Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[Duane Howell]

I simply can't imagine any manufacturer consenting to letting another manufacturer publicly and officially publish spec data for their product. It's a huge conflict of interest, it's like having the tobacco industry publish a health study on smoking. I wouldn't be surprised if there are legal issues too.

Your best bet would be some sort of third party database or organization, but I still think it would be rife with politics and conflicts of interest.

[Josh Ricci]

And besides this, what assurance would an independent customer have that the data was responsibly collected and the analysis was not slanted somehow to favor the manufacturer? Could you ever be certain? Even when some third-party standard is in place, such as CEA2010 distortion thresholds, independent practitioners have observed differences in their results from one data source to the next, which seem to be due to lurking variables in methodology and hardware. As far as I know these differences remain unresolved, and naturally each practitioner says "just compare measurements I made to other measurements I made", so CEA2010 distortion measurements are, for now, still not truly universal. If consultants and measurement specialists are unable to agree, when they are striving to add value to the standard by harmonizing their methodology, what will happen when every manufacturer is taking data separately? What if manufacturer A's measurements of third party C's subwoofer show meaningful differences from manufacturer B's measurements of third party C's subwoofer, and to make things more interesting, what if both A's and B's measurements differ from the published measurements by third party C of their own subwoofer? Who is right?

Something like what the OP describes was done by Community Loudspeakers back in its early days for off-axis measurements of speaker horns, and they did publish a book called the White Book containing their results, but I don't know if any copies of it are surviving and I've never seen one. But their White Book was the product of their efforts alone, and not a cooperation between manufacturers.

It can/should be done by an independent 3rd party in theory but the reality is there are a bunch of reasons it would be very difficult for it to work that way which I will outline below.  Verifiable measured data is very important. It should be completely unbiased in theory. Now if the data is being purposely skewed or completely fabricated to favor a product, that is a possibility, but it would be highly unethical.

As far as differences in methodology, etc...Yes this can account for some differences between data sets. You mentioned CEA-2010. There have been a number of instances over the years with data between multiple parties that didn't agree. The majority of these involved outlier data reported by people who were either very new at it or have a number of question marks in their data sets over the years that indicate that it simply isn't being captured or reported correctly and is of questionable trustworthiness. In those cases it usually gets investigated  and becomes apparent which is the "correct" set of data. This type of thing is very rare though. The MFG internal CEA-2010 data has been spot on in every case I've dealt with. By spot on I mean within +/-1dB generally over the operational bandwidth. You simply will never be able to get tighter agreement than this from something like CEA-2010. The hardware and even software differences don't amount to anything to worry about in my opinion. As long as the setup and methodology is good the results will be the same regardless.

The major cause for measurement variation is environmental and atmospheric conditions. This type of measurement collection is usually done outdoors. The second major factor is manufacturing tolerances in the speaker. Unless 2 testers get the exact same speaker this is going to cause some variation. For this reason I don't consider measured output differences as significant until >1dB. Burst output data like CEA-2010 is useful and it can be repeatable enough to be significant, but I find a lot of other measurements to be just as if not more useful and it's a large collection of different types of measurements that can really paint a picture of the overall performance of a speaker. All of those measurements will be subject to the same variance due to environmental and other factors as well but its the overall broad spectrum of performance differences between speakers that matter not overly exact minutia.

With all of that said what I would like to see in a perfect world is the industry move towards more standards similar to CEA-2010, which would then be adopted over time and used by the MFG's. I'd like to see standards covering: Sensitivity measurements, low frequency extension, frequency response, distortion, directivity and power handling. Everything would be based on measurements the measurements would be shown and there would be a suggested format that these should be reported in. Perhaps as a standardized PDF report since I'm dreaming here.

There are a number of problems with an independent 3rd party doing reports. (I speak from years of experience doing exactly this)
1.) Unless you've done a few of these you cannot fully comprehend the amount of work and time involved.
2.) There's so much product out there! (See above. You'd need a large team of people to truly cover even just the Pro market.)
3.) Its been fun for a while as a hobby/charity but is the whole team working for free indefinitely, year after year? (Doubtful)
4.) Generate income to grow and keep the thing going. How to do that? Crowd funding? Subscription? Advertising? Charge for the work/reviews? All of it? It's doubtful that crowd funding and /or subscription service would make enough revenue to offset the huge amount of time and effort. The pool of people that are interested enough in this type of thing is relatively small even if including Pro, consumer and mobile markets.
5.) If you take advertising or charge for a measurement workup from a speaker MFG it opens the door to the perception of being bought and bias towards them.
6.) If a product performs poorly or simply not quite as good as another the MFG may not supply future product for testing or may choose to not publicly release the data (NDA) or both.
7.) It is best not to rely on MFG's to provide product for the above reasons but purchase of products for testing out of pocket is also unsustainable once you get into products costing thousands of dollars each. Most owners of products balk at providing one for performance evaluation once they realize the lead time, shipping costs and potential for damage.

It makes most sense for the MFG's to do this stuff internally as part of product development. Most would assume that they are doing this but not sharing it publicly. You would be surprised how little some actually do.

Another problem is interpretation of the measurements regardless of who takes them. The average consumer/lay person will have no idea what they are looking at or how to interpret and make a useful comparison in most cases.

[John Roberts {JR}]

cough....

You can't handle the truth....

JR

[Chris Grimshaw]

I think it's more like "you really don't want to know the truth".

To pick a product, take a look at an EV sub where there's both a passive and active option. Note the wild differences in frequency response - any difference is being EQ'd by the built-in amplifier, and I've seen differences as great as 10dB on their mid-line 18" model.
So there'll be 10dB of EQ boost happening, which is 10x the power at some frequencies compared to others. Either you lose that as headroom, or you go for dynamic EQ, where you've got a cabinet with a frequency response that changes (drastically) with level. Neither of those are great options to work with, which is why I build my own speakers.

Chris

[Len Zenith Jr]

I think it's more like "you really don't want to know the truth".

To pick a product, take a look at an EV sub where there's both a passive and active option. Note the wild differences in frequency response - any difference is being EQ'd by the built-in amplifier, and I've seen differences as great as 10dB on their mid-line 18" model.
So there'll be 10dB of EQ boost happening, which is 10x the power at some frequencies compared to others. Either you lose that as headroom, or you go for dynamic EQ, where you've got a cabinet with a frequency response that changes (drastically) with level. Neither of those are great options to work with, which is why I build my own speakers.

Chris

You've seen impedance plots of cabinets then. With built in factory programmed DSP, if you only do your boosts where impedance is at a maximum, lets say 38 ohms, then there is little power added to your voice coil. You wouldn't lose your headroom. I'm sure the designers know how to pick and choose their battles.

[Chris Grimshaw]

You've seen impedance plots of cabinets then. With built in factory programmed DSP, if you only do your boosts where impedance is at a maximum, lets say 38 ohms, then there is little power added to your voice coil. You wouldn't lose your headroom. I'm sure the designers know how to pick and choose their battles.

They usually boost around the port tuning frequency, where impedance usually gets below the nominal rating of the cabinet.
Even putting thermal considerations to one side, the amp will still run out of voltage swing.

Chris

[Marcus Baeumler]

 Yes, the never-ending great debate…


I my personal opinion this is based on a few assumptions that I do not agree with and for me are the reasons why this will never happen.


1. It is possible to adequately and completely describe the audio performance, or maybe even more important to many, the audio experience of a loudspeaker system with a simple set of measurements that are also easy to understand.
-> I don´t think so…
Just think of something really simple as nominal dispersion: a simple number will not tell you anything. You will at least need a “dispersion over frequency” graph to really understand what is going on and even that is only half of the story.
Max SPL? There are probably 20 different ways to measure this and all of these different ways represent one part of the description of what somebody would expect as an audible experience when comparing one loudspeaker with another.
Short side story: a very highly regarded magazine here in Germany uses three-dimensional graphs to describe the output performance of amplifiers… Watts/Impedance/Crest factor.
And there are many more.
 
2. You will find two manufacturers that agree on a simple set of measurements that adequately describe the audio performance and audio experience of their loudspeakers.
-> Probably not and the reasons for that start with what I described above.
 
3. Everybody actually understands how these measurements relate to their expectations of the audible performance of a loudspeaker system.
-> Hard to judge objectively but on a broader scale I am not so sure. Nominal dispersion and Damping Factor are just some examples.
 
4. These simple measurement results are actually all that important.
-> Yes to many people but by far not for all. There were many posts above that describe a lot of the benefits that can be much more important when buying a system. I personally know of system sales for literally hundreds of thousands of dollars without a datasheet even being looked at.


Marcus

 

[John Roberts {JR}]

I think it's more like "you really don't want to know the truth".

I am repeating myself (again and again and...). Of course they want some version of the truth but are unwilling to educate themselves to understand how to evaluate multiple specifications and put them in correct context for their application (more is not always better).

To pick a product, take a look at an EV sub where there's both a passive and active option. Note the wild differences in frequency response - any difference is being EQ'd by the built-in amplifier, and I've seen differences as great as 10dB on their mid-line 18" model.
So there'll be 10dB of EQ boost happening, which is 10x the power at some frequencies compared to others. Either you lose that as headroom, or you go for dynamic EQ, where you've got a cabinet with a frequency response that changes (drastically) with level. Neither of those are great options to work with, which is why I build my own speakers.

Chris

As I have also long said I like to have the professional speaker engineers make the design tradeoff decisions for me after thorough evaluation of ALL variables. I haven't made my own speakers since the 70s. Of course there is nothing wrong with rolling your own if it makes personal sense (it hasn't for me for several decades).
=====

I have wrestled with this for years, professionally from the selling side (I supervised the speaker product manager for a decent sized company). He lost a lot of sleep and hair over trying to give customers what they think they want. This is an old and well discussed topic around here.
 
And one more time for the cheap seats most (all) major loudspeaker manufacturers try to present accurate specifications to satisfy customer's desire for information, but simple answers for complicated questions don't exist, and customers will make their purchase decision (often buying something else) long before you can finish explaining the difficult stuff. 

JR

[ProSoundWeb Community]