Speaker Measurements and What Constitutes Good or Great Speakers

0 Members and 1 Guest are viewing this topic. Read 7059 times.

DaveNote

  • Full Member
  • Posts: 615
  • Without music, life would be a mistake. Nietzsche
We see a constant theme in various postings here, especially about speakers. There are the technically oriented people who argue that if the measurements are "off" then the speakers can't sound right. There are others who say, the measurements notwithstanding, what we hear is what we hear, and the speakers are good.

I came across a very long and very techical three-part article written in the 90s by John Atkinson, who often does the measurements on speaker reviews for "Stereophile." Mostly far too technical for me. But obviously he is a great believer in the value of measurements, but he takes a measured (pun intended) view of them.

Overall, he writes, "The best-sounding loudspeakers, in my opinion, combine a flat on-axis midrange and treble with an absence of resonant colorations, a well-controlled high-frequency dispersion, excellent imaging precision, an optimally tuned bass, and also play loud and clean without obtrusive compression."

His conclusions about speaker measurements are these:

"While each measurement of a specific area of loudspeaker performance gives important information regarding possible sound behavior, it emerges that there is no direct mapping between any specific area of measured performance and any specific subjective attribute.(Emphasis added) As a result:

• Any sound quality attribute always depends on more than one measurement.

• No one measurement tells the whole story about a speaker's sound quality.

• Measuring the performance of a loudspeaker involves subjective choices.

• All measurements tell lies.

Most important, while measurements can tell you how a loudspeaker sounds, they can't tell you how good it is. If you carefully look at a complete set of measurements, you can actually work out a reasonably accurate prediction of how a loudspeaker will sound. However, the measured performance will not tell you if it's a good speaker or a great speaker, or if it's a good speaker or a rather boring-sounding speaker. To assess quality, the educated ear is still the only reliable judge. (Emphasis added)

And no matter how good any one measurement, if the beginning of the third movement of Beethoven's Fifth Symphony, where the composer introduces the trombones for the first time, or Jimi Hendrix's hammered-on tremolo at the start of "Voodoo Chile" on Electric Ladyland, doesn't send shivers down your spine, the loudspeaker is still doing something, somewhere, wrong."

Dave

James Tanner

  • Facilitator
  • Posts: 20864
  • The Demo is Everything!
    • http://www.bryston.com
So much depends on the room/speaker interface as well.

james

DaveNote

  • Full Member
  • Posts: 615
  • Without music, life would be a mistake. Nietzsche
So much depends on the room/speaker interface as well.

james

James, this is another reason why the "debates," arguments, whatever, among circle members about which is the better or best speakers are so futile. Best in what room? Better placed where in the room?

Dave

planet10

  • Industry Participant
  • Posts: 1945
  • Frugal-phile (tm)
    • planet10-hifi
One also must consider that most measures, take a snap shot of the "surface" and do not examine what is happening way down in level.

The analogy is that we are looking at the surface of a body of water, and not paying any attemtion to the bottom.

For instance when we do a FR sweep what is happening 40 dB down from that (at the same time). It is the subtle information "down there" that makes the difference between a good speaker & a great one.

Toole's Sound Reproduction likely has the best treatise on current measure to subjective correlation.

dave

Bobby_b

  • Jr. Member
  • Posts: 9
 
Ideally have a treated room or try and find a location that will take of advantage of the desirable reflections while diminishing the unwanted reflections, speakers will sound different from one room to another.

Cheers.

SoundGame

What's even more interesting is why, with measurement technologies so prevalent, so many speaker manufacturers just can't seem to even the basics right - horrible measurements are just not acceptable. 

The measurements don't tell you the entire story but they do get you going in the right direction.  Paul Barton of PSB has said - but don't quote me - that with the latest in measurements and auto CAD designs systems you can get a speaker about 80% to final but what makes the difference is that last 20%, which is by far, a trial and error / listen and adjust process.

Sasha

  • Full Member
  • Posts: 559
What's even more interesting is why, with measurement technologies so prevalent, so many speaker manufacturers just can't seem to even the basics right - horrible measurements are just not acceptable. 

The measurements don't tell you the entire story but they do get you going in the right direction.  Paul Barton of PSB has said - but don't quote me - that with the latest in measurements and auto CAD designs systems you can get a speaker about 80% to final but what makes the difference is that last 20%, which is by far, a trial and error / listen and adjust process.
Absolutely, and it is evident that majority of the proponents of "measurements mean nothing" approach are incapable of interpreting them, thus such aversion towards it.
Measurements are very good indicator of what may sound good and meet one’s objectives.

roscoeiii

Though no substitute for actual listening, paying attention to measureable characteristics has led to buying better and better sounding speakers IME. And as I have become more knowledgeable I have begun to move beyond freq. response and pay more attention to measurements such as off-axis response, phase coherence, time alignment, and decay plots. All are worth paying attention to, though as in most things, audio choices will require tradeoffs and compromises (according to budget, room etc).

DaveNote

  • Full Member
  • Posts: 615
  • Without music, life would be a mistake. Nietzsche
Absolutely, and it is evident that majority of the proponents of "measurements mean nothing" approach are incapable of interpreting them, thus such aversion towards it.
Measurements are very good indicator of what may sound good and meet one’s objectives.

Sasha, I agree with you and SoundGame. Measurements are a good indicator, as you write. But, as SoundGame and John Atkinson write, they don't tell the whole story and there is a portion (SoundGame says 20%) where subjectivity comes into play in designing and listening to speakers.

Dave

medium jim

Measurements aside, the human ear will always know which is the best speaker.  If people would just learn to trust their ears when buying and all of this speaker is the best, no this one is would vanish. 

At the end of the day, I only need to please my ears, not yours.  With that said, I agree that 80% is objective, but the vital and all important 20% is subjective.   Too much is referenced on stats, while they will for the most part weed out the also ran's, the end game still is subjective.

Nevertheless, there are things that are purely objective, those are the things that cannot effect the sound no matter how much some believe they do. 

Jim

roscoeiii

But like reviews, measurements can be useful in narrowing down the options to a manageable number for auditioning. (Not that you wouldn't possibly miss some excellent speakers by doing this)

James Tanner

  • Facilitator
  • Posts: 20864
  • The Demo is Everything!
    • http://www.bryston.com
I sure have been learning a lot about measurements and blind listening while I have been developing the Bryston speaker. :duh: :duh:

james

BobRex

What test measures signal compression?  I'm sure we can all name speakers that get so loud and then that's it.  Or other speakers that just have no dynamics.  Yet, I'm sure that these same speakers measure quite well with steady state signals.  Maggies, esp. older versions are speakers that comes to mind.  This to me is the essence of a good speaker, but I can't think of a single measurement that  indicates this feature.

Then there's the issue of time / signal coherence.  Theoretically, the ideal speakers are either Quad panels or Vandersteens (at least measurement wise).  After all, these are two of a small group that are truly coherent.  Hell, reviewers frequently talk about how listening to these speakers is a breath of fresh air, or some other superlative, and while these brands are highly regarded, less coherent speakers of the month continually make reference lists.  Why?

These are 2 examples of measurements that either aren't available or frequently disregarded.  So how valuable are other measurements?

planet10

  • Industry Participant
  • Posts: 1945
  • Frugal-phile (tm)
    • planet10-hifi
What test measures signal compression? ...

Then there's the issue of time / signal coherence.

If your only tool is a hammer, everything starts looking like a nail.

dave

DaveNote

  • Full Member
  • Posts: 615
  • Without music, life would be a mistake. Nietzsche
I sure have been learning a lot about measurements and blind listening while I have been developing the Bryston speaker. :duh: :duh:

james

James, this is a very interesting comment. Can you tell us what you've learned abiut measurements and blind testing?

Dave

James Tanner

  • Facilitator
  • Posts: 20864
  • The Demo is Everything!
    • http://www.bryston.com
James, this is a very interesting comment. Can you tell us what you've learned abiut measurements and blind testing?

Dave

Hi Dave

I can't see the speaker :lol:

james

DaveNote

  • Full Member
  • Posts: 615
  • Without music, life would be a mistake. Nietzsche
Hi Dave

I can't see the speaker :lol:

james

Very funny!  :lol:  Now, James, really, what have you learned about measurements and blind listening? It would be interesting to know, especially under this topic.

Dave

PRELUDE

Measurements do not tell you that how the speaker will sound like but it will tell you how it would perform and show you any problem that might be in existence.
BTW,Any time you buy or make any type of speakers regardless the price,name,size or etc....
There is a great chance to fail due to your listening habit and your room acoustics.

James Tanner

  • Facilitator
  • Posts: 20864
  • The Demo is Everything!
    • http://www.bryston.com
Very funny!  :lol:  Now, James, really, what have you learned about measurements and blind listening? It would be interesting to know, especially under this topic.

Dave

Hi Dave

It really removes the 'eyes' ability to seduce your reactions to a given speaker. I will try and give you a synopsis of what I have learned.

james

James Tanner

  • Facilitator
  • Posts: 20864
  • The Demo is Everything!
    • http://www.bryston.com
Very funny!  :lol:  Now, James, really, what have you learned about measurements and blind listening? It would be interesting to know, especially under this topic.

Dave

Hi Dave,

The process was interesting and it evolved in this manner:

1. I took a 3-way speaker that I currently like and listened to it first in a blind setup.
2. Then as each Model T speaker was developed I listened to it blind again in comparison to my reference
3. Started out with a simple 3-way version of the Model T active - 1-woofer, 1 mid, 1 tweeter.
4. Then tried 2 woofers - 2 mids 1 tweeter
5. Still not happy!
6. Then finally 3 woofers, 2 mids and 2 tweeters. - happy now.
7. Then replaced the reference speakers with the latest model T so that the positioning behind the screen was identical and listened some more
8. Since then we have been adjusting the front listening window curves, testing in the anechoic chamber and re-listening and assessing which combination of curves adds up to the best results (accuracy)
9. Now have them at home for further listening.
10.This has so far been with the Active Model T's but now we are building a Passive version to see how close we can come to the performance of the Active version without all the complexity the Active version brings with    it.

james