How to Choose a Loudspeaker -- What the Science Shows - Page 8 - AVS Forum | Home Theater Discussions And Reviews
Forum Jump: 
 5649Likes
Reply
 
Thread Tools
post #211 of 4433 Old 01-01-2019, 09:12 AM
Member
 
Makav3li's Avatar
 
Join Date: May 2009
Location: NJ
Posts: 174
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 94 Post(s)
Liked: 56
You may want to consider moving the main posting of this information to another forum. There are well known forums that are much more receptive to scientific discussion than avs. I fear that you'll only be bashing your head against a brick wall here.
avkv likes this.
Makav3li is online now  
Sponsored Links
Advertisement
 
post #212 of 4433 Old 01-01-2019, 09:21 AM
AVS Forum Special Member
 
Molon_Labe's Avatar
 
Join Date: Jun 2015
Location: Texas
Posts: 2,190
Mentioned: 105 Post(s)
Tagged: 0 Thread(s)
Quoted: 3347 Post(s)
Liked: 4027
Quote:
Originally Posted by R Harkness View Post
Right, sorry, now I remember you saying that befiore. I understand wanting a speaker to measure well, particularly if you are in the accuracy-or-bust school of thought. I prefer that the speakers I buy measure well - and I always look for measurements if available. But in my case a speaker lives and dies on how it sounds to me above all.
Going with ears alone, one is not sure if they are hearing an anomaly in the speaker (which may be pleasing) or if the room is coloring the sound to one's liking or disliking. Starting with a speaker that measures well, one is reducing factors that could be adding or subtracting to the source. I trust my JBLs are accurate, and I can season to taste with EQ. For some reason, the audiophile community has made the treble/bass knobs and PEQ taboo. The trend is to buy a speaker with "the sound" that is nirvana to one's ears. However, those adjustment knobs had better be at 12 o'clock or your cheating the system. One must keep changing and auditioning speakers until they "find it". I would rather buy a speaker that is proven accurate with great off-axis response at the onset vs the speaker merry-go-round game. Once I know I have the best measuring speaker that fits my decor, space, and budget, I can then use treatments and EQ to achieve "that sound". I can also have multiple profiles for the source or genre I am listening to. The tech/science approach with active equalization and room correction technology like Dirac Live is amazing.
sdurani, Vergiliusm, avkv and 6 others like this.

Last edited by Molon_Labe; 01-01-2019 at 11:49 AM.
Molon_Labe is online now  
post #213 of 4433 Old 01-01-2019, 09:43 AM
AVS Forum Special Member
 
Join Date: Mar 2017
Posts: 1,344
Mentioned: 15 Post(s)
Tagged: 0 Thread(s)
Quoted: 914 Post(s)
Liked: 276
Quote:
Originally Posted by aarons915 View Post
I didn't realize this was an option but some of you California people should take advantage of this. I'd love to see some measurements of Sierra 2 or other internet direct speakers that are popular. LS50's would be nice too but I'd be surprised if they didn't already have them on file.
What town do we have to take the speaker to?

Sent from my SAMSUNG-SM-G930A using Tapatalk
SouthernCA is offline  
Sponsored Links
Advertisement
 
post #214 of 4433 Old 01-01-2019, 10:35 AM
AVS Forum Special Member
 
callas01's Avatar
 
Join Date: Feb 2010
Posts: 4,048
Mentioned: 19 Post(s)
Tagged: 1 Thread(s)
Quoted: 1390 Post(s)
Liked: 935
Quote:
Originally Posted by SouthernCA View Post
What town do we have to take the speaker to?

Sent from my SAMSUNG-SM-G930A using Tapatalk
I think Northridge
callas01 is offline  
post #215 of 4433 Old 01-01-2019, 11:02 AM
AVS Forum Special Member
 
Join Date: Dec 2016
Location: Las Vegas, NV
Posts: 1,430
Mentioned: 33 Post(s)
Tagged: 0 Thread(s)
Quoted: 933 Post(s)
Liked: 1119
We had best wait to hear from Kevin on the subject of having any of your personal speakers being tested at the Harman facility in Northridge, CA.

I would not get too excited about the prospect until he addresses the matter. I say again, this is a time consuming process and the how facility is scheduled is up to the folks who run it.

In the meantime, get REW and see what your speakers and room look like.

For you serious enthusiasts and folks who like to take the full academic approach to learning about this, here is the link to the full ANSI/CEA-2034-A document. It's 60 pages and is quite technical. It costs $103 and must be purchased. Or, you can get Dr. Toole's books for less and get a lot more information.

https://webstore.ansi.org/Standards/CEA/CEA20342015ANSI
avkv and bearr48 like this.

Last edited by Rex Anderson; 01-01-2019 at 11:12 AM.
Rex Anderson is online now  
post #216 of 4433 Old 01-01-2019, 11:17 AM
Advanced Member
 
gene c353's Avatar
 
Join Date: Mar 2009
Location: S.F Bay Area
Posts: 927
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 177 Post(s)
Liked: 133
Frequency response is only one way to evaluate a speaker, but an important one. The more we can learn and understand it the better. And remember, they work for Harman. Of course they are going to use Revel, JBL, etc in there attempt to educate us. Otherwise this would turn into a "this speaker is better then that speaker thread". We already have enough of those. IMO they should use just a couple of speakers because this thread is about understanding FR and the way it's measured/evaluated, not about quality/comparison of sound. If someone hears a little "fanboy" in their tone once in a while it's only because they are proud of the work they have done (aren't you proud of the work you have done?). But these guys are engineers not marketing/sales/finance. There has always been a push/pull relationship between the two. Usually more pushing.

I always look forward to posts from industry insiders, including engineers like those from Harman, and reviewers like Kal Rubinson (got it right this time ). I hope they continue to contribute considering the way they are treated sometimes.

I've learned a few things from this thread, including using a single speaker in mono for evaluation. I'll have to try that someday, real soon.

What I'd really be interested in is to compare the charts measured at Harmans facility of a consumer level speaker like the JBL Studio 2 series to the charts measured in the average persons living room, where that speaker is likely to wind up.
avkv, Madmax67 and Lonewolf7002 like this.
gene c353 is offline  
post #217 of 4433 Old 01-01-2019, 11:19 AM
AVS Forum Special Member
 
Dave in Green's Avatar
 
Join Date: Jan 2014
Location: USA
Posts: 8,112
Mentioned: 147 Post(s)
Tagged: 0 Thread(s)
Quoted: 3678 Post(s)
Liked: 2883
As might be expected it was Canadian speaker companies that first started taking advantage of the scientific audio research at Canada's National Research Council facilities. I recall shopping for speakers back in the '80s and seeing Canadian speakers seemingly coming out of nowhere to get great reviews. Looking at the websites of some of Canada's most respected speaker companies shows they credit that NRC research for helping achieve their superior audio performance:

psbspeakers.com/articles/Paul-Barton-at-the-NRC

paradigm.com/en/research-development/nrc-research
Dave in Green is online now  
post #218 of 4433 Old 01-01-2019, 11:25 AM
Member
 
mitchco's Avatar
 
Join Date: Aug 2016
Location: Sunshine Coast Canada
Posts: 95
Mentioned: 9 Post(s)
Tagged: 0 Thread(s)
Quoted: 73 Post(s)
Liked: 137
Thanks Kevin @avkv I really appreciate your work, along with Floyd Toole and Sean Olive. For someone that lives in Canada and off the grid, it is next to impossible to listen to (any) speakers easily without travelling.

As a former audio pro turned amateur, I rely on measurements and reviews to help guide my purchases. I have been fascinated by the spinorama methodology and gathered quite a bit of information about it, along with Sean’s work on people’s subjective listening preferences on what makes for a good sounding loudspeaker (and headphone for that matter). I wrote an article that has links to quite a few of these presentations that point to AES papers on the research and measurements. It is quite a body of work, spanning many years: https://audiophilestyle.com/ca/revie...-r720/#science

In my most recent article, I compare the KEF LS50, which uses the spinorama methodology, to a pair of much larger, higher directivity loudspeakers. I also compared the KEF LS50’s with and without subs. Not only do the in-room measurements line up with the predicted in-room target, the LS50’s sound neutral to my ears.

I also made binaural recordings of each of the comparisons, including being able to switch back and forth between the LS50 versus LS50 with subs and LS50 with subs versus JBL 4722 with subs. One can hear clearly the difference directivity makes in choosing a loudspeaker, even if the speakers are eq’d similar. One can also hear the difference a sub makes, if ones headphones go down low enough.

https://audiophilestyle.com/ca/revie...cordings-r768/

So thank you for your science. Based on my measurements and ears, works for me. Please keep up the good work. I just wish other speaker designers and manufactures would use more science to predict in-room neutral responses. Neutral sounding speakers would be a good place to start. Most speakers I have listened to and reviewed, are too bright sounding to my ears.

Happy New Year!

Mitch
sdurani, DS-21, parapet and 6 others like this.
mitchco is online now  
post #219 of 4433 Old 01-01-2019, 11:30 AM
AVS Forum Special Member
 
michael hurd's Avatar
 
Join Date: Nov 2007
Posts: 3,413
Mentioned: 22 Post(s)
Tagged: 0 Thread(s)
Quoted: 635 Post(s)
Liked: 677
[QUOTE=garygarrison;57351584]
Paul Klipsch believed that having excursion of 1/16 of an inch or less in a woofer was preferable.
/QUOTE]




Kellogg, in 1931 proposed 0.1 cm ( 1 millimeter ) of cone excursion as a maximum. Beers and Belar examined this in 1942, published in April 1943, more than a decade later than Kellogg.



James Moir also presented his findings on cone excursion and doppler distortion September of 1973 in an AES pre-print.
fill35U likes this.

Ask your doctor if DIY is right for you. Side effects of DIY may include anxiety, elevated blood pressure, lightheadedness, rapid heartbeat, skeletal muscle flaccidity, euphoria, psychological dependence, insomnia, confusion, blurred vision, implusivity, uncontrolled or repeated movements.
michael hurd is offline  
post #220 of 4433 Old 01-01-2019, 11:45 AM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 763
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 794 Post(s)
Liked: 2864
Quote:
Originally Posted by Gooddoc View Post
Reading this thread, it doesn't seem clear that many find it valuable. If measurements were life-and-death, I suspect we would have a very high mortality rate.
A slightly "sick" version of that true statement, is that selecting loudspeakers without either double-blind listening data (almost impossible) or adequate measurements (truly rare) is a bit like Russian roulette - it is a high risk venture.

I attach the last figure in my book, Figure 18.8, which seems to summarize where we are. The "not good" loudspeaker was a creation of the US distributor of a very well known, respected brand that had competent loudspeaker engineering in Japan. It was ignored, and a design was apparently "phoned in" by a local contractor. At the time of publication, forum commentary on this speaker was leaning towards bi-amping as an improvement. Wishful thinking . . .

Those who doubt the utility and meaning of measurements will find the answers, and much more, in the preceding 467 pages.

For the cost of the bill of materials in this loudspeaker, competent design could have created more than acceptable sound, and competent measurements would have revealed it.

This, sadly, still goes on, and some of it at high prices, perpetuated by . . . well . . . you know who.

Happy New Year to all!
Attached Thumbnails
Click image for larger version

Name:	Figure 18.8 conclusion.jpg
Views:	243
Size:	124.1 KB
ID:	2503856  
Floyd Toole is offline  
post #221 of 4433 Old 01-01-2019, 11:46 AM
AVS Forum Addicted Member
 
R Harkness's Avatar
 
Join Date: May 2001
Location: Toronto, Ontario, Canada
Posts: 15,129
Mentioned: 17 Post(s)
Tagged: 0 Thread(s)
Quoted: 2756 Post(s)
Liked: 2675
Quote:
Originally Posted by Rex Anderson View Post
It is certainly a lot harder to get the double blind thing set up and done well, but John pulled it off at his home and did the shootout between Revel Salon2 and JBL M2 which turned into an epic thread here on AVS.
And look at what he had to go through and the resources he had at hand. It’s not for nothing that HK has to go to the lengths of building a speaker switcher device, special room, listener training etc. I think we need to admit the obvious that double blind testing speakers just isn’t practical and likely to happen for the average consumer/audiophile. And I am actually one of the few audiophiles I know who continually defends blind testing in the audiophile forums. And I have performed a number of eye-opening blind tests myself on some cables, DACs, CD players, music servers. Very few are motivated to do this and even I draw the line at attempting double-blind testing for speakers due to the impracticality.
Even if it was possible to set up one between a very few speakers it would leave out so many other speaker options I’d be interested in. As Prof Toole noted in the Revel shoot out thread:

Most people don't have the facilities, the money, the time, or the attention span, to undertake such meticulous tests.”

That really leaves the vast majority of us, if we are to take the HK advice, looking to one Speaker Company’s measurements and evaluations of their competitors.

Last edited by R Harkness; 01-01-2019 at 11:50 AM.
R Harkness is offline  
post #222 of 4433 Old 01-01-2019, 12:12 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 763
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 794 Post(s)
Liked: 2864
Quote:
Originally Posted by R Harkness View Post
As Prof Toole noted in the Revel shoot out thread:

Most people don't have the facilities, the money, the time, or the attention span, to undertake such meticulous tests.”

That really leaves the vast majority of us, if we are to take the HK advice, looking to one Speaker Company’s measurements and evaluations of their competitors.
First of all, I am not a "prof", but thanks for thinking so. I do have a PhD and lots of publications and industry awards (from organizations populated by Harman competitors), though.

Second: HK (Harman/Kardon) is a tiny company under the Harman International umbrella. It has a historically important place, though, as the original company created by Sidney Harman.

Third: a full spinorama is nice, but not at all necessary to recognize good sounding loudspeakers. The original data set that I created at the NRCC in 1983 utilized on-axis, plus averaged collections of progressively off-axis sounds, all published in my early papers:
Toole, F. E. (1985). “Subjective measurements of loudspeaker sound quality and listener preferences”, J. Audio Eng. Soc., 33. pp. 2-31.
Toole, F. E. (1986). “Loudspeaker measurements and their relationship to listener preferences”, J. Audio Eng. Soc., 34, pt.1, pp. 227-235, pt. 2, pp. 323-348.

Chapter 5 in my book discusses the evolution of the spinorama.

An evolved setup to make those original measurements was left by me at the NRCC in 1991 when I joined Harman, and it is still used by PSB and others to design speakers, and by Soundstage.com in their reviews. Other manufacturers still use the original data format, and there is nothing wrong with that, so long as they hit the targets - sometimes they don't - marketing sometimes wins over engineering.

You can find variations on this collection of on and off axis data from a few manufacturers who believe in measurements and are not embarrassed to show them. Look for smooth, flattish on-axis/listening window curves and similarly smooth and gradually changing off-axis performance. It does not require an engineering degree to "read" the essential information in the curves. There are lots of them in my book, good and bad, to provide practice.
Floyd Toole is offline  
post #223 of 4433 Old 01-01-2019, 12:34 PM
Advanced Member
 
TimVG's Avatar
 
Join Date: Apr 2011
Posts: 548
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 266 Post(s)
Liked: 399
With some time, effort, and a little bit of money, it's very much possible to get a decent idea of how a loudspeaker performs in your own home, using REW and a decent USB microphone. While not professional by any means, a combination of steady-state (low frequencies) and optimised time gated measurements both in the nearfield and at the listening position, over a listening window arc, will paint a decent enough picture of what is in front of you.

FYI - my humble Revel M105s measure superbly well, closely resembling the data on the spinorama chart.

It's always a good idea start off with a proven, neutral, loudspeaker that has a well behaved direct and reverberant sound field. Recordings are the great unknown and the biggest variable of it all. Use the bass/treble knobs, or a good equalizer to bring the recording to taste if you must, but to do so, it's much easier to start off with good loudspeaker.

Looking forward on seeing more data from @avkv and @Floyd Toole

On a different note: in the 'blind test' pdf there's a couple of spinoramas shown, of the magico, the 228be and the Salon2. In these graphs the 228Be appears better than the Salon2, which seems odd to me. Could it be the wrong graph?

I always assumed this was the Salon2 graph (can't remember how I got it). It looks different than the one in the pdf, much wider dispersion to start - and more behaved.

krabapple and wpbpete like this.

Last edited by TimVG; 01-01-2019 at 12:58 PM.
TimVG is offline  
post #224 of 4433 Old 01-01-2019, 01:06 PM
AVS Forum Addicted Member
 
Kal Rubinson's Avatar
 
Join Date: Mar 2004
Location: NYC + Connecticut
Posts: 30,174
Mentioned: 21 Post(s)
Tagged: 0 Thread(s)
Quoted: 1268 Post(s)
Liked: 1163
Quote:
Originally Posted by Floyd Toole View Post
You can find variations on this collection of on and off axis data from a few manufacturers who believe in measurements and are not embarrassed to show them. Look for smooth, flattish on-axis/listening window curves and similarly smooth and gradually changing off-axis performance. It does not require an engineering degree to "read" the essential information in the curves. There are lots of them in my book, good and bad, to provide practice.
For example, see the data that Bryston provides for their speakers.

Kal Rubinson

"Music in the Round"
Senior Contributing Editor, Stereophile
http://www.stereophile.com/category/music-round

Kal Rubinson is online now  
post #225 of 4433 Old 01-01-2019, 01:32 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 763
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 794 Post(s)
Liked: 2864
Quote:
Originally Posted by TimVG View Post
With some time, effort, and a little bit of money, it's very much possible to get a decent idea of how a loudspeaker performs in your own home, using REW and a decent USB microphone. While not professional by any means, a combination of steady-state (low frequencies) and optimised time gated measurements both in the nearfield and at the listening position, over a listening window arc, will paint a decent enough picture of what is in front of you.

FYI - my humble Revel M105s measure superbly well, closely resembling the data on the spinorama chart.

It's always a good idea start off with a proven, neutral, loudspeaker that has a well behaved direct and reverberant sound field. Recordings are the great unknown and the biggest variable of it all. Use the bass/treble knobs, or a good equalizer to bring the recording to taste if you must, but to do so, it's much easier to start off with good loudspeaker.

Looking forward on seeing more data from @avkv and @Floyd Toole

On a different note: in the 'blind test' pdf there's a couple of spinoramas shown, of the magico, the 228be and the Salon2. In these graphs the 228Be appears better than the Salon2, which seems odd to me. Could it be the wrong graph?

I always assumed this was the Salon2 graph (can't remember how I got it). It looks different than the one in the pdf, much wider dispersion to start - and more behaved.

The graph you show is ancient, having been done using the long retired MLSSA system. The curves in my book, Figure 12.1 are from the updated Harman designed system, which was excellent, but keeping it going through frequent OS upgrades required a resident computer engineer. The latest ones use the Klippel system which is someone else's responsibility. Theoretically they should deliver identical curves . However, in addition to production changes - the test objects are different - there are significant problems in making measurements on tall floor standing loudspeakers using a turntable in a chamber. An anechoic chamber is truly anechoic down to a frequency dictated by the length of the wedges. At lower frequencies there will be standing waves, causing errors of the same kind one finds in listening rooms, albeit better damped. Calibrating the chamber requires outdoor measurements (10 m tower or ground plane) compared to the same measurement at the rotational point in the chamber. It is usually done with a closed box woofer. Speakers with ports will interact differently with the standing waves in the chamber causing errors on any specific axis, so in general one should trust the sound-power curve.

With tall towers, if you align the preferred measurement axis at 0 deg. at the point of rotation you will end up with the woofers hanging out in space. When the speaker is rotated the woofers are not in the chamber calibration location at the point of rotation. So there are two important variables: the choice of "reference axis" (tweeter axis or some other point) and the fact that the bass measurement will not be calibrated. The smaller the loudspeaker the more truly accurate will be a simple spinorama measurement at low frequencies. The "work around" is to do multiple measurements, one with the origin of the reference axis located at the point of turntable rotation (resulting in erroneous bass data), and another with the woofers sort of over the point of rotation (resulting in erroneous mid-high frequency data). These data need to be digitally combined to yield a single set of data that is meaningful.

I was not there when the measurements were made, so I will leave it to Kevin to comment.

The good news is that in a listening room - any listening room - bass performance is dominated by standing waves (room resonances) and adjacent boundary issues. I devote two entire chapters, 8 and 9, to dealing with these problems. Anyone who has explored bass performance in rooms will realize that woofers and subwoofers are merely crude energy sources. Multiple subs and/or equalization will absolutely be necessary to achieve good bass, no matter what the anechoic curve looks like. That said, anechoic data can indicate the lowest frequency of useful output, and in general, there is a correlation between low-frequency cutoff and listener preference. This alone is a factor in the Salon2 preferences - it thinks it is a subwoofer.

Those who notice that the Salon2 is not as smooth as some other models need to look at the data relating to the audibility of resonances in Section 4.6.2 in my book. Absolute smoothness is not a requirement, but it is important that any amplitude variations are at or below the threshold of detection. They are. I have Salon2s knowing this, and they sound fine. Because they are 4-way systems they can play indecently loud without distress and the 4 inch midrange driver provides the potential of improved directivity.

Discussions of the significance of timbral differences between speakers at this level of performance approach the old question: How many angels can dance on the head of a pin? Music recordings and movie sound tracks are anything but consistent.
Floyd Toole is offline  
post #226 of 4433 Old 01-01-2019, 01:36 PM
AVS Forum Addicted Member
 
R Harkness's Avatar
 
Join Date: May 2001
Location: Toronto, Ontario, Canada
Posts: 15,129
Mentioned: 17 Post(s)
Tagged: 0 Thread(s)
Quoted: 2756 Post(s)
Liked: 2675
Quote:
Originally Posted by Floyd Toole View Post

Second: HK (Harman/Kardon) is a tiny company under the Harman International umbrella. It has a historically important place, though, as the original company created by Sidney Harman.

Third: a full spinorama is nice, but not at all necessary to recognize good sounding loudspeakers.

Thanks. The original posts seemed to be quite Harman Kardon spinorama-centric.




Quote:
Originally Posted by Floyd Toole View Post

You can find variations on this collection of on and off axis data from a few manufacturers who believe in measurements and are not embarrassed to show them. Look for smooth, flattish on-axis/listening window curves and similarly smooth and gradually changing off-axis performance. It does not require an engineering degree to "read" the essential information in the curves. There are lots of them in my book, good and bad, to provide practice.

Understood (I hope ). Among the speakers I have owned, and still own, are those from the deceased Canadian speaker company Waveform. The Waveform speakers were carefully designed by Dr. Claude Fortier with input by Paul Barton using the facilities at the NRC, with among their guiding principles:


* No substitute for blind listening
* Good acoustics = verifiable science
* Flat measured response
* A flat listening window
* Controlled wide dispersion
* Smooth room response


Reviews, even the crankiest objectivists measurements-first writers like The Audio Critic magazine and The Sensible Sound rated the Waveform speakers as the most accomplished speaker design they had encountered up to that time both in terms of measured success for their goal, and sound quality (company folded in 2000 I believe). One review with some measurements here FWIW:


http://www.audio-ideas.com/reviews/l...mach_solo.html



I love the sound of the Waveform speakers. What I find interesting is that, as far as I can tell, the design goals were very similar to those espoused in this thread and used for the Revel speakers. Yet I very much prefer the presentation of the Waveform speakers. There's a slight difference in there that - to my perception - renders the sound of voices and acoustic instruments with more believable tonal warmth on the Waveforms. I wonder to what degree my preference for the Waveform speakers could have been predicted and on what exact measurements.



But I am very curious about your recommendation: " Look for smooth, flattish on-axis/listening window curves and similarly smooth and gradually changing off-axis performance." And it's implication for different speaker designs - e.g. Electrostatic or Omni-directional speakers.


Stats are known not to put off as much sound to the sides as a conventional box speaker, so they are going to measure quite differently (if I have this correct) in that respect. What are the implications for electrostatics in the "preference" scale, given how they may tend to measure on axis vs their steep drop-off to the sides? I'm thinking Quad ESL 57s, 63s et al.



Then there are the attempts at omni-directional speakers. Among speaker designers, I've often seen it stated that an ideal speaker would be: the point source model, the pulsating sphere, ‘identical loudness in all directions’.
So you have companies like MBL attempting this with their radialstrahler models:

e.g.


https://www.stereophile.com/content/...r-measurements


But if your research recommends a gradual even sloping DOWN of frequencies off axis, what does this suggest for the preference rating of omni-directional speakers that put out essentially even energy radiation in to the room?


(I also own a pair of MBL radialstrahler omni-directional monitors and absolutely adore their sound btw. I wonder if you or HK ever blind tested that brand?).


Thanks.

Last edited by R Harkness; 01-01-2019 at 01:45 PM.
R Harkness is offline  
post #227 of 4433 Old 01-01-2019, 02:01 PM
AVS Forum Addicted Member
 
gajCA's Avatar
 
Join Date: Mar 2009
Location: Sonoma County, CA
Posts: 18,868
Mentioned: 221 Post(s)
Tagged: 2 Thread(s)
Quoted: 8800 Post(s)
Liked: 5988
Quote:
Originally Posted by Floyd Toole View Post

An evolved setup to make those original measurements was left by me at the NRCC in 1991 when I joined Harman, and it is still used by PSB and others to design speakers, and by Soundstage.com in their reviews. Other manufacturers still use the original data format, and there is nothing wrong with that, so long as they hit the targets - sometimes they don't - marketing sometimes wins over engineering.
Everyone but me likely knows already about Soundstage.com's archive of measurements for what must be close to 100 speakers.

https://www.soundstagenetwork.com/in...=16&Itemid=140

And amazingly they tested my little known Usher "bottom of the line" V602 towers.

Geoff A. J., California
gajCA is online now  
post #228 of 4433 Old 01-01-2019, 02:25 PM
AVS Forum Special Member
 
Bill-99's Avatar
 
Join Date: Oct 2003
Location: Columbus, OH
Posts: 1,512
Mentioned: 30 Post(s)
Tagged: 0 Thread(s)
Quoted: 833 Post(s)
Liked: 517
Quote:
Originally Posted by gworrel View Post
Many of the Internet Direct companies will allow in-home trials. The trick is just to buy two pairs with liberal return policies and compare them head to head in your home.

Back in the early to mid 80s, I was shopping for my first serious speaker buy. I bought a pair of Snell Type C and a pair of big AR towers from two separate stores with the clear understanding with both stores that I would be comparing the two and returning the pair that lost the comparison. The Snell speakers won out with better clarity in the bass as I remember. When I returned the AR speakers, the salespeople were incredulous. They tried to belittle the Snell speakers by pointing out that the Snells relied on complicated electronics inside the speakers. So in their eyes, a complicated crossover was a sign of inferior drivers or something like that. After the Snells I went DIY and have not looked back. I especially realize now how the crossover isn't everything, but it is a big thing.

Of course, I didn't compare the speakers back then double blind or in mono, so I've learned that from this thread. The other thing I've learned from this thread is that I should finish reading Floyd Toole's book which I started but haven't finished. A fresh new year and a new goal already.
"Many of the Internet Direct companies will allow in-home trials. The trick is just to buy two pairs with liberal return policies and compare them head to head in your home."

A big problem with this approach is that it's quickly faced with the sheer number of speakers available, making meaningful in home testing impossible for most consumers.

Consider, at the top of the speaker section here on AVS is a thread for specific product threads, most of them list manufacturer names. On a quick pass, I found 84 different manufacturers. How many speakers are there per manufacturer? Spot checks of a few different companies gave a range of 7 to 29 -- you can get different numbers by who you select, of course. But suppose the average is about 15 speakers per company. That translates to 1260 different speakers potential customers could select. Even if this is off by half, that's still over 600 speakers to consider. I've never met anyone that listened to even a quarter of that number of speakers before selecting something for their home theater. So how do potential buyers reduce the number of choices?

I went through this exercise in March/April 2018 while upgrading my L/C/R/SL/SR and it was a challenge. Available data was spotty and I was instead forced to cull based on price, speaker type, reviews, availability, auditions, AVS input, and SAF. Lack of data made a lot of these choices feel arbitrary. Did I get the best choice for money invested? Without data, how could I possibly answer that with any certainty? After an investment of just under $10k msrp, that's not a very satisfying answer.
unretarded and Lonewolf7002 like this.

Just one more upgrade and things will be perfect.
Bill-99 is online now  
post #229 of 4433 Old 01-01-2019, 02:30 PM
Member
 
Floyd Toole's Avatar
 
Join Date: Jun 2017
Location: California
Posts: 763
Mentioned: 80 Post(s)
Tagged: 0 Thread(s)
Quoted: 794 Post(s)
Liked: 2864
Quote:
Originally Posted by R Harkness View Post
But if your research recommends a gradual even sloping DOWN of frequencies off axis, what does this suggest for the preference rating of omni-directional speakers that put out essentially even energy radiation in to the room?
.
I could answer your questions here but I am tired of extensively rewriting the contents of my book. So, please just get it and read it. I am confident that your questions will be answered, including those about panels, Quads, and omnis. Nothing is "magic".

As for the sloping down thing, the research does NOT "recommend" it. If you get to read my book you will learn that a downward sloping curve is the inevitable resulting steady-state room curve for highly rated forward-firing loudspeakers. Figure 12.4 explains in detail. Equalizing a lesser loudspeaker to match that room curve does not guarantee equivalent sound quality.

In contrast, Figure 7.20 shows that the in-room curve of an approximately omni Mirage M1 follows the sound power very well indeed, and it is flatter. So-o-o-o. I go to great lengths in the book to deter people from thinking that the steady-state room curve is a reliable indication of sound quality. It is not.

If you don't want to buy the book, here is an open access JAES paper that addresses the basic issue of anechoic vs room curves and the folly of believing in Room EQ, something that the cinema industry is finally recognizing (Chapter 11).
Toole, F. E. (2015). “The Measurement and Calibration of Sound Reproducing Systems”, J. Audio Eng. Soc., vol. 63, pp.512-541. This is an open-access paper available to non-members at www.aes.org. http://www.aes.org/e-lib/browse.cfm?elib=17839
Floyd Toole is offline  
post #230 of 4433 Old 01-01-2019, 02:49 PM
Advanced Member
 
TimVG's Avatar
 
Join Date: Apr 2011
Posts: 548
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 266 Post(s)
Liked: 399
Quote:
Originally Posted by Floyd Toole View Post
The graph you show is ancient, having been done using the long retired MLSSA system. The curves in my book, Figure 12.1 are from the updated Harman designed system, which was excellent, but keeping it going through frequent OS upgrades required a resident computer engineer. The latest ones use the Klippel system which is someone else's responsibility. .............

Thank you for elaborating on the measuring methods, Dr. Toole. Owning and having read your book, I'm quite familiar with the content. I'm sure once new data (with the Klippel system) becomes available on the Salon2, it'll paint a more clear picture - comparing apples to apples and all.

PS: If you recall- you included some measurement data in your book of older monitors, one of them the JBL 4320 with the horn/lens combination. You called it quite well behaved (for its time I assume) apart from needing an equalization tilt. One of my guilty pleasures is playing around with these old systems, and based on your high resolution data in the book, and my own measurements I was able to turn my DIY project with the same horn/lens (plus the 2405 slot tweeter) from the 'standard' curve into this, through a modern DSP processor. medium <800hz data omitted due to irrelevance/accuracy. It's no full spinorama, but these old parts will not have sounded better in their lifetime. Showing the on-axis response, listening window and early reflections curve in 1/24 octave smoothing.


Last edited by TimVG; 01-02-2019 at 01:47 AM.
TimVG is offline  
post #231 of 4433 Old 01-01-2019, 03:39 PM
AVS Forum Special Member
 
CruelInventions's Avatar
 
Join Date: Mar 2004
Location: Chicago-ish
Posts: 6,551
Mentioned: 7 Post(s)
Tagged: 0 Thread(s)
Quoted: 1210 Post(s)
Liked: 1419
Quote:
Originally Posted by R Harkness View Post
Thanks. The original posts seemed to be quite Harman Kardon spinorama-centric.

You've been corrected twice, not sure why you keep using HK for your representation of Harman. It's been distracting me from your otherwise thoughtful contributions, like observing a well-tailored man walking around with a piece of toilet paper trailing from his shoe.
CruelInventions is online now  
post #232 of 4433 Old 01-01-2019, 03:43 PM
AVS Forum Special Member
 
chikoo's Avatar
 
Join Date: Dec 2011
Posts: 1,213
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 87 Post(s)
Liked: 40
Quote:
Originally Posted by torii View Post
many, many smart people have lost their shirts/bank rolls on science...I thought we all could agree flat sounds like shiit but guess not
Humans, by their very nature, love curves, bumps and slopes , in the right spots, and very few prefer flat

Last edited by chikoo; 01-01-2019 at 04:12 PM.
chikoo is offline  
post #233 of 4433 Old 01-01-2019, 03:45 PM
Member
 
neutralguy's Avatar
 
Join Date: Dec 2009
Posts: 104
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 79 Post(s)
Liked: 68
While I'm a big fan of the sound science that's been well summarized here, I do think the premise of this thread has been overstated from a scientific standpoint. The science has not shown how to choose a loudspeaker based on measurements. For how to choose a speaker, a reasonable goal would be overall satisfaction with purchase over some period of time, and that is far from what the research has even studied. What the Harman research has shown is how to choose a speaker that will be preferred under blind listening in a reference room, and that is far from translating directly into consumer satisfaction.

A few areas of discrepancy:

- Consumers don't listen under blind conditions. If speaker A is perceived as sounding better than B because it looks more impressive or has me so convinced because of better marketing, would I really be happier having B in my home?

- Consumer rooms are not reference rooms. Has Harman repeated the same listening tests in a sampling of rooms of its customer base?

- Sound quality is only one of many reasons that affect overall satisfaction with the speaker. Looks, value, reliability, perception, and many other factors can all play a role.

Perhaps some of these discrepancies explain why although Revel's speakers are unbeaten in the lab, they don't seem unbeaten in popularity.

It would be like saying 0-60mph times predict perceived acceleration, therefore it tells you how to choose a car. That would be an unsupported claim. To support it we would have to show that 0-60mph times predict overall satisfaction, say via the JD Power surveys.

Don't get me wrong, Harman speakers sound great to me, but if the topic is about science we do want to be precise about what exactly the science has shown and what it has not.
neutralguy is offline  
post #234 of 4433 Old 01-01-2019, 03:51 PM
Advanced Member
 
modenacart's Avatar
 
Join Date: Nov 2008
Posts: 518
Mentioned: 1 Post(s)
Tagged: 0 Thread(s)
Quoted: 282 Post(s)
Liked: 180
Quote:
Originally Posted by neutralguy View Post





It would be like saying 0-60mph times predict perceived acceleration, therefore it tells you how to choose a car. That would be an unsupported claim. To support it we would have to show that 0-60mph times predict overall satisfaction, say via the JD Power surveys.


This analogy has nothing to do with what has been stated in this thread. It is just some random car stat.




Sent from my iPhone using Tapatalk
modenacart is offline  
post #235 of 4433 Old 01-01-2019, 04:06 PM
AVS Forum Special Member
 
Milt99's Avatar
 
Join Date: Jan 2002
Location: The Great Pacific Northwest
Posts: 5,858
Mentioned: 6 Post(s)
Tagged: 0 Thread(s)
Quoted: 496 Post(s)
Liked: 491
Like most things in life, no well-intentioned deed goes unpunished.

A sincere Happy New Year.
bearr48, Lonewolf7002 and mik2h like this.

 
This isn't Nam Smokey, there are rules here!
Milt99 is offline  
post #236 of 4433 Old 01-01-2019, 04:17 PM
AVS Forum Special Member
 
chikoo's Avatar
 
Join Date: Dec 2011
Posts: 1,213
Mentioned: 0 Post(s)
Tagged: 0 Thread(s)
Quoted: 87 Post(s)
Liked: 40
Quote:
Originally Posted by neutralguy View Post

It would be like saying 0-60mph times predict perceived acceleration, therefore it tells you how to choose a car. That would be an unsupported claim. To support it we would have to show that 0-60mph times predict overall satisfaction, say via the JD Power surveys.
In every test that the BMW 3 series was voted as the best drivers car, it never was on top of any of the measurements that were used, including hp, torque, 0-60, top speed, slalom speeds, etc

Is BMW willing to spell out what makes the 3 series so lovable? Of course not. Why would they?
chikoo is offline  
post #237 of 4433 Old 01-01-2019, 04:48 PM
AVS Forum Addicted Member
 
R Harkness's Avatar
 
Join Date: May 2001
Location: Toronto, Ontario, Canada
Posts: 15,129
Mentioned: 17 Post(s)
Tagged: 0 Thread(s)
Quoted: 2756 Post(s)
Liked: 2675
Quote:
Originally Posted by CruelInventions View Post
You've been corrected twice, not sure why you keep using HK for your representation of Harman. It's been distracting me from your otherwise thoughtful contributions, like observing a well-tailored man walking around with a piece of toilet paper trailing from his shoe.
Ah, I have indeed misunderstood the distinction. I have thought it was the HK division of Harman that has been doing
the tests. I did not see any corrections in replies to my posts, save that HK was a small division within Harmon - which
didn’t disambiguate my misunderstanding.

Ok Harman it is. Thanks.
CruelInventions and mik2h like this.
R Harkness is offline  
post #238 of 4433 Old 01-01-2019, 05:00 PM
Member
 
buckchester's Avatar
 
Join Date: Dec 2006
Posts: 161
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 79 Post(s)
Liked: 48
Great thread. Thank you to all the experts who have provided extremely valuable information for us all, especially Dr. Toole.

If I am interpreting things correctly, then I assume this first speaker likely sounds better than this second speaker, since the frequency response appears to be smoother.

First speaker:
https://www.soundstage.com/measureme...m_studio20_v3/

Second speaker:
https://www.soundstage.com/index.php...nts&Itemid=153


Would the resident experts here agree with this?

Thank you.
buckchester is offline  
post #239 of 4433 Old 01-01-2019, 05:02 PM
Member
 
Join Date: Jul 2018
Location: It looks like ML took his ball and went home. :)
Posts: 191
Mentioned: 2 Post(s)
Tagged: 0 Thread(s)
Quoted: 105 Post(s)
Liked: 51
Quote:
Originally Posted by torii View Post
well I guess so...most people know what they like and dont need science to tell them
Or place such a high regard for room correction software as though audio loudspeakers that have been around since before this technology came into vogue are worthless without said software. It's a gimmick, but it sells, apparently.
bootster is offline  
post #240 of 4433 Old 01-01-2019, 05:07 PM
AVS Forum Special Member
 
Molon_Labe's Avatar
 
Join Date: Jun 2015
Location: Texas
Posts: 2,190
Mentioned: 105 Post(s)
Tagged: 0 Thread(s)
Quoted: 3347 Post(s)
Liked: 4027
Quote:
Originally Posted by chikoo View Post
Humans, by their very nature, love curves, bumps and slopes , in the right spots, and very few prefer flat
Absolutely. The problem is that a bump or curve found in the wrong place is called a tumor. Without measurements, your playful groping in the dark might not be what you think it is
sdurani, Vergiliusm, Muza and 10 others like this.
Molon_Labe is online now  
Sponsored Links
Advertisement
 
Reply Speakers

Tags
cea 2034 , double-blind , listening tests , loudspeaker measurements , spinorama

Thread Tools
Show Printable Version Show Printable Version
Email this Page Email this Page


Forum Jump: 

Posting Rules  
You may not post new threads
You may not post replies
You may not post attachments
You may not edit your posts

BB code is On
Smilies are On
[IMG] code is On
HTML code is Off
Trackbacks are Off
Pingbacks are Off
Refbacks are Off