Audio Asylum Thread Printer

Get a view of an entire thread on one page

Critic's Corner

Will magazines ever improve their measurements in the near future?

96.255.226.137

There are a couple interesting articles in Stereophile about the limitations of the methods they use:

http://www.stereophile.com/features/105kh/index.html
http://www.stereophile.com/features/405time/index.html

But they haven't changed anything. I've noticed in recent times, some speakers measure so well on the standard Stereophile tests that it's hard to find fault with them or tell them apart. It seems that something more is needed. Measures of dynamics and distortion probably wouldn't help. It would also be interesting to see if they could extend the waterfall plots down to lower frequencies and see what that reveals.

Soundstage magazine has access to an anechoic chamber. They post some distortion measurements which Stereophile doesn't, but they don't post any waterfall plots at all which is a shame if the anechoic chamber would let them produce better ones than Stereophile.

Is there any hope either of these mags might expand their test regimes?

Follow Ups:

I think the (roughly) 8.5" x 10" measurement works fine. Sorry. n/t - hesson11 15:12:24 11/04/09 (0)
n/t
Keith Howard's article - KlausR. 04:48:58 11/04/09 (0)
As I see it Howard's main argument is:

"cabinet resonances are known to be a major factor, perhaps THE major factor, in determining a loudspeaker's sound quality..."

and

"as these resonances are known to be a prime determinant of a speaker's sound quality..."

I’d like to see the evidence for these claims. Interestingly, Howard does not cite any of the relevant literature, in particular these two:

Bastyr et al., “On the acoustic radiation from a loudspeaker’s cabinet”, Journal of the Audio Engineering Society 2003, p.234

Behler et al., “Investigation of cabinet vibrations of bass loudspeakers“, 31st Annual convention of acoustics (German Society of Acoustics), Munich 2005

Bastyr:

“Note the differences between the total sound pressure level and the SPL from the drivers alone at the frequencies of 120 and 200 Hz. At 120 Hz the total pressure is higher than the pressure from the drivers alone, whereas at 200 Hz the total pressure is lower than that from the drivers alone.”

“Due to its high radiation efficiency, resulting from its comparatively large surface, the cabinet can affect the total radiated pressure despite it low level of input power.”

“This means that while the wall vibration amplitude of the braced production cabinet is smaller, the location of the bracing is such that the production cabinet is a more efficient radiator of sound than the unbraced cabinet. More precisely, the radiation efficiency of the production cabinet is nearly double that of the unbraced cabinet at this frequency [300 Hz].”

Behler:

“The bending wave velocity, which is low compared to the speed of sound in air, does not allow for an efficient radiation [of the cabinet surfaces] because the sound pressures of the partial surfaces, which vibrate in opposite phase, are cancelling each other. This is the case for the smallest wall thickness [8 mm]. For greater wall thicknesses the radiation is weaker anyway.”

“It can be seen from the graph that the strong dips are caused by the internal cabinet modes and not by bending waves of the cabinet.”

“Cabinet vibrations have been shown by means of Laser Scanning Doppler Vibrometer and have, at low frequencies and small wall thickness, a demonstrable but small influence on sound radiation.”

From the above it appears as if cabinet vibrations are not a source of problems as a matter of principle, but that it strongly depends on the design of the cabinet.

Klaus
Resources for understanding the current ones would be helpful. - bjh 08:09:16 11/02/09 (2)
Organizations could perhaps make web resources available (perhaps collecting existing ones into a easy accessible location) that provide intreprations of what existing measurements proport to provide.

Everything matters, don't forget to tweak your placebos!
- RE: Resources for understanding the current ones would be helpful. - John Atkinson 10:40:39 11/02/09 (1)
  > Organizations could perhaps make web resources available (perhaps
  > collecting existing ones into a easy accessible location) that provide
  > intreprations of what existing measurements proport to provide.
  
  You can find my thoughts on how to measure loudspeakers and how to
  interpret the graphs at
  
  http://www.stereophile.com/features/99/
  http://www.stereophile.com/features/100/ and
  http://www.stereophile.com/features/103/ .
  
  John Atkinson
  Editor, Stereophile
  - Thanks. nt - bjh 11:55:53 11/02/09 (0)
    .
    
    Everything matters, don't forget to tweak your placebos!
RE: Will magazines ever improve their measurements in the near future? - Pjay 07:07:55 11/01/09 (6)
Measurements are a tough call. I like them. But a comparative analysis between measurements and listener comments shows there is some level of disconnect. Some moderately measuring items get good reviews.
Not being an electronics wizard, I still sometimes have to sit and think "does this mean this is good or it is bad?" For measurements to get better in a magazine format, they have to get better at relaying compartive info for the non-tech. Keyword is comparative. Then the question becomes: how far do we lower the bar?
I am not sure where this would go on the eletronics side, but on the speaker side, there are similarly measuring speakers that sound very different. One of my first projects was a kit that offered "ruler flat MLS". I measured the result and that is what it had. Amazing! It also had perfectly normal imp curve. Yet the speakers never engaged me and I found myself prefering a number of other speakers I had laying around the house. Now that I am almost a decade behind these, I think the issue was lack of dynamics or too much compression. Yet in measurement blocks like the ones in many magazines, these would have looked great.
As technology improves, a lot of speakers are measuring better. How do we improve the compartive analysis between the Alpha B1 and the Magico small speaker and the Usher small speaker? On paper these may not measure very differently but in a room, they would be very different. This is why we need the balance between the objective and subjective.
Maybe what we are missing the the comparative subjective chart, which I happen to have been playing with: above. More Blue is better.
P

My speaker building site

Edits: 11/01/09
- RE: Will magazines ever improve their measurements in the near future? - hahax@verizon.net 20:06:20 11/01/09 (0)
  All you have made apparent when describing the kit measurements is that there needs to be additional measurements. the ones you mention are all static and overlook dynamics.
- Amen... Nt - Frihed89 08:58:17 11/01/09 (0)
  Nt
  "Live free or die"
- I wonder what measurement indicates "big" sounds - bjh 07:23:52 11/01/09 (3)
  I've noticed a lot of people like "big" sound. I have a friend who is selling Wilson Watt/Puppy 7 to go with Jamo R 909 because the latter has ... yeah you guessed it.
  I'm practically in mourning as I adore the Wilsons and thus far have major misgivings with the Jamo's; I heard then over an evening but with APL digital player (too aggresive for my taste) and NuForce amplifiers (hate em').
  
  Everything matters, don't forget to tweak your placebos!
  
  Edits: 11/01/09 11/01/09
  - Define "big" to your ears. - Frihed89 08:59:45 11/01/09 (2)
    I can't help but agree with your general conclusions; I am just interested in what you hear as "big".
    "Live free or die"
    - RE: Define "big" to your ears. - bjh 13:38:15 11/01/09 (1)
      Well a sort of all enveloping effect, sound all around you sort of thing, especially the bass which is sort of everywhere. It gives the sensation of larger scale but at the expense of that sort of panaromic soundstage view with image specifity (a feature of much recorded music)... for example large scale like Wilson MAX II but sloppier, the MAXX II seems to do scale but without loosing the look into soundstage effect; but even the MAXX II seems less acurate or controlled compared to WATT/Puppy 7 it seems to me.
      Most omnidirectional systems I've heard seem to cast the big sound effect I might add.
      I suppose I would say in general big sound emulates live PA/sound systems, good ones.
      
      Everything matters, don't forget to tweak your placebos!
      
      Edits: 11/01/09 11/01/09
      - That's a good way to put it, says my ears. - Frihed89 01:16:18 11/02/09 (0)
        
        In Reply to: RE: Define "big" to your ears. posted by bjh on November 01, 2009 at 13:38:15
        
        By the way, the BC Penny, which is omni-directional does not have the bass-emphasis you mention. If you ever run across a small danish 2-way omni-directional monitor, called the Namia (or something like that) made by Kibri, give it a listen. With a little power, it's lovely.
        "Live free or die"
"wouldn't help"="would help" - riker1384 17:17:05 10/31/09 (0)
That was a typo, I think distortion measurements would be useful.
RE: Will magazines ever improve their measurements in the near future? - das@soundstage.com 12:04:34 10/31/09 (5)
You make some interesting points.

First, one thing I do want to point out that's not often apparent is that there is much more information *within* our existing suite of measurements that, if you're extremely knowledgeable about measurements, will be able to take from them. Take the frequency-response measurement, for example. People often use it simply to assess linearity but, in fact, if you know where and how to look, you can see resonances without having to resort to a waterfall, you can determine diffraction from resonance and so on. Hand a very skilled measurements-savvy designer a sheet with our current measurements, he'll be able to tell you more than the average audiophile or reviewer.

Other than that, there is always room for improvement and expansion. Our main concern: cost. As you point out, we use an anechoic chamber and the cost for this is not insignificant. In fact, I don't know of any magazine in the world who has access to equipment of this caliber -- partly because of logistics but often because of costs. It costs us a lot of money to do it there -- and we pay by the hour. Increase the measurements so that we double the time, we double the cost, basically.

For us, having this chamber is crucial. We could actually produce more measurements in an inferior environment, but we feel it's a matter of quality over quantity.

DS @ SoundStageNetwork.com
- Exactly - robert.gaboury@videotron.ca 06:05:39 11/03/09 (4)
  I do not think more measurements would be of any use to the average audiophile. Most do not have the skills to interpret correctly the meaning of a graph. It is also noteworthy that nobody actually found a way to characterize the "musical" or "dynamic" quality of a loudspeaker by measurements alone.
  
  This is particularly evident when Stereophile reviews a speaker where the listener likes what he hears while JA gathers and interpret catastrophic engineering flaws of the same speaker.
  
  Floyd Toole (at least when with the NRC) worked on this aspect for quite a long time, and frustrating as it may be (from an R&D perspective), ultimate sound quality still cannot be measured accurately in a repeatable manner.
  
  I remember reading that Toole (I'm not sure if it was with NRC or Harman) ran listening tests of an ugly speaker VS a beautiful one, but it clearly demonstrated that people do listen with their eyes, as the ugly speaker did not sound as good as the other one.
  
  On measurements, a simple impedance sweep can tells a lot if you know how to read the curve, but it's hardly entertaining from a reader's perspective.
  
  I guess it all brews down to the intent of the designer, and what the goal was (or budget, which has a tremendous impact on design), and to the manufacturer's perception of value relative to curves and measurements.
  
  Another contradiction of this hobby is that some very nicely engineered products (i-e any PSB speaker) do not scale high of the audiophile excitement scale.
  - RE: Exactly - Old Listener 08:49:23 11/03/09 (3)
    > I do not think more measurements would be of any use to the average
    > audiophile. Most do not have the skills to interpret correctly the
    > meaning of a graph. It is also noteworthy that nobody actually found
    > a way to characterize the "musical" or "dynamic" quality of a
    > loudspeaker by measurements alone.
    
    It is unfortunate that the "I know nothing but I hear everything" set has so overrun the audiophile world. Their level of hostility to knowledge, experiment and measurement seems to be increasing.
    
    > Floyd Toole (at least when with the NRC) worked on this aspect for quite
    > a long time, and frustrating as it may be (from an R&D perspective),
    > ultimate sound quality still cannot be measured accurately in a
    > repeatable manner.
    
    I think that you misrepresent Toole's work. The reports of his work that I have seen shown that listeners generally found that speakers that measure well were preferred to speakers that had serious flaws in measured frequency response or exhibited resonances.
    
    > On measurements, a simple impedance sweep can tells a lot if you know
    > how to read the curve, but it's hardly entertaining from a reader's
    > perspective.
    
    If I'm going to spend a few thousand dollars on a pair of speakers, I appreciate very much getting solid information to narrow the field to a manageable number of products and show me what I can get at a particular price level.
    
    > Another contradiction of this hobby is that some very nicely engineered
    > products (i-e any PSB speaker) do not scale high of the audiophile
    > excitement scale.
    
    PSB is still in business selling to rational consumers. Some of those consumers are reading Stereophile to get some factual information to make buying decisions.
    
    Bill
    - RE: Exactly - robert.gaboury@videotron.ca 09:38:35 11/03/09 (2)
      If you're to spend a few thousands dollars on a speaker pair - or car, I hope you do a test drive on top of factual information gathering.
      
      As Toole effectively stated:
      
      The final arbiter of quality, nevertheless, is always the subjective evaluation.
      http://www.theaudiocritic.com/back_issues/The_Audio_Critic_2 (Open in New Window)
      - I listen if I can manage it - Old Listener 11:47:05 11/03/09 (1)
        
        In Reply to: RE: Exactly posted by robert.gaboury@videotron.ca on November 03, 2009 at 09:38:35
        
        > If you're to spend a few thousands dollars on a speaker pair - or car, I
        > hope you do a test drive on top of factual information gathering.
        I don't think my words indicated that I would not like to listen before purchasing. Here is what I said:
        "If I'm going to spend a few thousand dollars on a pair of speakers, I appreciate very much getting solid information to narrow the field to a manageable number of products and show me what I can get at a particular price level."
        The audio dealer network in my area is getting pretty thin. Some of those dealers are no longer competent to set up equipment so that it sounds decent. Maybe I can borrow a pair of speakers over the weekend but I can't see managing to have several pairs of speakers on loan on the same Sunday. And some cost effective products are only available via direct sales.
        All those factors make it even more important now to have a reliable way to narrow the field. I do want to listen to speakers in my listening room before I purchase. That requires some effort for each purchase candidate so I want to pick out a few good choices and try them out.
        Sometimes one or all of the best choices are only available via direct purchase. Maybe there is a 30 day trial with a money back guarantee but I don't want to use that any more than necessary. Maybe I can arrange to listen to a friends speakers before buying something via direct sales. However, sometimes I have to do my research and take a flyer.
        Som emonths ago, I bought a pair of Audioengine A5 powered speakers for my home office without hearing them in advance. I didn't want a separate amp and speakers in that situation. There were several plausible choices in my desired price range; none were available for a local audition. My choice worked out well for near field listening.
        Bill
        
        Edits: 11/03/09
        
        Valid point - robert.gaboury@videotron.ca 12:01:33 11/03/09 (0)
        
        In Reply to: RE: I listen if I can manage it posted by Old Listener on November 03, 2009 at 11:47:05
        
        In my area, dealers are getting desperate. The good old time will not come back, but a new business model is being created and those who adapt will eventually succeed, because survival is not enough.
RE: Will magazines ever improve their measurements in the near future? - John Atkinson 10:56:28 10/31/09 (0)
> There are a couple interesting articles in Stereophile about the limitations of the
> methods they use:
> http://www.stereophile.com/features/105kh/index.html
> http://www.stereophile.com/features/405time/index.html
>
> But they haven't changed anything.

I admit that there will always be room for improvement. The problem is that I don't have
access to a large enough flat space on a routine basis to include a ground-plane
measurement in my test regime. However, I feel the current combination of nearfield
and farfield acoustic measurements, and vibrational analysis of the speaker's cabinet
panels, does uncover midrange problems - the subject of Keith Howard's articles - in
a qualitative if not quantitative basis.

John Atkinson
Editor, Stereophile

Post a Followup:

FAQ

Post a Message!

Forgot Password?

Moniker (Username):
Password (Optional):
	Remember my Moniker & Password (What's this?) Eat Me
E-Mail (Optional):
Subject:

Message:	(Posts are subject to Content Rules)

Optional Link URL:
Optional Link Title:
Optional Image URL:
Upload Image:
E-mail Replies:	Automagically notify you when someone responds.

To view your new posting or follow-up, click on the RELOAD or REFRESH button on your browser.