLEDGE. I will stipulate that you know a lot more than I, or I wouldn't be asking

Are B&W (for example) guilty of something by selling speakers that don't measure as well as they could, or are they simply aiming at an uninformed 'victim' of their marketing strategy? Or (as I am inclined to believe, truth be told), are they tuning their speakers for a particular sound (no doubt researched to no end to appeal to as broad a market as possible), and measurements be damned. This will not doubt lead to discussions of 'is complete and total accuracy of reproduction the pinnacle of the speaker makers art?'

Related to this, can we as consumers use measurements as we shop? WPC ratings, and basic FR numbers are largely useless IMHO, sensitivity ratings less so, but even then, the demands the speaker will place on the amp aren't usually predictable from the spec sheet in most speaker ads. So what can we use? How do non-techies 'use' specifications and measurements to evaluate a speaker?

Thank you for participating

 

[Steve81]

A little light reading that you might find helpful:
Why We Measure Audio Equipment Performance | Audioholics

 

[Irvrobinson]

I have only heard the 802D and the 800D, so I can't comment on all B&W speakers. For these two models that I have heard, IMO they demonstrate a classic emphasis in the upper midrange and the bass that makes them sound more exciting in a dealer demo. They appear "voiced" to sound exciting, rather than completely neutral, and I seem to be particularly sensitive to it. I have a friend who owns 802Ds, and he thinks the Salon2s I use sound too reticent and unexciting for his tastes.

It is no surprise that speaker buyers have taste-related preferences. The same is true for musical instruments. Pianos sound different, for example, and people have preferences for how they think a piano should sound. Some people like a brighter, more forward sound, while others tend to like richer-sounding pianos with more bass emphasis. For years people have noticed that German Steinways sound brighter than American Steinways. No one considers one Steinway better than another, it is simply a matter of taste.

The problem with speakers is that they are not supposed to have a character of their own, and "voicing" is another word for coloration. The impression of coloration is the primary issue I have with the 802D and the 800D. (I also find their imaging to be less than competitive at their price levels, but that could be room-dependent.) In my extensive experience, recordings sound "more different" from one another on the Revels (and the KEF 207/2 and the Sound Labs and the YGs I've heard) than they do on the B&Ws, which is my unscientific indicator I use during auditioning. If someone likes the B&Ws better than other speakers I completely understand it, but if neutrality is the goal then the 800D and 802D seem a notch lower on the quality scale than the others I've listed.

 

[j_garcia]

I agree that "Voicing" = coloration and it is the designer's taste. What you are hearing is the sound profile that the designer(s) have chosen. So just like wine or beer, not everyone likes the same thing. I've listened to a fair number of B&Ws and don't find them unpleasant, though I don't mind what many refer to as the "British sound" either. That said, I have never liked any of the B&Ws enough to own them.

 

[ski2xblack]

This is a complex topic, far too much for a forum post to adequately cover (particularly from me, I'm just a hobbyist music lover like yourself). It's such a complex topic that people have written entire books about it. One such book is Toole's, which I picked up some years ago and have found it a fantastic resource.

It's also helpful to understand your frame of reference. We tend to like what we are most familiar with. You can adjust your frame of reference by choosing more linear speakers, or using live unamplified instruments as the reference. Just based on the guys who post here, there seems to be a convergence of preferences with linearity the more the frame of reference is live, un-amplified music. (This brings up a caveat/grype of my own, that many speakers fail to respect the dynamic range of real instruments. I prioritize dynamic range up there with frequency response, maybe even higher.)

So measurements, in lieu of actually hearing the speakers, can give an idea of how they compare to what you are familiar with, and if you're slightly more sophisticated, perhaps an idea of how they'll play in your environment. Because speakers with flat on axis and very smooth off axis measurements tend to play well in many different environments, it can be used to identify the contenders from the pretenders and narrow down the choices when making buying decisions.

All that being said about their usefulness, the whole point to this endeavor is goosebumps as the music takes us on a journey. The only measurements I fully trust are the ones taken at the listening position in my room, and those taken by my ears/brain. Mid/far field measurements from the listening position within that particular room are the closest thing to what your ears get. But that concerns system calibration, a different topic than judging speakers based on published measurements.

 

[AcuDefTechGuy]

Pianos sound different, for example, and people have preferences for how they think a piano should sound. Some people like a brighter, more forward sound, while others tend to like richer-sounding pianos with more bass emphasis. For years people have noticed that German Steinways sound brighter than American Steinways. No one considers one Steinway better than another, it is simply a matter of taste.

So will the same exact note (say C-major) sound differently on 2 different Steinway pianos? Which note is more accurate?

If so, then how are speakers any different?

My preference is pretty simple - whatever sounds crystal clear + awesome bass.

I think B&W speakers (Diamonds, CMs, 600s, etc.) can sound great, but only in a good room and setup. They probably sound bad 50% of the time at stores like Best Buy or shops & dealers.

 

[Irvrobinson]

So will the same exact note (say C-major) sound differently on 2 different Steinway pianos?

If so, then how are speakers any different?

Yes, the same note on two pianos can and often will sound different, because a piano key, like Middle C, does not just produce a fundamental frequency, there are harmonics. Pianos are designed to produce different overall sounds, and so is every other instrument. Two violins can sound different playing the same note. I know from personal experience that flutists listen to different head joints on their own flute and choose which one they like by sound. Speakers are completely different, they are supposed to be neutral.

 

[AcuDefTechGuy]

As much money as B&W have, I think they can easily make all their speakers flat.

But they choose NOT to.

Some people prefer flat-measuring speakers, some people prefer a little more vacillation and excitation.

 

[AcuDefTechGuy]

Speakers are completely different, they are supposed to be neutral.

Pretend we're in a debate class and I am suppose to take the opposite stance because I'm the class clown.

On who's authority are you basing that statement on?

Who is to say that speakers are supposed to measure FLAT & neutral, instead of fun and exciting?

Did the lord himself say so?

Is it written in law?

 

[3db]

Yes, the same note on two pianos can and often will sound different, because a piano key, like Middle C, does not just produce a fundamental frequency, there are harmonics. Pianos are designed to produce different overall sounds, and so is every other instrument. Two violins can sound different playing the same note. I know from personal experience that flutists listen to different head joints on their own flute and choose which one they like by sound. Speakers are completely different, they are supposed to be neutral.

What I think you are saying is that Middle C is still a MiddleC and its fundamental frequency if tuned properly is accurate across all piano makes. Its the different harmonics that fill in to create the different sound of a middle C that makes it sound different among pianos. The different sound of the Middle C is not inaccurate. It becomes a preference at that point.

Loudspeakers however should be as neutral as possible so that they can as accurately reproduce what is being played on the recording. A neutral speaker is easier to tune with a room than a speaker that is purposely voiced to a certain sound.

 

[AcuDefTechGuy]

Loudspeakers however should be as neutral as possible so that they can as accurately reproduce what is being played on the recording.

But how do we know what the actual original recording was in PERSON?

Which piano, violin, guitar did they use?

Will the instruments sound differently if you sit 3 ft away vs 6 ft away vs 12 ft away vs 24 ft away?

Will the instruments sound differently depending on the venues?

How do we know the exact conditions?

 

[Irvrobinson]

Pretend we're in a debate class and I am suppose to take the opposite stance because I'm the class clown.

Pretend we're in debate class? You mean this isn't a debate class? And you're not one of the class clowns? I'll try...

On who's authority are you basing that statement on?

My second-grade math book. 1+1 = 2, but 1+0 = 1. If a piano = 1, do you want your speaker reproducing, conceptually, 1 or 2?

Who is to say that speakers are supposed to measure FLAT & neutral, instead of fun and exciting?

Only those people that want a speaker to be a reproducer rather than a musical instrument of some sort.

 

[Irvrobinson]

But how do we know what the actual original recording was in PERSON?

Which piano, violin, guitar did they use?

Will the instruments sound differently if you sit 3 ft away vs 6 ft away vs 12 ft away vs 24 ft away?

Will the instruments sound differently depending on the venues?

How do we know the exact conditions?

The differing instruments are irrelevant. The problem is that if your speakers are colored they will apply those same colorations to every recording, regardless of what the original instruments or conditions were. The colorations will produce a "sameness" to the reproduced performances, and that is audible by its relative *absence* on more neutral speakers.

 

[AcuDefTechGuy]

My second-grade math book. 1+1 = 2, but 1+0 = 1. If a piano = 1, do you want your speaker reproducing, conceptually, 1 or 2?

Only those people that want a speaker to be a reproducer rather than a musical instrument of some sort.

If every piano sounds differently, than how can ALL PIANOS be = 1?

If every piano sounds differently, how do you know the EXACT sound being reproduced? How do you know it is just that one Steinway and not another?

Isn't the PRIMARY purpose of this hobby for FUN and enjoyment?

Who is to say that a speaker has to measure flat or sound a certain way in order for people to enjoy and have fun?

We're not talking about performing surgery or building bridge or dispensing the correct medication here. It's not life and limb.

Speakers are supposed to be FUN. And everyone has a different definition of fun - flat, neutral, emotional, colorful, etc.

 

[AcuDefTechGuy]

The differing instruments are irrelevant. The problem is that if your speakers are colored they will apply those same colorations to every recording, regardless of what the original instruments or conditions were. The colorations will produce a "sameness" to the reproduced performances, and that is audible by its relative *absence* on more neutral speakers.

How do you know they are "colored"?

How do you know what the exact original sound is to compare to?

What if the original sound is "colored" depending on the instruments, distance, venues?

 

[Irvrobinson]

If every piano sounds differently, than how can ALL PIANOS be = 1?

If every piano sounds differently, how do you know the EXACT sound being reproduced? How do you know it is just that one Steinway and not another?

Isn't the PRIMARY purpose of this hobby for FUN and enjoyment?

Who is to say that a speaker has to measure flat or sound a certain way in order for people to enjoy and have fun?

We're not talking about performing surgery or building bridge or dispensing the correct medication here. It's not life and limb.

Speakers are supposed to be FUN. And everyone has a different definition of fun - flat, neutral, emotional, colorful, etc.

I think I just answered this in post #13. Look, ADTG, if you like colored speakers, because they're fun, that's fine with me.

 

[Irvrobinson]

How do you know they are "colored"?

How do you know what the exact original sound is to compare to?

What if the original sound is "colored" depending on the instruments, distance, venues?

Because in some cases I made the recordings. I heard the original instruments.

 

[English210]

Move the piano from venue A to venue B, and how does the sound change. It will of course. Move those 'colored' B&W speakers from an anechoic chamber into my live-as-heck room, and they will change their sound. Back into the dealers dead as a doornail room and the sound changes again.

The measurements haven't changed, the design colorations, whether there on purpose or through faulty design, are still there, but given a 'wrong-enough' environment, those same B&W's everyone pans for measuring badly could well sound more accurate than brand X, Y, or Z, correct?

 

[j_garcia]

ALL speakers are colored because by definition they are a reproducing device. The goal would be "least colored". That coloration of the sound however, is part of the preference that individuals have for certain speakers.

 

[3db]

But how do we know what the actual original recording was in PERSON?

Which piano, violin, guitar did they use?

Will the instruments sound differently if you sit 3 ft away vs 6 ft away vs 12 ft away vs 24 ft away?

Will the instruments sound differently depending on the venues?

How do we know the exact conditions?

All irrelevant because at that point, all we can do is play the recording. A recording is a point in time best acoustic capture possible. We are powerless to change the recording. But if I don't like the sound of teh recording coming form the speaker and I want to add more bass, midrange or treble or take some away, its easier done on speaker that is neutral than having one of those factors emphasized.