What do you all think of acoustic waveguide technology?

[Nuance AH]

Just for reference, if anyone is interested in reading Sean Olive's "Differences in Performance and Preference of
Trained versus Untrained Listeners in Loudspeaker Tests: A Case Study," PM me and I'll send it your way. However, I don't think he's the all mighty guru of who's studies, tests and opinions should be considered fact. Still, it's a great read. Check it out if you have time.

 

[Nuance AH]

I will put this into perspective and make clear the bottom line: it's easiest and cheapest to produce a maximally flat on axis and relatively wide off axis response set by using the proper waveguide. The superior method is to use very large radius on the horizontal corners, in the range of at least 3" radius. This allows a wider dispersion and also helps the upper band of the mid-bass i most cases. Cost to do this as compared to molding a waveguide face plate for the tweeter is enormous.

The description of using DSP to correct linearity behaviors is mostly accurate, but seemingly over-stated. On any single system, DSP will allow a better and more ideal target response to be realized. Even more importantly, it allows easy and accurate adjust ability of baffle step correction and treble graduation response; which must ideally be adjusted differently for every acoustic environment and placement, as well as to listener preference in regards to their choice of recordings.

-Chris

Thank you; now that's what I was looking for. Finally, back on topic.

 

[avaserfi]

Seemingly, the bulk of the article presents the notion that any loudspeaker design not utilizing a waveguide design (such as theirs) causes a variety of diffraction "types" and "events"...including, perhaps especially, flat baffle designs (which the author calls actually a "bad [hemispherical] waveguide design"). Their point? Impedance levels and boundary conditions are both optimized with SP speakers, facilitating extended bandwidth and edge diffraction elimination.

Of course they will say such things, they are after all trying to sell their design .

Diffraction issues are multifaceted, yes? So ignoring for a moment the existence of the spheroidal waveguide, what elements in your speaker designs do you utilize to eliminate or minimize diffraction issues such as discontinuity of air pressure at the baffle plane, impedence mismatches, pressure zones causing secondary radiation, etc? Or is it even that much of an audible distortion problem? Looking for wisdom here.

There is a very good article on the subject here. This article essential shows that if a non-wave guide is used the baffle itself will act as a waveguide. To prevent diffraction issues at minimum 3" radii should be used on the corner to be fully effective. Of course, this method is not economically feasible relative to use of a waveguide, but properly implemented will achieve far superior results.

-Andrew

 

[Nuance AH]

Of course they will say such things, they are after all trying to sell their design .

Now that I agree with.

There is a very good article on the subject here. This article essential shows that if a non-wave guide is used the baffle itself will act as a waveguide. To prevent diffraction issues at minimum 3" radii should be used on the corner to be fully effective. Of course, this method is not economically feasible relative to use of a waveguide, but properly implemented will achieve far superior results.

-Andrew

Has this method been tried and tested thoroughly, including using a wide variety of listeners to prove a 3" radi provides an noticeably audible difference over, let's say a 1.5" radi?

 

[avaserfi]

Has this method been tried and tested thoroughly, including using a wide variety of listeners to prove a 3" radi provides an noticeably audible difference over, let's say a 1.5" radi?

There is no need. The measurable differences between the diffraction effects of a 1.5" radii and a 3" radii can be correlated to known thresholds for frequency response anomalies as found by Bücklein.

-Andrew

 

[Nuance AH]

There is no need. The measurable differences between the diffraction effects of a 1.5" radii and a 3" radii can be correlated to known thresholds for frequency response anomalies as found by Bücklein.

-Andrew

So it has been proven to be audible to all human beings, then, or at least the majority of them? Forget the assumed correlation. Have tests and experiments been performed to prove this? I am not talking hypothesis through science here; I'm speaking of actual testing with many, many human subjects (such as what Olive did in his blind speaker test)?

 

[avaserfi]

So it has been proven to be audible to all human beings, then, or at least the majority of them?

Rather, it would be proper to say that this has been proved for human beings with defined 'normal' hearing noting that there will likely be outliers who are more and less sensitive to such anomalies.

Forget the assumed correlation. Have tests and experiments been performed to prove this? I am not talking hypothesis through science here; I'm speaking of actual testing with many, many human subjects (such as what Olive did in his blind speaker test)?

All of the work cited within this thread is not simply hypothesis through guessing, but rather, science based on actual listening tests using thousands of individuals in hundreds of listening groups. Of course, the author's start with a hypothesis and test it, but there is no actual way to test such hypothesis in a credible manner, without using human subjects. The step of correlating the results of this research to actual measurements successfully is only possible because of the strict, generalizable, manner that these tests are conducted. Simply put, it is not untested theory.

I strongly suggest you actually read this research (I have provided a thorough list of references) and my posts within this thread as I specifically deal with these topics.

- Andrew

Please note that it is not me drifting off topic, I am simply answering the questions asked.

 

[Nuance AH]

Rather, it would be proper to say that this has been proved for human beings with defined 'normal' hearing noting that there will likely be outliers who are more and less sensitive to such anomalies.



All of the work cited within this thread is not simply hypothesis through guessing, but rather, science based on actual listening tests using thousands of individuals in hundreds of listening groups. Of course, the author's start with a hypothesis and test it, but there is no actual way to test such hypothesis in a credible manner, without using human subjects. The step of correlating the results of this research to actual measurements successfully is only possible because of the strict, generalizable, manner that these tests are conducted. Simply put, it is not untested theory.

I strongly suggest you actually read this research (I have provided a thorough list of references) and my posts within this thread as I specifically deal with these topics.

- Andrew

Please note that it is not me drifting off topic, I am simply answering the questions asked.

I am not speaking of Toole and Olive's work, of which I have read many a time; nor am I talking about our previous converstaion (which we've taken to PM). I am talking about the last link you provided concerning the audible differences of a 3" roundover in comparison to smaller roundover, such as 1.5" I am simply asking if tests have been performed with listening subjects under controller conditions.

By the way, this does apply to the topic because it started based on the approach taken by acoustic waveguides.

Since Toole and Olive have been mentioned, I'll keep the question simple for you (so there can be no misunderstanding). Have there been tests and trials performed (similar to those that Toole and Olive have conducted) that prove a 3" roundover makes an audible difference, and not only an audible difference but rather a positive audible difference, in sound quality?

I am not trying to debate here, but rather simply find out the answer to my question. If the answer is yes, could you please provide a link to where I can acquire these studies? Thanks!

 

[avaserfi]

I am not speaking of Toole and Olive's work, of which I have read many a times, not am I talking about our previous converstaion (which we've taken to PM). I am talking about the last link you provided concerning the audible differences of a 3" roundover in comparison to smaller roundover, such as 1.5" I am simply asking if tests have been performed with listening subjects under controller conditions.

Without specificity in the question I could not have known. I thought you were asking about the research in general.

As far as I know there has been no specific perceptual research with regard to audibility of these different diffraction techniques.

As previously stated (post 45), this is because there is no need. Credible perceptual research is not cheap, so such an endeavor would be wasteful.

-Andrew

 

[Nuance AH]

Without specificity in the question I could not have known. I thought you were asking about the research in general.

I quoted your statement above, the asked the question. There wasn't anything confusing about it.

As far as I know there has been no specific perceptual research with regard to audibility of these different diffraction techniques.

As previously stated (post 45), this is because there is no need. Credible perceptual research is not cheap, so such an endeavor would be wasteful.

-Andrew[/QUOTE]
The same could have been said of Toole and Olive's work back in the day. You don't know that "such an endeavor would be wasteful." You are drawing conclusions based on assumptions.

Either way, just forget it. If you can find those links, I'd love to read all about it.

 

[avaserfi]

The same could have been said of Toole and Olive's work back in the day. You don't know that "such an endeavor would be wasteful." You are drawing conclusions based on assumptions.

In no way are these the same situations. There was no in depth understanding of how measurement correlated to listener perception before Toole, Olive and others. This data can be used specifically in regard to understanding diffraction effects on audibility.

One primary purpose of all research is to be generalizable, otherwise it would be useless. This is exactly what is being done when the results from the credible perceptual research (frequency response anomaly audibility) is being correlated to measured results from another study (diffraction effects).

Sadly, I have gotten caught up in this debate again. Rather than bring this thread farther off topic, I will attempt to refrain from debating the merits of credible perceptual research. Instead, I will attempt to focus on specific details of waveguides. Due to this no responses with regard to perceptual research will be had, from me, unless directly related to waveguides.

-Andrew

 

[mazersteven]

IMO this is a good discussion. One that is informative, and educational. I don't care what side of the fence you are on. It's good to read both perspectives on this issue.

 

[Nuance AH]

IMO this is a good discussion. One that is informative, and educational. I don't care what side of the fence you are on. It's good to read both perspectives on this issue.

I do to, but we've since taken it to PM. I suppose a new thread could be started if called for.

 

[Tomorrow]

IMO this is a good discussion. One that is informative, and educational. I don't care what side of the fence you are on. It's good to read both perspectives on this issue.

+1.

I find myself adopting both sides of the fence, though. (We all know how oddly Tomorrow thinks, anyway! )

I'm a firm believer in perceptual science. But I also know that as in all human senses testing, these well conceived experiments and concommitant conclusions are averages of the human experience. Tendencies are what result. Those results are only fact if you are discussing tendencies, therefore.

With speakers, there are innumerable influences upon individual speaker preference that have naught to do with a near distortion free, flat FR loudspeaker (e.g. room acoustics, amplitude of audio, amplification equipment, source material, music preferences and other biases, individual hearing anomalies, etc. I may be uniformed here, but no testing I'm aware of has controlled for all known contributing factors.) Digital EQ can resolve only so much of that. So I guess my take is that speaker preference is pretty situational...where the speaker-listener interface/preference is important, as is the ability to fine-tune the well-designed speaker signal with EQ given the individual listeners setting/equipment/biases.

And Brandon...I think there is fun to be had on both sides of this hobby fence....fun in the hunt for a desirable speaker and fun in the design of the best audio for the situation. We know from Tim's hunt that is detailed over on AVS that his perfect speaker(s) in the shop can, and did, turn pretty sour in his home.

EDIT: Finally, I guess we've all come to agree that a well designed waveguide is preferred but can be expensive to implement. Although it should be noted that there is little expense in some of the better implementations, such as the JBL EOS and PT waveguide designs.

 

[annunaki]

Whether you sound arrogant or not is besides the point. And even if you did, I wouldn't let it bother me. I've dealt with plenty of self proclaimed no-it-all's, and I've learned to cope (not saying you are, but you get the idea).

As far as the original thread, I am not, nor was I interested in it, but yet we somehow got back on that same topic. I created this thread concerning acoustic waveguides, not recommended B&W speakers. I noticed that some people were continuously making recommendations and arguing points while never having heard the equipment in question, so I asked why that is. Those are the only two things I wanted to know about, but yet we came full circle back to that thread? And if I recall, that thread was started because it had already gotten another thread off topic. Hmm...that's kind of humorous.

I also do believe in controlled tests and that the majority will agree (already did agree in Sean Olive's double blind test) that a flat FR is preferred, but that's not my argument. My argument is that most people will not purchase based on measurements only, or on someone else's recommendation who's never heard the speaker in question. They will use their ears, or at the very least take the suggestion of someone who's actually put some time in with said speaker(s). If you gave the majority a choice of choosing based on each, they'd choose their own ears in most cases.


I agree with all of this (well...most of it), so don't get the wrong idea. I always make multiple suggestions and help the person find a speaker that suits their needs. I never only recommend a speaker that I like. I also think that picking up a speaker then tweaking it to fit your needs is a great idea, though not practical for many people. I do not, however, agree with continuously recommending a speaker to fit someone's needs without every having heard it. One or two based on great feedback is a given, but over and over again concerning multiple speakers? It just doesn't make sense because objective measurements only tell us so much, especially when adding different room sizes and acoustics into the equation. We don't all listen in the same room, with the same acoustics, using the same gear, with our ears at the same level at the exact same listening distance. That is why I keep "pushing" my point, avaserfi.



avaserfi - I am familiar with the white papers and sources you cited; try not to assume. I don't believe everything I read, however, especially if it's from the same sources.

Again, we'll agree to disagree. In my opinion you are forcing your own opinions on people (or rather your opinion of other people's opinions, since you cite Floyd and Toole so often), and I don't agree with that, especially if you've never even listened to the equipment at hand. Sure, you've got some credible sources, and a good teacher (WmAx) but who says they are the be all end all final source? You keep using the same sources. This is why you and I will agree to disagree. There is no point to further argue the issue, because I won't budge and neither will you, so that's where I'll leave it. I am not one to argue something so subjective, even if you say it's quite objective (which is still isn't, at least amongst the majority).

Simply put, measurements don't tell the entire story. If you'd like to further discuss this, we can take it to e-mail or PM. It's been entertaining, but you guys take the fun out of the debate (and no, not because I'm wrong, because I'm not ).

Not to derail this discussion, but based upon the research done by Toole & Olive and other sources cited, I am unsure as to how you feel it is subjective research or an opinion of theirs???

What the reseach tends to show is that subjective listening preferences can be recreated with a linear output device and DSP to shape the "subjective" listening curve. This was proven through the blind listening tests. Andrew and Chris are simply pointing out that fact.

 

[fmw]

I'm not sure what a digital speaker is. All that aside, I quit listening to speakers in audio showrooms many years ago. I learned back then that room acoustics had more effect on sound than the speakers did so listening to speakers in someone else's room didn't correlate to what I would hear at home. I've bought speakers based on measurements and specs ever since and I'm not going back. So you can agree to disagree with me as well.

 

[Tomorrow]

I'm not sure what a digital speaker is. All that aside, I quit listening to speakers in audio showrooms many years ago. I learned back then that room acoustics had more effect on sound than the speakers did so listening to speakers in someone else's room didn't correlate to what I would hear at home. I've bought speakers based on measurements and specs ever since and I'm not going back. So you can agree to disagree with me as well.

What specs and measurements do you look for before purchase? Many speakers have quite similar specs/measurements but are voiced quite differently. And how do you determine which spec differences would be audible to you (and in your room) and which would not?

 

[fredk]

I think thats a very good question Tomorrow. Things like on and off axis response over the appropriate range are good starting points, but how do you evaluate speakers on measurements alone when some of the relevant measurements are not commonly avialable?

On waveguides, it seems to me from what I have read here that it is like so many other speaker technologies, if well implemented they can achieve excellent results: yet another path to audio nirvana.

 

[Nuance AH]

Not to derail this discussion, but based upon the research done by Toole & Olive and other sources cited, I am unsure as to how you feel it is subjective research or an opinion of theirs???

What the reseach tends to show is that subjective listening preferences can be recreated with a linear output device and DSP to shape the "subjective" listening curve. This was proven through the blind listening tests. Andrew and Chris are simply pointing out that fact.

Right, I understand that and agree with it. Perhaps some of my wording was confusing?

Either way, I agree with you. What I don't agree has already been listed in a previous post. That's all.

Also, I am done debating the topic because it keeps getting swayed to other topics. Back to waveguides...if you are so inclined.

 

[Tomorrow]

Here's a good look at one of the SP Technologies waveguides. I happen to think this speaker looks pretty cool. (But then I like the looks of women with freckles, too. So what do I know?! )