• The move to the new server is done. There are some software and database maintenance updates in process. This has us passing the hat around to help out. We appreciate any donations. Seriously, even a dollar helps. The payment page may be found here - https://www.audiokarma.org/support.html

Accuracy in Loudspeakers---Define

I just knew this would be fun to watch.

So far, its Zilch 2-0. :D

Can an accurate speaker be built?

No.

But not because accuracy can't be defined - but because of the limits of technology.

Accuracy is simple to define, guys - in any arbitrary system it is the perfect reproduction of the input at the output.

The proverbial accurate amplifier is a straight line with gain - input translated to output with no changes (other than the gain). The perfectly accurate speaker transforms its electrical input to an acoustic output with no distortion - no changes.

The continuing advancements in audio technology shrink the gap between what our speakers can do, and what they should do, to be perfectly accurate reproducers. But, this is an asymptotic function - there is always going to be a difference, vanishingly small though it may become. I.E., we can only approach perfection, but never truly and absolutely reach it.

Discussing the desirability of a sufficiently accurate system, or the potential for universal acceptability of such a system, or at what point 'vanishingly small' is small enough to deliver an essentially 'perfect' reproduction - now, there's some of the really fun questions...

I nominate Zilch's first post for AK's "Best Post of the Year" award. :thmbsp:

Back to the barcolounger... :lurk:
 
poohsan said:
So it's not so much that an accurate speaker cannot be built as much as the issue is we have no viable way of measuring a speaker's accuracy.
Click to expand...
With a bit of gear, knowledge, and practice, you can do it yourself, actually, for major parameters.... :yes:
 
CUlater said:
Accuracy is simple to define, guys - in any arbitrary system it is the perfect reproduction of the input at the output.
Click to expand...

But how is this perfection judged? On the display of an oscilloscope or SPL meter? or in the ear of the listener?

Why do tone controls or graphic equalizers exist? could it be that the human ear varies from person to person?

One mans perfection or accuracy is another mans imperfection or inaccuracy, it's what makes life interesting.
 
Tom Brennan said:
Quads are somewhat damped dipoles, there are those who can argue reasonably that dipolar dispersion is inaccurate. The Quads suffer severe dynamic limitations, again an inaccuracy.
Click to expand...

Dipolar dispersion, help me here, are you referring to the area outside of the 'sweet spot'? And if so, would you consider the area *inside* the sweet spot to be 'accurate'?

Wrt dynamic limitations, definition? Difference between 'loud' and 'quiet' notes (I'd define that as dynamic range). Inability to play at high volumes? If it's a volume thing, then would you consider them 'accurate' within the volumes they're intended for?

Tom Brennan said:
I'm not disputing your preference, indeed I hold the Quad in high regard, just pointing out the impossibility of the Quad, or any speaker, being accurate.
Click to expand...

I guess I'm kind of struggling with the term 'acurate'. The first time I sat in the sweet spot of a pair of Quads, and listened to Julie Andrews (JULIE ANDREWS!! I'm a kick ass rock and roll guy, for crying out loud!!), I felt like she was standing there, between the speakers. It was about as perfect sound reproduction as I've ever heard, and would have satisfied my definition of 'accurate'.

So my question is, would the Quads (or any other speaker) ever be considered 'accurate' within certain limitations (which could vary by speaker)?

Interesting conversation..

bs
 
Zilch said:
With a bit of gear, knowledge, and practice, you can do it yourself, actually, for major parameters.... :yes:
Click to expand...

Indeed you can measure what devices do and certainly a designer needs to know what his devices do but how do you decide if a device is accurate or going to aid a system in accuracy?

Let's say you have two horns, you objectively determine that one has a 90 degree pattern and the other a 120 degree pattern. Now what? Which one is part of an accurate system? Either? Neither? Who knows?

The objective data must be put to subjective use, I see no way around that.
 
I have to say that today, a truly accurate speaker cannot be made but we can supplement for that in the way we record and reproduce the sound. If you have never heard a Holophonic recording I highly suggest you listen to the samples on this site and the sample here as well. They make my laptop speakers image and that is an incredible trick. Also, listen to the violin in this file.
 
Skinley said:
Why do tone controls or graphic equalizers exist? could it be that the human ear varies from person to person?
Click to expand...

That has to do with personal preference, not accuracy. Some people don't like a speaker that's accurate. Some like a boom-tizz sound that is obviously not accurate but it's what sounds good to them. Some like rolled off treble because it's mellow and easy to listen to, but it's not accurate.

Ray
 
To me, it's like the term "High Fidelity". How high is high? Yes, some stereos are more faithful to the source than others, but I'd really like to see (hear) a stereo system that anyone couldn't tell the difference between it and a live performance. Even the best that technology can offer, is to me, miles away. That's not to say there aren't moments when some instrument or singer is reproduced in a fairly convincing manner, but High-Fidelity? What is the bias point?
 
Accurate?

If you found a speaker that measures extremely well, would it absolutely sound like the best speaker you ever heard? I doubt it.
You also have to take into consideration the system config. Are we talking about comparing them with identical electronics? We all know that speaker "A" could very well sound better than speaker "B", if it was paired with "Model XXX" amp, or vice versa.
I believe that what is the end-all speaker for one person, will not necessarily trip the next guy's trigger. So...., one could argue that the most accurate is the one that "does it for you", or the one that measures closest to the ideal set of predefined/universal technical parameters.
I've had it happen to me more than once. I've listened to what should be the best speaker I've ever heard, and they didn't do a thing for me. I've also put together a speaker that is the most "accurate" for my needs, wants and taste.
That is my definition of the most accurate speaker. If you don't like what you hear, then it doesn't matter how good they do, or do not measure..
It is a circular debate because of that very thing. I believe it boils down to subjective taste.
I'll say that some people just cannot grasp that concept. They dismiss it as something fundamentally wrong with the person.
If you're selling speakers, you want to advertise them as having the best measurements possible. If you're buying speakers, then you will listen and make a choice based on your own personal taste.
While not mutually exclusive of one-another, if you have one, it does not guarantee the other.
 
Skinley said:
But how is this perfection judged? On the display of an oscilloscope or SPL meter? or in the ear of the listener?
Click to expand...
Ear of listener is notoriously unreliable, and has no standard for comparison. That becomes a statistical exercise of considerable complexity:

http://www.harman.com/wp/pdf/HarmanWhitePaperMLLListeningLab.pdf

Transfer function uses source as one input and microphone as the other. You dial in the delay and push the "Display Differential" button.

Figuring out what it means is the tough part.... :p:

Skinley said:
Why do tone controls or graphic equalizers exist? could it be that the human ear varies from person to person?
Click to expand...
Not so much as we might suppose, actually. They are there to compensate for system inacurracies in situ and/or to adulterate the sound according to personal preferences....
 
Zilch said:
Well, this shook out to subjectivist/objectivist mighty quickly.

Accuracy is inherently objective, and bears no relationship whatsoever to how anything sounds to anybody, which does not appear in the formula. The assertion that accuracy resides in the realm of personal taste or preference is, well, preposterous.

This is only a dilemma for subjectivists who just can't get a grip on the fact that anything can be more absolute than their own opinion, in which conceit lies their only claim to credibility:

http://www.zaphaudio.com/evaluation.html

Move along now, folks -- there's nothing more to discuss here.... :p:
Click to expand...

The only problem with this, and a lot of arguments in this space, is that the author assumes that 'accurate' equals 'better'.

I think that there's a difference between what's accurate, and what's best for you, to your ears, in your listening room.

If I'm listening to kick ass rock and roll, at ear splitting volumes, then my K-Horns are 'best'. For other things, at other times, something else wins.

The problem is that people confuse the two, or are afraid to admit that what they love, or what they invested their precious dollars in, may not be the most accurate things out there.

bs
 
bshorey said:
The only problem with this, and a lot of arguments in this space, is that the author assumes that 'accurate' equals 'better'.

I think that there's a difference between what's accurate, and what's best for you, to your ears, in your listening room.
Click to expand...
Hmmm. :scratch2:

I don't believe I've either stated or implied so in this thread, but that's been studied, too, and I generally subscribe to these findings:

http://www.harman.com/wp/pdf/AudioScience.pdf

Different subject, of course, but I do consider loudspeaker accuracy a worthy pursuit.... :yes:
 
What a great thread. Defining it is simple doing it is not. Basically the input and output signal should match exactly in phase, amplitude and frequency content. Like that's going to happen any time soon:no: We are not really there with measurements either. We still don't have a measurement set that correlates to what we hear and is accepted by the industry. THX certification of theaters and home HT speakers is about as close as we have come.

Normally what we consider accurate is looking at an On Axis Frequency Response curve for flatness but that only tells part of the story. Without a complete set of Polar Measurements in both the Horizontal and Vertical axis you have no idea what the system power response is. This makes single point on axis measurements are very limited in there usefulness.

Also realize that measurements are not that easy to interpret in a meaningful way. You need to know the conditions the measurements were taken under and what if any smoothing was applied. They can be taken and manipulated to make a speaker look great or show the truth holes and all.

You still have to look at power compression, directivity and phase accuracy beyond your basic F/R curve.

Bottom line is we don't have the technology yet and there is no real standardization in the published curves that many manufacturers have in their brochures. So you can't always depend on the measurements published to even use the basic FR curve because you don't always know the conditions. Some of the most complete set's of measurements are the ones published by Revel. Even looking at those an audition is the only way to know for sure if the speakers will work for you or not.

Rob:beer:
 
Is speaker building considered an art?
Click to expand...

Depends on what side of the fence your are on:D I see it as a combination of art and science. Science to get you the basic building blocks and show you what's happening. The art being making the correct choices when you have to make trade offs in performance. Making a good sounding systems takes experience.

Rob:yes:
 
markb said:
Is speaker building considered an art?
Click to expand...

It shouldn't be at least not in the traditional sense. It seems that many confuse artistry and emotion involved in an original performance with an understanding of electro-mechanical systems and intelligently chosen compromises that go into making a reproduction. Not to deny that good engineering doesn't involve some artistry but there's a whole lot less of that body-wrenching stuff going on.
 
You must log in or register to reply here.
Back
Top Bottom