How do you test the quality of a speaker?


Apart from how flat and wide the frequency response is, what other measurements do you need to test the quality of a speaker? Why dont these measurements ever appear on manufacturers websites? Surely that would be the definite evidence of the quality? 

Could it be possible that we are being deceived? 


kenjit
I listen to them/speakers with the music I appreciate/like/love - PERIOD(.)

DeKay

I listen to them/speakers with the music I appreciate/like/love - PERIOD(.)
That wasnt my question. 
By the quality of what you hear.
By the quality of what you hear.
Thats not the answer I'm looking for
Henry Ford estimated accounts payable by weighing the stack of bills, so I just weigh the packaging....
That WAS your question, and that WAS my answer as I "measure" by listening to MUSIC.

Do you have a better way of choosing speakers to own, and actually listen to music through?

If so, then please enlighten me/US.

DeKay
No, I let the speaker lab techs  do all that specs,  for me. 
I trust in labs I know are superior craftsmen. Like Honda, I know for a  fact they are superior, .
same with speakers, i buy whats a  superior lab. 
kits kits kits, thats my game
No, I let the speaker lab techs do all that specs, for me. I trust in labs I know are superior craftsmen.
Thats fine but you havent answered the question. What other measurements do you need to test the quality of a speaker? 
How do you test the quality of a speaker?

By listening to it. 

Why would I want to measure a speaker? Any speaker? Why? If it measures bad but sounds good, am I gonna not buy what sounds good simply because it doesn't measure good? Really?

Or what if it measures good but sounds bad? I am gonna buy it and when people come over turn to them and say yeah I know but would you look at this graph?

What is the point of measuring a speaker, kenjit? Seriously. Why? Even you yourself say they all need to be hand tuned to each persons ears. So why not just listen and go with what sounds good? Cut out the middleman. You said it yourself, we all hear differently. So why not just go with what we hear?
 I "measure" by listening to MUSIC.
Thats not what I meant by measurements. I am talking about using equipment such as microphones to test the performance of the speaker. 
Do you have a better way of choosing speakers to own, and actually listen to music through?If so, then please enlighten me/US
You havent told me what measurements are currently used to assess the quality of a speaker. Are you suggesting there are none other than a frequency response chart?
@millercarbon We need to use scientific methods to assess our speakers. Listening alone is inadequate. Are you suggesting we just take a piece of wood, cut it into pieces, stick them together and stick a bunch of capacitors and inductors together until it all sounds right to our ears? That is not how its done. Even cutting a piece of wood requires measurements to check you are cutting the correct length and size. You cant simply judge by looking can you? A crossover can be measured to check it produces the required slope, which ultimately affects the sound we hear. We know that a frequency response measurement of the speaker affects what we hear. Unfortunately it does not tell us everything we hear. The obvious question is, what other measurements do we need to objectively assess what we hear?
Take a look at a Stereophile review for example and you will gain insight on how to measure speakers in meaningful ways.

Here is a good one:

https://www.stereophile.com/content/tekton-design-impact-monitor-loudspeaker-measurements
"Thats not what I meant by measurements. I am talking about using equipment such as microphones to test the performance of the speaker. "

"Are you suggesting there are none other than a frequency response chart?"

How do you gather that I was suggesting ("Are you suggesting there are none other than a frequency response chart?") such?

Please give me a clue as to how you reached enough of a conclusion to ask.

Duh:

I was SIMPLY making a note that I MEASURE speakers by listening to them with music that I like/prefer to listen to.

Why would I need a microphone and/or other gear to accomplish the MEASUREMENTS that are most appropriate for the end use/function of the speakers?

I have used mikes and sound generators years/decades ago to set up systems, but this was for speakers that the end user already liked/preferred (the speakers/systems were, in the end, always finally tuned/equalized by ear).

DeKay


There are a few that provide, more than one 1 meter, 1 watt, tone burst at given axis points and at a given frequency. That is the conventional way.

NOW some are adding frequency responses in 10 watt increments. That information along with a dead silent enclosure and final assembly, and XO trim and match. There is a lot of information on some.  BE aware a LOT of them CHEET to get the numbers though.. LOL Yes they do..

3 drivers almost always leads to some weird wiring CRAP,  to get a flat response.

I suggest you get a Kit from Parts Express for 2-400.00 dollars. It comes with software, mic, and calibration.
A lot of the stuff to give you what you need, not me. I have my testing EQ, that works well for me. I just updated a few things, added a second mic, and coupled that with an active OXO with on the fly correction...
You should have software to test the individual drivers, Q, impedance curve, breakup, la-te-da...make the corrections to your crossovers and you TUNED your own stuff..

The quality of the question you ask, often  reflects to quality of the answer you get. Better questions, better answers.. 

Regards

Regards
If it measures bad but sounds good, am I gonna not buy what sounds good simply because it doesn't measure good? Really?

What sounds good is not an accurate descriptor of whats going on. Frequency response is a precise descriptor. The result should be the same every time you do the measurement if you use the same method. 
It would not vary according to the time of day, your mood, or your hearing. Measurements tell us reality. Our ears often do not.

If we had the ability to measure the quality of a speaker, we would not need to try to use our fallible hearing to compare whether speaker A is better than speaker B. The answer would be incontrovertible unlike our opinions.

Measurements are needed to confirm what we think we hear. 
"Thats not what I meant by measurements. I am talking about using equipment such as microphones to test the performance of the speaker. "

"Are you suggesting there are none other than a frequency response chart?"

How do you gather that I was suggesting ("Are you suggesting there are none other than a frequency response chart?") such?

Please give me a clue as to how you reached enough of a conclusion to ask.

Duh:

I was SIMPLY making a note that I MEASURE speakers by listening to them with music that I like/prefer to listen to.

Why would I need a microphone and/or other gear to accomplish the MEASUREMENTS that are most appropriate for the end use/function of the speakers?

I have used mikes and sound generators years/decades ago to set up systems, but this was for speakers that the end user already liked/preferred (the speakers/systems were, in the end, always finally tuned/equalized by ear).

DeKay


Toss the proverbial measurements into the proverbial cocked hat. Sure, the size, cost and aesthetics of a set of speakers play a definite role in the consumer’s decision to open up the proverbial pocketbook, but the purchaser’s decision essentially boils down to an aural beauty contest. Let the designers and manufacturers crunch the numbers and let the publicity department spin the bs. If the buyer has any sense left after being slapped silly by the buying process, the buyer ends up buying what tickles his/her ear.
" Measurements are needed to confirm what we think we hear."

Why?

kenjit OP
1,027 posts08-23-2020 9:43pm@millercarbon We need to use scientific methods to assess our speakers. Listening alone is inadequate.
Why do you need to use "scientific methods" to asses speakers ?
Why is listening not enough, after all. What do you do when using them..........................LISTEN...DUH.
That is what they are made for .. listening !
Only listening tells you how they sound .. not measurements
Post removed 
How do you gather that I was suggesting ("Are you suggesting there are none other than a frequency response chart?") such?
Because you are not telling me what other measurements there are. Why dont you tell me, if you know the answer?

Why would I need a microphone and/or other gear to accomplish the MEASUREMENTS that are most appropriate for the end use/function of the speakers?
Because our ears are not good enough to check the quality of the speakers. Also different people hear different things even if the speakers are the same. Measurements can be much more reliable consistent and they are better at identifying subtle differences in sound and also help us to understand the reasons for these differences. 

Only listening tells you how they sound .. not measurements

So if i measure the response of the speaker and it starts to fall off at 24db at 200hz, do you think that wont tell me the speaker will sound very thin and light?  If the response shows that it has a severe peak at 1khz, do you think that will not be audible? There is obviously some correlation between response and sound quality. Its just not close enough.

" Measurements are needed to confirm what we think we hear."

Why?
There is no consensus about what people hear and even if there is it could be completely wrong. Measurements tell us the facts. Facts are what we want not opinions. If you dont care about the facts, you can do what you like. No measurements needed.
I put on Hells Angel’s Cracker Factory and turn it up to 11.
There is no consensus about what people hear and even if there is it could be completely wrong. Measurements tell us the facts.

Well, no, as a matter of fact you are dead wrong. Measurements do not tell us "the" facts. Not at all. Measurements tell us only one very tiny subset of the facts. 

We can measure a rose down to its atoms, and beyond, to electron shells, protons, neutrons, quarks. So let me ask you kenjit, in all these measurements, where is the rose? Eh?

Oh and by the way, do you know what a consensus is? Its a form of measurement. Think about it. For a change.


Danny Richie explains and demonstrates how he uses measurements to design loudspeakers (and their crossovers) in his series of GR Research Tech Talk Tuesday videos, viewable on YouTube.

Stop asking the same question a dozen different ways; you have been told numerous times where the information you desire can be found, yet you continue to ignore it. Go to YouTube, do a search for GR Research, and do your homework assignment.

If you're too lazy to do that, please stop asking the same question over, and over, and over again. You're a dreadful bore.

Why should danny richie be the only person who can answer the question? Does GR research speakers represent every other speaker company? How do they know that their measurements are the only relevant ones? 

So let me ask you kenjit, in all these measurements, where is the rose? Eh?
Its all there in the measurements. But we are talking about sound waves here. No need to bring atoms and electrons into the discussion. We are interested in the movement of air. Thats all sound is. As long as we can accurately measure the correct data, thats all there is to it. 
you ask a question, we answer it and you keep saying that's not the answer you are looking for, therefore you already know your answer.
so state your answer and be done with it instead of trying to start arguments.
K, it's a tool. It's a TOOL. Ok, Quit being a TOOL.. USE the tools.

All crap aside, the forums you start have come from a bunch of joking around to some pretty good stuff. There is a WHOLE lot more to it than "JUST" measurements.

Danny at GR is a pretty strict, measurement man. BUT he has TRAINED his ears to hear a potential problem. Some live by the mic alone, some live by "how it sounds" alone.  The question is which is the best?
I say you can't have one without the other, BOTH in concert, gives me, a finished BOX, with drivers. Now it's time to, "Tune that  box, To that Room", That is where it gets real interesting..

KNOW your room!, Treat as mush as you can deal with (partner included) and go from there. If you know your room. There is NOT a lot of measuring going on. WHY? You know your room. You're not setting up rooms for a living.

Just one friggin' room. It's not that hard to do and it tells you what NO installer can know. A good setup man KNOWs certain things about room SIZE, you know the rest. Behind the scenes type stuff. (Adams family comes to mind). :-),

NOW where's the BEEF? Cotton out of the ears, and in the mouth, start using the TOOLS you have. TWO ears and a one mouth, use them proportionately. Listen, then share what you learn..

You Rang!!

Regards
Kenjit:

You are wasting your time on this topic. 

Would you buy a speaker just based on how well the specs read and not by listening to them. It is your ears that finally say this is right speaker. Wake up and smell the rose!
Wake up and smell the rose!

Heh. Good one. He won't get it. But good one.

Its all there in the measurements. But we are talking about sound waves here. No need to bring atoms and electrons into the discussion. We are interested in the movement of air. Thats all sound is. As long as we can accurately measure the correct data, thats all there is to it.

Okay. I’ll play along. Except we first need to get a couple things straight.

Number one, "sound waves". Waves are physical phenomenon, like on water or in air. In air waves are compression and rarefaction- changes in pressure. Sound is something completely different. Sound is a human quality- biological, psychological. Sound is the word we humans give to the perception of the physical phenomenon of waves in air. We do not "hear" dB. We perceive sound. These are completely separate and different things. Your constantly confusing and conflating them is causing you so much trouble you don’t even know. So try and keep them straight. Please.  

And two, more of a housekeeping thing but again since you’re determined to dig into this: We do not "accurately measure the correct data". Data is the result of measurement. Data itself is not what we measure. We measure the pressure waves. Which by the way are comprised of atoms, another one you got wrong. We measure dimensions, and we measure events. We record them on tables or graphs. The tables and graphs are the data. The things we measure and the resulting data are completely different things. Mixing these up is causing you more trouble than you know. So try and keep them straight. Please.  

Okay, so now we got that straight- Please tell us, so we can dispense with all this listening, what are the correct things we need to measure?
If you had ears you would know.
Forget about what's going on.
You will NEVER know that.
Trying to help him realize that for himself.
I think kenjit is a Russian troll perfecting his craft on a AI response program to turn loose on all of us. We're all in some weird Beta testing phase.

All the best,
Nonoise
Hmmm.  I thought he was left behind by aliens who didn't want him on the craft on the return flight back to their home planet.
Please tell us, so we can dispense with all this listening, what are the correct things we need to measure?
That is my question to this forum. 

We do not "hear" dB. We perceive sound. These are completely separate and different things.
Why do you think they are completely separate? Everything that can possibly be heard is just movement of air. 

We measure the pressure waves. Which by the way are comprised of atoms, another one you got wrong.
I dont agree that we need to know what each atom of the air is doing in order to understand the way soundwaves work.

We have already been designing and evaluating the performance of speakers by listening tests for a long time. What we need are more measurements not less. Once we have found the exact set of measurements that matter, we can then dispense with listening tests completely. We are not there yet. But my question is, how far have we progressed in this respect? Are there newer measurements being used now that didn't exist say 20 years ago? How has our progress in understanding the way we hear sound affected the measurements we use when it comes to evaluating sound quality?
Why do you think they are completely separate? Everything that can possibly be heard is just movement of air.

This is where we need to focus. Your confusion about what is what. Sound is human perception. Yes sound comes to us in waves. But sound and waves are still two completely different things. Instruments measure waves. Instruments do NOT measure sound! Only the human being listening with his ears can measure sound. You need to stop conflating and confusing these two very different things. 

Let me give you an example that may help you to see this distinction- the Fletcher Munson Equal Loudness Contours. https://en.wikipedia.org/wiki/Equal-loudness_contour  
Please note the first sentence:
An equal-loudness contour is a measure of sound pressure level, over the frequency spectrum, for which a listener perceives a constant loudness when presented with pure steady tones
In particular, notice measure of sound PRESSURE LEVEL for which a listener perceives a constant loudness. 

What the equal loudness curves are telling us is that while your measurements of air pressure waves are linear, our human hearing is not. This is something you should be well aware of, having lectured us endlessly about it. Well here it is, and if you want to begin to understand why listening rules over measuring you need to look hard at these curves and think and understand their meaning. 

What they really mean is you cannot possibly ever find any set of measurements that will tell you as much about the quality of a speaker as listening. Cannot be done. This is just one reason. There are many others. But this one comes with a chart you can look at. Study it. Please.
There is an interview of Roy Delgado from Klipsch by Steven Guttenberg in the Klipsch forums. Delgado describes measuring speakers and then listening to them after building them to “spec.” He comments that speakers specs are great for engineers but sometimes that does not translate into the sound that people like. So, build a speaker that people like how they sound, and you will sell speakers. Does that answer your question, no. It does give a different perspective on the idea of specs meaning something different for engineers or the technically astute. We have heard the same adage used with other audio equipment as well, specs aren’t good, but damn, it sounds good. You may feel deceived if you believe that a company is willfully is holding back specs for whatever reasons. I have seen 50k plus speakers advertised without detailed specs and I don’t think people are being deceived. 
@kenjit, if the experts and manufacturers don't know or have yet to be known ways of answering your question, how do you expect people on this forum to have the answer?
It's like asking us if we know the cure for cancer.
Not sure about measurements.

 I will sit and hear them playing my fav music.

 Then I will give the knock on wood test, if it sounds super hollow, I it’s not good.
or if the drivers inside the cabinet have their own enclosure. 

I measure speakers with a tape measure.  Height, width and depth.  Got any other brain busters?
b-limo, everyone knows you need a laser device to get accurate measurements !!!!!!
@oldhvymec 

You forgot the most important.

We should all listen to TOOL.  Then evaluate the quality of said speaker.  Tool at about 90db +/- 5db should do the job.
Important announcement:

Kenjit once owned, or still does own, Green Mountain speakers.

DeKay
I measure speakers with a tape measure. Height, width and depth. Got any other brain busters?

Don’t forget the scale. Most important speaker measurement there is. Too heavy you could throw your back out.

Also very important, possibly the most important thing you can measure, and here’s the tool to do it with:
https://www.amazon.com/New-GM-6-Gloss-Meter-Glossmeter/dp/B00S97P966/ref=asc_df_B00S97P966/?tag=hypr...

kenjit, if you insist on measurements instead of listening then of course you must insist on measurements instead of looking as well, right? Right??

As my Mama use to say: "You can lead a horse to water, but you can’t make him drink." The op is more a jackass than a horse, stubbornly more interested---it would appear---in arguing than in learning. The information he claims to thirst for is available, yet he insists on claiming it doesn’t exist. Weird guy.

Just one example: the use of the spectral decay ("waterfall") plot is ubiquitous in loudspeaker design. It is used to examine and determine the time domain behavior of individual drivers, enclosures, and complete loudspeakers. In numerous GR Research Tech Talk Tuesday YouTube videos, Danny Richie shows how he conducts his spectral decay measurements, explaining and demonstrating how he improves the time domain behavior in loudspeakers sent to him for evaluation, displaying the before and after response on his computer monitor.

Another measurement tool Richie explains and demonstrates is how the phase relationship between loudspeaker drivers affects the speaker’s frequency response. He explains and shows how he designs his crossovers to optimize the phase relationship between drivers, to create the best time and amplitude performance the drivers are capable of.

I have numerous times suggested the op watch those videos, yet he continues to insist that the waterfall plot test is not commonly use by designers. Mere intellectual laziness, or something more serious? Some people are just a lost cause. Like Mr. T, I pity the fool. ;-)


trudat
48 posts
08-24-2020 3:11pm
@oldhvymec 

You forgot the most important.

We should all listen to TOOL. Then evaluate the quality of said speaker. Tool at about 90db +/- 5db should do the job.

<<<<<<<<<<<<<<<<<<<<<<>>>>>>>>>>>>>>>>>>>>>>>>

Excuse me of being as thick as two short planks, but this went right over my head...No idea!

I'm pretty good at TUI translation too.
TYPING UNDER the INFLUENCE. You got me..  :-)

Regards
Here's a great test for any speaker: make it play the words person, woman, man, camera, TV.

Play half an hours worth of music and then ask it to repeat those five words. It it does, it proves nothing of note.

All the best,
Nonoise
 the use of the spectral decay ("waterfall") plot is ubiquitous in loudspeaker design
Funny how B&W, Wilson audio, Yg acoustics, Vivid Audio, Magico, Kef, and many more do not publish this data. How is a consumer supposed to compare speakers and decide which speaker has the best spectral decay measurements if it is being hidden?
OK....back to 'testing'.....without getting 'testy' about it....

Equipment of 'known values' driving speakers of 'purported values'.
Microphones of known values ic'd to test equipment of known values.
Perform a series of test routines known to deliver a certain level of values that have, in the past, delivered results of a certain degree of reliablity.
Do all of the above in a space of agreed upon values of qualities of response over a period of time in previous testing routines with the same parameters as outlined above...

There's the 'sci' approach.  With diligence and 'luck', you might match 'measured' with 'purported', +/- some degrees of 'acceptable variance'.

And the speakers in test still may sound to you like yesterdays' 'recycled oatmeal'....your choice of output....;)

Or, as noted previously and in multiple posts....

You can just Listen to the damn things and decide if you want to live with them.

The former routine is generally a bit beyond the scope of most to perform except by the most fastidious audioholic with time and $ to burn....

...imho....do what you must to scratch that itch...

J