How should a speaker measure on axis??

[Robh3606]

I was reading some posts at AA and was directed to a review at Stereophile of the new Wilson Audio speakers. The price is about 45K for the pair. I thought the measurements section had some interesting comments. I then took a look at the new B+W speaker and the Tetra as well. What do you guys think?? I am going to reserve my comments for now.

Rob:)



http://www.stereophile.com/loudspeakerreviews/805wilson/index4.html



http://www.stereophile.com/loudspeakerreviews/805bw/index3.html



http://www.stereophile.com/loudspeakerreviews/805tetra/index3.html

[Don C]

Maybe we should repost that article by Floyd Toole here. The gist: Flatter response sounds better.

[Zilch]

None of those would make it past the garage testing laboratory here, I don't believe.

WHEW! I'm glad I stick with two-ways. :p

[Loved the "Honk" part....]

[Mr. Widget]

Rob, Zilch, you should read this document.

http://www.harman.com/wp/pdf/Loudspeakers&RoomsPt2.pdf

Anyone really interested in this stuff should read all of these white papers... there is an incredible wealth of information. The published talks by Floyd Toole are particularly interesting. They really explain a lot of the stuff that I have been slowly becoming aware of over the years through my trial and error audio exploration.

http://www.harman.com/wp/index.jsp?articleId=default

There is a hell of a lot to digest here, but it really clears up a lot of the misconceptions that are promoted by the audio community.

Widget

[Mr. Widget]

Maybe we should repost that article by Floyd Toole here. The gist: Flatter response sounds better.

http://www.reed-electronics.com/tmworld/article/CA475937.html

It is interesting but doesn't really have the necessary detail... read the "white papers" they really tell the story.

Widget

...on reflection, I suppose this article does tell enough of the story for speaker buyers... but coming from the perspective of a speaker builder, I was really thankful that Dr. Toole and Harman were willing to take the time to share so much in their white papers.

[boputnam]

:spin:

A friend was recently thrilled with getting a pair of the ProAc Response D80 (http://www.proac-loudspeakers.com/d80.php). OK, they're ugly, sure, but what do they sound like?

"Frequency Response
20hz to 30Khz"

Really? But is that at +/-3dB...??? I'd bet +/-12dB, or worse (20Hz, my butt!). ProAc no-where specifies the dB envelope for any of their products.

A prior pair of the ProAc "Future Point Five" never impressed me - they had no bass reponse at all, and using my SmaartLIVE the crossover point was a whopping notch (about -10dB) at exactly 700Hz. I tired to talk to their technical support about this, but only got: :blink: I sensed they were not accustomed to direct questions about their engineering - "you mean, you actually measured something...?" Many audiophiles don't believe in EQ, so the notch just lived there, in this otherwise great listening room powered by monster Krells.

Anyway, as Lancer sez, "I sure got the JBL's I want", and properly set-up they far surpass anything else out there.

[boputnam]

Rob, Zilch, you should read this document.

http://www.harman.com/wp/pdf/Loudspeakers&RoomsPt2.pdf

Don't forget Part 1!! :scold:

http://www.harman.com/wp/pdf/Loudspeakers&RoomsPt1.pdf

[Robh3606]

Hey Widget

I have them saved and use them as references. You are right there is a wealth of information there. I like the JBL Tech Notes too. Lots of good information and good references in the back. The Revel sight used to have a couple of good papers up too. Don't know if it's still there or not. Having said that it's clear what camp I am in. I will take flat on axis with uniform power response. I am always surprised when I see a SOTA design that seems to ignore this. I quess design philosphy has it's place but it's always a surprise when it seems to conflict with good engineering practice.

Rob:)

[DavidF]

Hey Widget

...I will take flat on axis with uniform power response. ...
Rob:)

Agreed, since I need to use toe-in and and close-to-side-wall positioning in my little space. But much easier said than done. Very difficult with some crossover slopes, especially with a two-way design. Always have to know where the correct axis points, as well, to establish a reference going off axis.

David F

[JBLnsince1959]

About what!?
I've got the JBL's I want, don't you? :p

:jawdrop:

that's the first time I've heard anyone say they have all the jbl speakers they want :D

surely there must be something JBL you want and don't have...

[Mr. Widget]

...and properly set-up they far surpass anything else out there.


Now really? Anything/everything??? :hmm:


Widget

[boputnam]

Now really? Anything/everything??? :hmm:
Widget:yes:

And I am one with expensive tastes, boy-howdy... :o:

[edgewound]

The human ear is a remarkable device...and can be a very failing device. You can set up a number of different "flat" measuring speaker systems that all have great engineering and construction practices incorporated into them....yet they will all sound different to the ear...and everyone else's. Because all of our respective ears have been "calibrated" differently. All the sophisticated measurement tools lack one characteristic...subjective listening tastes. That's why there are so many speakers on the market. I think the white papers all have some great scientific info to absorb....but the bottom line is...at the end of the day the white paper is used as a sales tool (Toole?) to sell more loudspeakers. What do your ears say?

[Robh3606]

"All the sophisticated measurement tools lack one characteristic...subjective listening tastes. That's why there are so many speakers on the market. I think the white papers all have some great scientific info to absorb....but the bottom line is...at the end of the day the white paper is used as a sales tool (Toole?) to sell more loudspeakers. What do your ears say?"

Hey Edgewound

How true think how boring things would be if it all sounded the same. Of course they are a sales tool however what I like about the Harman stuff is all the peer reviewed references that their stuff is based on. In many ways it set's what they do apart from the rest. I still think JBL is leading the way in many respects and to see many of the authors of those papers as employees both present and former is impressive to say the least. As far as what my ears say well I would be preaching to the choir around here.

Rob:)

[Mr. Widget]

...what I like about the Harman stuff is all the peer reviewed references that their stuff is based on. In many ways it set's what they do apart from the rest. I still think JBL is leading the way in many respects and to see many of the authors of those papers as employees both present and former is impressive to say the least.
I agree... and I agree with Mr. Edge's "calibrated" comments... after all, Bo found his perfectly "calibrated" speakers...:yes:

I really would like to know exactly what the "random musical selections" are that Harman uses for their shuffle comparisons and they make a point that they get the most consistent results while using ears that they have trained. I wonder just how biasing that "training" might be.

It seems that if the results of Harman's exercises were complete and accurate... according the the "sales toole", JBL and Harman speakers would be the run away best sellers. I think the tests are very good but even Dr. Toole admits they do have a ways to go yet. I do think that they are on the right track, but perhaps are still not asking all the right questions.

Widget

[Ian Mackenzie]

:spin:


Anyway, as Lancer sez, "I sure got the JBL's I want", and properly set-up they far surpass anything else out there.

Well you not prone to excessive exaggeration like some I know so that is probably true...Muhahahah.

Having taken the day off to compete with the winter blues I took a shine to that rather large cup of popcorn.

I thought the paragraph on "trained listeners" was the most interesting aspect of the whole article.

Of all the people JBL gurus I have met Bo does have a "trained ear" as he can detemine not only a frequency variation, but also if the musical instrument sounds true.

This was the situation as I understand it at a previous group evaluation of project May.

As the article say "trained listeners share untrained listerners preferences in speakers, but they tend to be harder to please".

I heard he cracked actually at that meeting.......yes he is a difficult listener to please at the best of times but then he is a 4345 listener.....LOL.

The rest as Bo says is correct set up in the listening room, this is where the nuisances variables are so important.

Alas, as I found out in blind test I did last weekend with a trained listener he was dissatisfied with the loss of midrange transparency and spatial ambience in a full passive crossover simulation of a 4343 and entirely satisified with bi amp mode.

The techical reason by the way was the way in which a single capacitor was used in the crossover network.(see 4343-4344 upgrade discussion thread)

The system frequency response however was identical!!

Back to more head scratching....:hmm:

Ian

[Mr. Widget]

Alas, as I found out in blind test I did last weekend with a trained listener he was dissatisfied with the loss of midrange transparency and spatial ambience in a full passive crossover simulation of a 4343 and entirely satisified with bi amp mode.

This directly relates to my questions about musical selection and "training"... it is one thing to hear a neutral sounding speaker with nice timbral balance, but the midrange transparency and spatial ambiance seem to frequently be missing from so many of these system that we hold so dear.

Widget

[Ian Mackenzie]

I agree... and I agree with Mr. Edge's "calibrated" comments... after all, Bo found his perfectly "calibrated" speakers...:yes:

I really would like to know exactly what the "random musical selections" are that Harman uses for their shuffle comparisons and they make a point that they get the most consistent results while using ears that they have trained. I wonder just how biasing that "training" might be.

It seems that if the results of Harman's exercises were complete and accurate... according the the "sales toole", JBL and Harman speakers would be the run away best sellers. I think the tests are very good but even Dr. Toole admits they do have a ways to go yet. I do think that they are on the right track, but perhaps are still not asking all the right questions.

Widget

Yep,

The logic was a bit flawed but a nice PR job for the shareholders.

I mean why not use a set a really good headphones and figure it out from there. Screw what idiot customers think, unless the whole deal is a white wash on market research for selling a sound at a price point that most people want to hear. Nor did they discuss how badly people set up their speakers at home or timbre accuracy or tonal colourations and the like...that has in my view never been a focus on the JBL sound.

People are people and some will always buy what looks nice and run for the west coast sound..they just love those dynamics.

[Ian Mackenzie]

This directly relates to my questions about musical selection and "training"... it is one thing to hear a neutral sounding speaker with nice timbral balance, but the midrange transparency and spatial ambiance seem to frequently be missing from so many of these system that we hold so dear.

Widget

Well as i said in may last post I think so too.

I heard a friends $1000 Dali's the other day and they seemed to have more timbre and warmth than a JBL per say.

That's annoying isn't it!

Oddly I did the big band thing again on Sunday and I thought it was all ..well nothing too surprising..even dry sounding..maybe our beloved JBLs have it right after all.

Perhaps we just expect something more listening at home!

Ian

[Mr. Widget]

Perhaps we just expect something more listening at home!


Reproduction at home can never be quite what live music is... they are just different experiences. Live music is typically muddy, smeared, jacked up on the bottom and missing the highs.... but who cares, there is something wonderful about really getting what the musicians are feeling.


Widget

[Robh3606]

If you are interested in what the Training is look here

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

and here

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

[Mr. Widget]

I have read those, but I still can't say that I am satisfied that I have a complete grasp of their training and whether or not the training might not bias listeners toward one type of listening experience over another. The discussions of statistical analysis were thorough enough and enlightening, but their description of exactly what they are asking their listeners to focus on other than frequency content and anomalies really isn't discussed. If in fact they are only teaching the listeners to listen for a flat response then it is no wonder that the most linear systems win...

I certainly agree that a flat response with a slightly downward tilt from the bass to the treble is highly desirable, but it is only one of many things I listen for in a loudspeaker.


Widget

[Ian Mackenzie]

Agreed,

The more I think about it the more I think its hype, sure a panel is important but so is the interaction of a given room...you hear 1/2 the room after all.

If a system is that far out of wack that it takes a national focus group to determine if the systems are acceptance then they need to start with a clean sheet of paper.

It seems to me Tool & co are attempting to market the most pleasing system for a given price point in the hope its quick sale before the saleman at the En Mass appliance Mart has time to switch around all the other systems.

I really don't think that this adds up to an accurate and certainly not the best speakers for the money. Besides, only a very small bunch of designers/manufacturers build a very small range of really good loudspeakers for an equally small number of audiophiles.

[Robh3606]

Ian/Widget

I am curious that this put off as hype. You did see the AES papers on blind testing and speaker evaluations?? That not just some advertising copy that is peer reviwied and published work.

What exactly do you listen for?? Part of the problem is the media is fundimentally flawed. Half the audiophille terms are to describe soundstage, depth and imaging. All three are completely artificial and a product of the recording techniques and the skill of the production team. All three are directly affected buy smooth frequency response and power response. The LSR series was designed along the same philosophy as the referenced papers as were the Performance series and the 9800. They were all evaluated and voiced using the techniques in those papers using the room as an evaluation tool. Do you really think it's all hype?? I don't see anyone else besides BW doing RD like they do and if you look at the measured responses at the start of the thread the two philosophies BW and Harmans are in line. Look at the measurements. The 2 other speakers are botique type companies aimed at the audiophille market. Do you want a speaker based on sound design techniques or someones elses take on what they think is the best based on subjective impressions only where engineering takes a back seat???

Rob:)

[Ian Mackenzie]

Very Interesting points Robert,

But whole flat on/off axis ever which way thing has nothing to do with a speaker sounding accurate...why do some prefer aluminiun diaphrams and others pure dynamic non horn loaded systems.

Also being able to pick the eyes out of a recording with very linear monitor is not necessarily a good musical experience.

Besides a well designed cone system wins lands down on all scores imho..time to make a good 4 way active crossover and start sampling drivers.

yawn:snore: \

Ian

[Lancer]

Do you want a speaker based on sound design techniques or someones elses take on what they think is the best based on subjective impressions only where engineering takes a back seat???I want a loudspeaker based on G.T.'s cumulative practical experience along with his having worked with, or been inspired by, Keele, Smith, Gander, Eargle, Margolis, Augspurger, Moro, Button, Sorensen, Daniels, May, Locanthi, etc. :p
From the 4315 till today, it's been a fantastic ride. :yes:

[JBLnsince1959]

I want a loudspeaker based on G.T.'s cumulative practical experience along with his having worked with, or been inspired by, Keele, Smith, Gander, Eargle, Margolis, Augspurger, Moro, Button, Sorensen, Daniels, May, Locanthi, etc. :p
From the 4315 till today, it's been a fantastic ride. :yes:

and what a "cumulative practical experience" it is (plus he has the education and brains). Stop and think about how many different designs they have done over the years. A lot of high-end speaker companies take one design and keep tweating it, BUT JBL ( GT) has tried and studied so many different designs over the years for speakers. L300, L220, L212, 4435, other studio montiors, 4430,5's, 250's and other TI's, Cascade, XPL, L series, TiK, K2 and other "Statement" speakers, performance series and the list goes on. I'm not sure of any other company ( or person) and has had that much experience with so many designs.

[Robh3606]

"I want a loudspeaker based on G.T.'s cumulative practical experience along with his having worked with, or been inspired by, Keele, Smith, Gander, Eargle, Margolis, Augspurger, Moro, Button, Sorensen, Daniels, May, Locanthi, etc. :p
From the 4315 till today, it's been a fantastic ride. :yes:"

Hello Lancer

Yes a balance of both engineering and listenning skills .

Hello Ian

"But whole flat on/off axis ever which way thing has nothing to do with a speaker sounding accurate...why do some prefer aluminiun diaphrams and others pure dynamic non horn loaded systems."

I disagree with you on the "whole flat on/off axis ever which way thing has nothing to do with a speaker sounding accurate...". I am not saying it's just that but I think it is an excellent starting point from an engineering standpoint. If the speaker is difficult to place due to poor off axis response and is tonally incorrect from the start because of poor on axis response combined with poor power response where do you go from there???? The whole idea of the uniform power response is more predictable room integration. The uniform power response also makes EQ more feasible and effective. Those are plus's in my book because I want speakers that will work better in my room.

Define accuracy?? If you can't get the Frequency Domain correct it's not accurate under any definition.

"why do some prefer aluminiun diaphrams and others pure dynamic non horn loaded systems"

I don't know?? Some guys drop jars of rocks on their speakers.

"Also being able to pick the eyes out of a recording with very linear monitor is not necessarily a good musical experience."

Well it all depends on what you personally determine is a good musical experience. I like clarity and tonal accuracy. I also like listenning to music on a boom box and my brother in laws Bose systems. It's all relative to where you are, what you are doing, and your frame of mind. I can enjoy music under damn near any circumstances however most are not what I would define as accurate reproduction.

Rob:)

[Mr. Widget]

"I am curious that this put off as hype." "The LSR series was designed along the same philosophy as the referenced papers as were the Performance series and the 9800. They were all evaluated and voiced using the techniques in those papers using the room as an evaluation tool. Do you really think it's all hype??"-Rob

I don't believe I ever said it was hype. I did agree with Edgewound that the "White Papers" were a sales tool... very different from hype. I would imagine the white papers are a sales tool for the the Harman companies, but they are also probably a sales tool for the design and engineering teams to keep getting the funding from those that hold the purse strings. These documents show the bean counters where the money is going and why they should continue spending it on these research facilities.

"You did see the AES papers on blind testing and speaker evaluations?? That not just some advertising copy that is peer reviwied and published work."-Rob

Sure, but it isn't peer reviewed like the medical journals. The AES papers are frequently full of opinion, and erroneous "facts"... just go back and look at them over the years... I am not painting Harman's work with this brush, simply pointing out that presenting a paper to the AES isn't the same as discovering a new law of physics.


"I don't see anyone else besides BW doing RD like they do and if you look at the measured responses at the start of the thread the two philosophies BW and Harmans are in line."-Rob

I have always been an advocate of keeping the science in the design of loudspeakers and audio in general. The early work at the BBC was fundamental in equating what we hear and what is produced by speakers, and the work done by the National Research Council of Canada has been extremely important. (Didn't Dr. Toole come from there?) I do see the work done at B+W and Harman as continuing this very important work in the same vein.

One of my concerns is that the engineers and designers at Harman have possibly adjusted their experiments consciously or subconsciously to impress the in house bean counters... testing like this is very expensive. Even though Harman is a huge company, they are still cost benefit ratio driven.

"What exactly do you listen for?? Part of the problem is the media is fundimentally flawed. Half the audiophille terms are to describe soundstage, depth and imaging. All three are completely artificial and a product of the recording techniques and the skill of the production team. All three are directly affected buy smooth frequency response and power response."-Rob

I listen for a neutral tonal balance. I want a speaker that does not have a sonic signature. (they all have one, it is a matter of degree) I want a speaker that can present the most delicate detail without masking it with colored sound, distortion, or sluggish performance. I want a speaker that has the ability to play thunderously loud or soft and still sound essentially the same. ( Fletcher and Munson not withstanding)

I agree that holographic imaging is an audiophile illusion and it isn't real... but that's what stereo is all about, creating an illusion. It is not how we hear in live music, it is something else entirely, but when done well it can be quite magical.



"I want a loudspeaker based on G.T.'s cumulative practical experience along with his having worked with, or been inspired by, Keele, Smith, Gander, Eargle, Margolis, Augspurger, Moro, Button, Sorensen, Daniels, May, Locanthi, etc."-Lancer

This is certainly a large part of it. Talented and experienced designers having the best testing facilities in the world can yield fantastic results. There is still one component that is possibly lacking. The best audio companies have been those with an Avery Fisher, Saul Marantz, or James B. Lansing at the helm... I am not sure a large multinational corporation really has the ability to take the risk necessary to really produce greatness.



Widget

[Mr. Widget]

Back to the original question...

I bet that the Wilson Maxx2 would measure better using the JBL axial spin system in their anechoic chamber than John Atkinson was able to do with MLSSA in his "lab". Since the speaker has so many drivers it is very difficult to measure outside of a real anechoic chamber.

I have listened to them and I really doubt that many would find them bad sounding.... though I did think the bass was a bit exaggerated.

Widget

[DavidF]

Quote:
Originally Posted by Robh3606
Do you want a speaker based on sound design techniques or someones elses take on what they think is the best based on subjective impressions only where engineering takes a back seat???


I want a loudspeaker based on G.T.'s cumulative practical experience along with his having worked with, or been inspired by, Keele, Smith, Gander, Eargle, Margolis, Augspurger, Moro, Button, Sorensen, Daniels, May, Locanthi, etc. :p
From the 4315 till today, it's been a fantastic ride. :yes:

I find the question and answer above the crux of my position on this matter. No one has yet produced a one-size-fits-all speaker system. Accuracy as such is still relative, a threshold beyond which are personal preferences and bias. The compromises involved in speaker design are much better understood today but they have not been eliminated or blended into a universally accepted recipe for accuracy.

Ian mentioned his experience with the Dali and the warmth and timbre associated with that design. This suggests to me a purposeful design element to satisfy an audience that relates to certain experiences in live (acoustic?) music. But will this system work for all types of music genre, in my room as well as someone else’s? Widget mentioned a preference for a total system response tending tipped downward, going from bottom to top. This does seem to be a more natural response tendency further exemplified by the extensive THX study on theater sound (admittedly focused on speech articulation in large rooms but does also address the total media aspect in the theater as well as the home environment). But does this work for horns and direct drivers equally, in large rooms, dead rooms, so on? Then there is the old East Coast-West Coast sound, the BBC curve, the “best-with-tube curve”, even a National Research Council sound (NRC in Canada whence Dr. Toole had considerable influence).

Each of these generalized sound styles follow a bias, or preference which seems nicer to say, carried through to the design philosophy as to what seems natural, but maybe not accurate. This bias is often sustained as a brand philosophy (AR- voice of the concert hall, Altec- Voice of the Theater!). So I understand Rob’s concern that designing for accuracy starts with fundamentals such as consistent frequency power response. But I also suggest that consistent axial/power response is but one of many, many variables that come into play. Response lobing, phase response, group delay, TIM distortion, FM distortion, defraction effects, etc. Studies have indicated that the ear is forgiving to certain frequency response anomalies but quite sensitive to time and spatial distortions. If a designer has to fudge on one element to improve on another, so it goes. Always a compromise to consider.

I would speculate that Dr. Toole’s primary function at Harman is very similar to that with the NRC, to further the study and understanding of the total audio/acoustic system. This, and to give Harman first shot at the benefits of all this research. Not hype by any means but I doubt he was commissioned to design the perfect loudspeaker system. That remains with the engineers and technicians who carry with them experience, trained ears, and, yes, bias whether personal or part the company’s brand heritage. So, yes I can also support Lancer’s position for holding faith in the heritage of design philosophy at JBL, if for no other reason that it sounds good to me.

David F

[Robh3606]

"I bet that the Wilson Maxx2 would measure better using the JBL axial spin system in their anechoic chamber than John Atkinson was able to do with MLSSA in his "lab". Since the speaker has so many drivers it is very difficult to measure outside of a real anechoic chamber."

Yes I would like to see it measured there too. I would also like to see a couple of the clasic monitors in there too but we know that's not going to happen. You do have to wonder how much of the bloom down low is room related.

Rob:)