Home Theater Forum and Systems banner

1 - 20 of 41 Posts

·
Banned
Joined
·
4,838 Posts
Discussion Starter #1
A Response to Floyd E. Toole's AES Article "The Measurement and Calibration of Sound Reproducing Systems"

Mr. Toole has done much for the audio industry over the years, as is clear from his excellent recent article in the Journal of the Audio Engineering Society, The Measurement and Calibration of Sound Reproducing Systems. Unfortunately, the good Mr. Toole appears to have left an entire class of listeners out of his study altogether - the Soundstage Listener, for whom a natural, lifelike soundstage and precise imaging are a "holy grail" in speaker setup and calibration.

The importance of achieving an appropriate balance between the objectives of having good frequency response (FR) performance and good soundstage and imaging (SS&I) performance is hinted at in section 2.0 (p513), but is terribly under-emphasized. In fact, it is my opinion that this single factor is the thread which, when pulled at to understand its implications, potentially leads to the unravelling of Toole's thesis, that "a calibration methodology that could be applied throughout the audio industry" (abstract) is achievable.

Of course listeners prefer "the delivery of an accurate, neutral, direct sound" (P 538). However this point tends to be used in support of the rule of thumb to always point the speaker at the listener, as is typical of equilateral triangle configurations. While this approach is clearly the best way to get flat FR delivery, it rarely gives good (SS&I) performance for most loudspeakers. In fact, the best SS&I performance with most loudspeakers is achieved with a significant amount of off-axis aiming of the speakers, as much as 15° to 20° or more, usually resulting in noticeable HF rolloff at the listening position (LP).

One might ask why a listener would be willing to live with that kind of FR sacrifice in any listening situation. The answer is simple: With appropriate attention to detail in setting up speakers, SS&I performance can be achieved that is so engaging that it becomes the primary goal of the listening experience, even to the point of the listener being willing to sacrifice FR performance to achieve it.

Not having seen the precise speaker setups that Toole has used in his research, one can only make an educated guess, based upon other information and statements made, that his research and studies have all been conducted with the LP on the speaker axis.
  • "Flat on-axis frequency response is clearly the engineering objective for most of these systems." p512
  • The general discussion on pp515-517.
This is unfortunate. What is implied is that all of the research he has conducted has been under circumstances in which the listeners have been exposed to SS&I performance which, for most speaker types involved, has been less than optimal, perhaps even poor, relatively speaking. It is unfortunate because, in my experience, when listeners are given the choice between flat, even on-axis FR with mediocre-to-poor SS&I performance vs. top-notch, viscerally-engaging SS&I performance with some minor sacrifice in FR, the majority of listeners will choose the setup favoring the SS&I performance every time.

The implications relative to Toole's research are as follows:
  • The importance of using a loudspeaker that "radiates a flat, smooth, direct sound, and that has gradually changing or constant directivity" (p525) does not change. While the following point has not been researched, it is completely reasonable and likely that these same characteristics would contribute to good SS&I performance. In fact, the "gradually changing or constant directivity" characteristics would support the worthy goal of minimum change in FR when the speakers are aimed away from the LP to get better SS&I.
Beyond that, the direction one would take to calibrate a sound system potentially takes turn where SS&I are a high listening priority.
  • The placement and aiming of the speakers is undertaken with SS&I performance being the primary goal, and with no regard for high-frequency rolloff that results from off-axis aiming. This is an entirely by-ear process at this time. It is conceivable that there could be a measurement method which could lead to a good SS&I set up result, but I have yet to see it happen and have tried it numerous times. Even finding a sweet spot with good frequency response matching and good phase matching between left and right speakers in a stereo pair does not seem to be enough to ensure good SS&I performance. Further, that sweet spot that does give good SS&I performance may, when measurements are taken, oddly show less-than-the-best matching between the left and right frequency response and phase measurements, yet still display striking SS&I performance characteristics. This is all due to some relationship between direct and reflected sound paths in a given listening room that gives the hoped-for result, and at this point there is no simple formula that leads to a measurement method that can predict a good SS&I set up. There is no doubt however that several experienced listeners will agree when the goal has been achieved, as I have been among such groups that have set up speakers in that way to achieve SS&I performance day in and day out and always agreed when that sweet spot for SS&I performance had been achieved.
  • Once SS&I performance has been achieved, then one is left with the question of what to do about the roll-off in the high-frequency response.
    • One alternative is to ignore it.
    • Another is to apply automated DRC to compensate for it. As is suggest in Toole's paper, there are limits to what can be achieved without creating undesirable side effects, so less is better here. In general, the application of automated DRC to compensate for the HF variations under these conditions has been found to be beneficial.
    • The third alternative is to use simple minimum phase filtering, a single band HF shelf filter, for instance, to boost or smooth frequencies equally between the left and right speaker to achieve the flattened frequency response that is desired.
    All three of these alternatives have been applied very satisfactorily. That high-frequency roll-off can be quickly tuned out in the listeners mind, and after a few minutes of listening forgotten about altogether. This is not to say that flatter frequency response would not be preferred, but simply to say that some degree of HF roll-off is easily tolerated where the means to correct it is not readily available.
Mr. Toole is to be applauded for his goal and the marvelous research that he has undergone over the years toward achieving it. Few have the patience and focus to accomplish what he has. But to leave out of his study altogether the pickiest type of listener of all seems a major oversight. It is clear that to include them (us) complicates his work immensely. It would be folly to suggest that he has overlooked us on purpose. Yet we are not an invisible contingent - although admittedly our numbers might not be impressive, they grow steadily as the word gets out.

If it turns out that we are somehow included in that study with full consideration for our seemingly very different setup and calibration goal, then I will apologize. If such is the case, however, one would think its mention would be more easily recognized, and that it would receive more than the cursory coverage given in section 2.0 (p513).

I for one will be curiously watchful of Mr. Toole's future writings and research to see how he includes this group in his research and in the "Unified Calibration Method" he is seeking to achieve.
 

·
Registered
Joined
·
2,261 Posts
Thanks for bringing this article to our attention, Wayne, and for your keen observations. Dr. Toole is, indeed, a researcher extraordinaire. I've followed his publications and activities with great interest both before and after purchasing the original Revel Salon speakers. A few of his writings are listed here. As a scientist he naturally leans toward the objective, but recognizes that objective design alone fails to satisfy discriminating listeners (of SS&I) which have one foot on the subjective side. Purely objective design also tends to ignore human preferences (i.e. flat response is not pleasing to the ear). But how and what to measure? Dr. Toole again hints at, but doesn't elaborate on, spatial psychoacoustic SS&I phenomena in his paper Audio Science in the Service of Art:

"To design audio products, engineers need technical measurements. Historically, measurements have been viewed with varying degrees of trust. However, the value of measurements has increased dramatically as we have found better ways to collect data, and as we have learned how to interpret the data in ways that relate more directly to what we hear. With measurements we can set objectives, telling us when we are successful. Some of these design objectives are very clear, and others still need better definition. All of them need to be moderated by what is audible. Imperfections in performance need not be immeasurably small, but they should be inaudible. Achieving this requires knowledge of psychoacoustics, the relationship between what we
measure and what we hear."

Another passage taken from the same paper seems to support SS&I listeners comprise a significant portion of the listening community: "...all sound radiated by a loudspeaker, in whatever direction, eventually reaches the listener, and all of it will influence what we perceive in terms of sound quality and spatial and directional effects." (my emphasis).

I think the doctor's focus is on objective data, and on pleasing his colleagues in the AES. Trying to convince them of unmeasurable psychoacoustic phenomena is not on his agenda.
 

·
Banned
Joined
·
4,838 Posts
Discussion Starter #3
It doesn't help that there is no clearcut way to measure for successful SS&I implementation, nor a truly clearcut way to set it up - DO THIS and you will get a great soundstage.

There are those who deny it exists, but get a group of experienced listeners in a room and they will agree when it is right.
 

·
Registered
Joined
·
363 Posts
A Response to Floyd E. Toole's AES Article "The Measurement and Calibration of Sound Reproducing Systems"

(...)

Unfortunately, the good Mr. Toole appears to have left an entire class of listeners out of his study altogether - the Soundstage Listener, for whom a natural, lifelike soundstage and precise imaging are a "holy grail" in speaker setup and calibration.

(...)
In fact, the best SS&I performance with most loudspeakers is achieved with a significant amount of off-axis aiming of the speakers, as much as 15° to 20° or more, usually resulting in noticeable HF rolloff at the listening position (LP).

(...)
I totally agree with you. Since I deserted the "sacro-saint" rule of the equilateral triangle (religiously applied since 40 years in my case!):rubeyes:, a couple of weeks ago, my listening experience with my HT and my stereo system for musik is so much more satisfactory.:bigsmile:

One might ask why a listener would be willing to live with that kind of FR sacrifice in any listening situation. The answer is simple: With appropriate attention to detail in setting up speakers, SS&I performance can be achieved that is so engaging that it becomes the primary goal of the listening experience, even to the point of the listener being willing to sacrifice FR performance to achieve it.
You bet ! When you are watching a movie and you can pinpoint the trumpet or the piano of the soundtrack and hear clearly the dialogue ...what a trip !:sn:

Of course there is always place for FR improvement too.
 

·
Registered
Joined
·
71 Posts
Kudos for one of the better thoughts on sound I know I've read in a long time, Wayne. I've used a 3:4 triangle for over 30 years...

Interestingly, SS&I is terribly important, but it could be that its presence is itself an indicator of other behaviors not found in conventional wisdom or the usual vernacular, and impossible to locate in the objective data.

There are those who deny it exists, but get a group of experienced listeners in a room and they will agree when it is right.

As many of us have said - even if mostly privately - there's good hifi and then there's the undeniable illusion of real music. The difference is neither subtle nor predictable by the usual methods.
 

·
Registered
Joined
·
168 Posts
You are certainly touching the limitations of the current psychoacoustic science. I'm certain someone like Floyd Toole has his own ideas about SS&I but as a scientist he needs to gather data to prove it with statistical evidence before he can publish anything.

In my opinion the challenge in developing a clearcut way to setup and measure for optimal SS&I lies in the listener preferences (some value timbre more than spaciousness and vice-versa). Even when listener preferences were compiled, it will be difficult to achieve without previously developing recording standards since the listener preferences will vary with recording material.

I'm sure you all read that once also but IIRC the widely accepted metrics were imaging and spaciousness/envelopment and he seemed to present them as somewhat tradeoff since the former is improved by absence of reflection while the latter is provided the reflections. As you mentioned there is the FR on top of that.

About the setup, audiophiles and hi-fi shops have aimed speakers straight ahead or so for as long as I can remember whereas studio speakers followed the old triangle rule. This is in-line with their respective objectives (plus many “hifi” designed are unbearable on-axis).

Have you tried the positioning method of Earl Geddes, that is more toe-in to remove early reflection (good for timbre+imaging)? That way you are also listening off-axis and the reflections contribute to the spaciousness but being delayed they do not harm timbre+imaging as much as when the speakers are aimed straight. His approach implies constant directivity speakers like his to keep good FR off axis as well but I have found it worked well even with conventional cone+dome speakers.

@Lumen thanks for the AC link, I had all the files on PC but wanted to share it with a friend.
 

·
Plain ole user
Joined
·
11,121 Posts
Having worked with Wayne many times positioning speakers, I can tell you that he has tried such positioning, along with many others. He rules out very little as possibilities, and we have tried some pretty outrageous things. My experience is that he rarely fails to end up with what I personally would, because we seem to prioritize similar sound. I think it would be possible to isolate SS&I scientifically, but it would be a tedious matter and would involve some pretty boring listening tests.

Perhaps in the next sessions that we have we can do some blind speaker positioning tests to see if we come up with the same conclusions with respect to position and direction.
 

·
Moderator
Joined
·
1,669 Posts
Hi Wayne,

Just caught this thread. I have no idea how what you said has anything to do with the Toole paper, which is mainly about EQ/so called "room correction" products. What am I missing here?

I would be happy if we could do some blind tests that simply showed that we all agreed when the soundstage and imaging were at/near optimum.
You might as well do a blind test for optimum tasting beer.:)
"Sounstage and imaging" is going to be a preference test, no more, no less.
Hopefully you grasp that the S&I begins for most part (rare exceptions of true live multi-mic recording), with a wholly artificial construct in a production studio, based on ILD and ITD developed from the days of Blumlein. I really wish most audiophiles could go to a studio to see how things are "constructed" spatially, but unfortunately, 99.999% do not.
So now we have our electrical construct media that we must now reconstruct to our pleasure. That involves those particular speakers/3D polar response, that particular room, the recording...and that particular listeners personal preference. There is no "optimum", other than the listeners preference, for what they believe is optimum.
IOW, things Toole wisely knows comes down to things well beyond just the loudspeakers....and outside his scope.:)

Btw, the top Revel does not have extremely wide/flat horizontal dispersion by accident.

That particular feature is there for reasons...one being very much related to S&I. I really think Harmans research (based on NRC/Toole/Olive et al) is more misunderstood than understood to the casual eye.

cheers,
 

·
Banned
Joined
·
4,838 Posts
Discussion Starter #10
Hi Wayne,

Just caught this thread. I have no idea how what you said has anything to do with the Toole paper, which is mainly about EQ/so called "room correction" products. What am I missing here?
All due respect, I pretty much predicted your response.

I will (re)summarize:
  • Toole is "seeking a calibration methodology that could be applied throughout the audio industry." (abstract)
  • Toole's research assumes (this is easily inferred from the text of his paper) that an on-axis listening position (LP) would be universally preferred.
  • On-axis LP usually gives poor SS&I. This depends upon the design of the speaker and the upon the room.
  • Setup, measurement, and calibration methods are different for the audience that prioritizes SS&I highly. There is an entire listening audience that has been ignored in his research. His "universal calibration methodology" is almost certain to overlook most of the more particular listeners on the planet.
You might as well do a blind test for optimum tasting beer.:)

"Sounstage and imaging" is going to be a preference test, no more, no less.
I completely disagree. My experience has shown that the difference between good SS&I vs. poor SS&I is quite recognizable by the general listener - once he has been exposed to both and given the opportunity to experience a truly top-quality soundstage with sharp imaging, and most listeners have not had that opportunity.

Hopefully you grasp that the S&I begins for most part (rare exceptions of true live multi-mic recording), with a wholly artificial construct in a production studio, based on ILD and ITD developed from the days of Blumlein. I really wish most audiophiles could go to a studio to see how things are "constructed" spatially, but unfortunately, 99.999% do not.
I grasp it quite clearly, I have worked in multitrack studios since my teens, I have recorded, mixed, mastered, and am sitting in my personal recording studio as I write.

So now we have our electrical construct media that we must now reconstruct to our pleasure. That involves those particular speakers/3D polar response, that particular room, the recording...and that particular listeners personal preference. There is no "optimum", other than the listeners preference, for what they believe is optimum.
IOW, things Toole wisely knows comes down to things well beyond just the loudspeakers....and outside his scope.:)
Toole has gone to great lengths to "prove" that listeners prefer "the delivery of an accurate, neutral, direct sound" (P 538). I am simply stating that there is another class of preference which most of those listeners have not even been exposed to, and that it makes a difference to the conclusions of the research they have done.

Btw, the top Revel does not have extremely wide/flat horizontal dispersion by accident.

That particular feature is there for reasons...one being very much related to S&I. I really think Harmans research (based on NRC/Toole/Olive et al) is more misunderstood than understood to the casual eye.
And perhaps to the experienced as well.

There is no doubt their work holds great value. But it is not the be-all end-all that many claim it to be. Listener preference is what their work is ALL ABOUT! To claim to define listener preference while ignoring a whole area of listener preference that is closely related is a big miss of the listener preference mark, in my opinion.

And I totally agree that the speaker design characteristics you mention are desirable.

If I might, I will simply state that SS&I, while more difficult to measure via instrumentation than frequency response, is not difficult for a listener to recognize. I have seen it time and again, heard reactions, gotten positive feedback to setup suggestions - when the SS&I experience is achieved, it is almost universally applauded. And the same characteristics are almost universally preferred. That has been my experience. Has it been researched with double-blind study? No. I have not had the opportunity. But Toole and all have. And they have apparently chosen to overlook it.

A.J. - I appreciate your response, and while your views are clearly as valid as mine, we are not in agreement on a number of key points. Rather than have this turn into a sparing match, I suggest we agree to disagree and move on. You are, of course, more than free to open a thread to express your own views.
 

·
Moderator
Joined
·
1,669 Posts
Toole is "seeking a calibration methodology that could be applied throughout the audio industry." (abstract)
Right, spectrally, recordings are all over the place. Too much bass, too little, bright, dull, etc. I think we've all experienced that.

Toole's research assumes (this is easily inferred from the text of his paper) that an on-axis listening position (LP) would be universally preferred.
Regarding spatial reproduction specifically? Please quote that directly, for no ambiguity, thanks.

On-axis LP usually gives poor SS&I.
Supporting evidence? I'm an AES member and can access any paper. I can't recall any like that.

This depends upon the design of the speaker and the upon the room.
If you mean what affects SS&I, then of course. We agree. How does Toole diverge here?

Setup, measurement, and calibration methods are different for the audience that prioritizes SS&I highly. There is an entire listening audience that has been ignored in his research. His "universal calibration methodology" is almost certain to overlook most of the more particular listeners on the planet.
I have no clue how you have concluded this from any Toole paper. Direct quotes would be helpful.
As someone who designs speakers, specifically to vary spatial reproduction aspects, I see absolutely nothing in Tooles work which I've followed for 20+ years, to warrant any such assertion.

I completely disagree. My experience has shown that the difference between good SS&I vs. poor SS&I is quite recognizable by the general listener...
That is a preference!:)
Just like Toole has found most prefer smooth rather than ragged on axis FR, off axis FR, smoother amplitude bass, etc, etc.. Prefer. What are you disagreeing with??:huh:

I grasp it quite clearly, I have worked in multitrack studios since my teens, I have recorded, mixed, mastered, and am sitting in my personal recording studio as I write.
Great, so you know that there is no standard, for placing the drummer back left 2', 3' or 4' in the SS, yes? So when you playback/listen to a recording you didn't do, is the drummer supposed to be 2', 3', or 4' back? Which is "better"? Or is it just what you prefer?

Toole has gone to great lengths to "prove" that listeners prefer "the delivery of an accurate, neutral, direct sound" (P 538). I am simply stating that there is another class of preference which most of those listeners have not even been exposed to, and that it makes a difference to the conclusions of the research they have done.
If you are taking exception to them now preference testing only in mono (vs stereo), especially your favorite MLs, then you could simply have said so. I agree! I have raised this issue (and others) with Sean Olive directly. But what that has to do with the Toole EQ paper you seem to be responding to, I have no idea.

I think we may agree far more than you think, but I just don't see the connection to the Toole paper referenced.
See you at the next party.:)

cheers
 

·
Banned
Joined
·
4,838 Posts
Discussion Starter #12
Right, spectrally, recordings are all over the place. Too much bass, too little, bright, dull, etc. I think we've all experienced that.

Regarding spatial reproduction specifically? Please quote that directly, for no ambiguity, thanks.
"Flat on-axis frequency response is clearly the engineering
objective for most of these systems. Those that deviate
significantly earn lower ratings in double-blind subjective
evaluations. Although there is more to be considered, a
flat direct sound delivered to listeners is the basis for most
reproduced sound."
p512

Supporting evidence? I'm an AES member and can access any paper. I can't recall any like that.
Direct experience with hundreds of speakers. The exception is the horn-loaded mid/tweeter with high directivity.

If you mean what affects SS&I, then of course. We agree. How does Toole diverge here?
He does not.

I have no clue how you have concluded this from any Toole paper. Direct quotes would be helpful.
As someone who designs speakers, specifically to vary spatial reproduction aspects, I see absolutely nothing in Tooles work which I've followed for 20+ years, to warrant any such assertion.
Already stated as clearly as I know how. Nowhere does he mention off-axis listening angles. Nor does he mention soundstage/imaging more than in passing. The two are connected. If they were considered important in his research, it seems he would say so. If you have seen him say so, please show me where.

That is a preference!:)
Just like Toole has found most prefer smooth rather than ragged on axis FR, off axis FR, smoother amplitude bass, etc, etc.. Prefer. What are you disagreeing with??:huh:
Argumentative.

Great, so you know that there is no standard, for placing the drummer back left 2', 3' or 4' in the SS, yes? So when you playback/listen to a recording you didn't do, is the drummer supposed to be 2', 3', or 4' back? Which is "better"? Or is it just what you prefer?
Very argumentative.

If you are taking exception to them now preference testing only in mono (vs stereo), especially your favorite MLs, then you could simply have said so. I agree! I have raised this issue (and others) with Sean Olive directly. But what that has to do with the Toole EQ paper you seem to be responding to, I have no idea.
""
 

·
Moderator
Joined
·
1,669 Posts
"Flat on-axis frequency response is clearly the engineering
objective for most of these systems. Those that deviate
significantly earn lower ratings in double-blind subjective
evaluations. Although there is more to be considered, a
flat direct sound delivered to listeners is the basis for most
reproduced sound
."
He's stating the obvious that has been confirmed by others as well. If the system is spectrally chaotic, starting in the studio, exactly how can there be any standardization spectrally?
Are you arguing for SS&I "standards" that Toole is somehow neglecting?? When you just said it will depend on a host of variables, room, speaker directivity...and preferences?

Direct experience with hundreds of speakers. The exception is the horn-loaded mid/tweeter with high directivity.
That is very different from Toole et al work and unrelated to the limitations of steady state measurements and so called "room correction"/EQ, the topic of the paper. I think your criticism of this paper is misdirected and should be aimed more at their loudspeaker ranking method, using mono, which would put you on firm ground imho and in agreement with myself and others.

Already stated as clearly as I know how. Nowhere does he mention off-axis listening angles. Nor does he mention soundstage/imaging more than in passing. The two are connected. If they were considered important in his research, it seems he would say so. If you have seen him say so, please show me where.
2 SOUND FIELDS IN ROOMS

Sounds arriving at a listening location include charac-
teristics that are traceable to the sound source (e.g., spec-
trum/amplitude response and directivity) and to the room
through which the sound is communicated (the type and
location of acoustical materials and the effects they have
on the reflected sound field). Listeners identify aspects of
sound quality (timbre), and also, because of binaural hear-
ing, spatial attributes: localization, imaging, envelopment,
etc. Some evidence suggests that the timbral and spatial
perceptual dimensions may be comparable in importance
,
possibly not completely separable (discussed in [1] sections
8.2.1 and 20.1). Sound cues contributing to spatial percep-
tions are not revealed in steady-state amplitude-response
measurements and the lack of directional discrimination
makes time-windowed measurements ambiguous. There-
fore, it is necessary to understand the spectral and direc-
tional properties of the sound sources as well as the sound
reflecting behavior of listening venues in order to anticipate
acoustical and perceptual events
[2].
What exactly do you want Toole et al to test/standardize for, with what speakers, in what room, to determine what you will prefer with all your senses, in your room, with your preferences?

cheers,
 

·
Plain ole user
Joined
·
11,121 Posts
AJ, I think you miss the point. When Wayne is optimizing SS&I it is not some attempt to find a magical state where the "performance" is most perfectly trasmitted to onesperception. It is simply an attempt to find the placement that gives the overall best impression of creating a sound space in which performers are playing music. Certainly it is not a characteristic of the recording alone, nor is it a characteristic of any single component. In a given room with a given system, however, one tries to achieve the best possible experience of the music, which for many of us is improved when a sense of space is perceived.

To suggest that testing of perception is pointless is a revelation of the very limited perspective from which you operate. Wayne, and the rest of us, have always been willing to challenge our assumptions about what we experience. We are quite compfortable negotiating the space between the objective and the subjective. Toole and others have avoided that space by lumping all perception into a notion of preference. What Wayne has attempted to do is to take one aspect of the listening experience and focus on optimizing it to maximize his enjoyment, because it is the part of his own listening preference that appears to be the biggest variable. I tend to agree to a large extent. It is likely that Toole's conclusions about off axis response align with Wayne's experience and views on the matter. The difference is that he never really addressed the matter, that I know of.

One of the most important things about the way we interact here at HTS is that we truly read for understanding, trying to get what others are trying to say, rather than trying to pick apart the words of others to advance our own perspective. You don't have to agree, in fact the very reason that you and Wayne are on staff is that you bring a differenct and valuable perspective to the table, but you sometimes come very close to ridicule, which is not helpful. Please keep that in mind. Challenge, always. Belittle, never.
 

·
Premium Member
Joined
·
1,562 Posts
RE the subject paper:
With the discussion here I was motivated to revisit the paper to try to clarify my understanding. The conclusions were not stated in a way that I found easy to follow. Toole indicates there are other issues and more work to consider. That is why a more definitive finding was not offered. Nevertheless, I tried to pull out and simplify the main points that were made to see if they help this hobbyist's understanding. Since I assembled this for me, I decided to share it here. Feel free to correct me. I hope didn't overlook something important.

Regarding theaters and dubbing stages:
> Flat direct SPL response of the speaker is necessary.
> Smooth controlled speaker directivity is necessary.
> It is not possible to infer a flat direct sound response based on the steady-state room response.
> The LF below approximately 300Hz needs to be EQed to smooth the SPL irregularities due to room modes.
> No other EQ is necessary. [Possibly there is room for gentle EQ shaping to the preferred house curve.]
> The steady-state SPL response will fall several dB starting from 50Hz to ~2kHz as a function of room reflectivity and speaker directivity.
> Per Fig 14, trained listeners prefer a steady-state house curve which slopes down ~9dB from about 50Hz to 13kHz and then falls a little faster to 20kHz.
> The X-Curve is flat from 50-2k Hz and then drops about 9dB from 2kHz to ~13kHz
> The X-Curve therefore differs significantly from the preferred curve.
> There is no need for adjustments to a house curve due to venue size.
> More work is needed before any specific changes to the X-curve or EQ practices can be recommended.

I was not able to pullout much of interest specifically regarding home theater and music rooms except a hint that it may be possible to establish a unified standard that works for all. Wouldn't that be nice!

This doesn't help much to those of us who are looking for a deeper understanding to improve our personal home listening experience. It does reinforce the directional advice that is often given and cites significant data and detail to help with the understanding of it.
 

·
Moderator
Joined
·
1,669 Posts
AJ, I think you miss the point. When Wayne is optimizing SS&I it is not some attempt to find a magical state where the "performance" is most perfectly trasmitted to onesperception. It is simply an attempt to find the placement that gives the overall best impression of creating a sound space in which performers are playing music. Certainly it is not a characteristic of the recording alone, nor is it a characteristic of any single component. In a given room with a given system, however, one tries to achieve the best possible experience of the music, which for many of us is improved when a sense of space is perceived.
Leonard, I haven't missed that the Toole paper is about room/EQ measurements. How is Waynes spatial desires a "response" to a paper about EQ/measurements?
If he had wrote a response to Harmans specific method of ranking speakers via mono tests, then I could see his point...and agree.

When Wayne is optimizing SS&I it is not some attempt to find a magical state where the "performance" is most perfectly trasmitted to onesperception. It is simply an attempt to find the placement that gives the overall best impression of creating a sound space in which performers are playing music. Certainly it is not a characteristic of the recording alone, nor is it a characteristic of any single component. In a given room with a given system, however, one tries to achieve the best possible experience of the music, which for many of us is improved when a sense of space is perceived.
Right. Where have I disagreed?

To suggest that testing of perception is pointless is a revelation of the very limited perspective from which you operate.
That's a strange accusation given that I operate based nearly entirely around perceptual testing, of the scientific variety, not ad hoc purely personal experience.

Wayne, and the rest of us, have always been willing to challenge our assumptions about what we experience. We are quite compfortable negotiating the space between the objective and the subjective. Toole and others have avoided that space by lumping all perception into a notion of preference. What Wayne has attempted to do is to take one aspect of the listening experience and focus on optimizing it to maximize his enjoyment, because it is the part of his own listening preference that appears to be the biggest variable. I tend to agree to a large extent. It is likely that Toole's conclusions about off axis response align with Wayne's experience and views on the matter. The difference is that he never really addressed the matter, that I know of.
Toole may or may not, but it is certainly not the focus of this paper. Hence my involvement in the thread, having read the paper. I've now stated several times that had Waynes criticism been of other aspects of Tooles work, we would probably be in agreement.

One of the most important things about the way we interact here at HTS is that we truly read for understanding, trying to get what others are trying to say, rather than trying to pick apart the words of others to advance our own perspective. You don't have to agree, in fact the very reason that you and Wayne are on staff is that you bring a differenct and valuable perspective to the table, but you sometimes come very close to ridicule, which is not helpful. Please keep that in mind. Challenge, always. Belittle, never.
I know Wayne well enough to pick up the phone and call him to clarify any position and I certainly wouldn't belittle him. But I will defend the scientific method that I try to operate under, as long as I can post here.

cheers,
 

·
Moderator
Joined
·
1,669 Posts
Regarding theaters and dubbing stages:
> Flat direct SPL response of the speaker is necessary.
> Smooth controlled speaker directivity is necessary.
> It is not possible to infer a flat direct sound response based on the steady-state room response.
> The LF below approximately 300Hz needs to be EQed to smooth the SPL irregularities due to room modes.
> No other EQ is necessary. [Possibly there is room for gentle EQ shaping to the preferred house curve.]
> The steady-state SPL response will fall several dB starting from 50Hz to ~2kHz as a function of room reflectivity and speaker directivity.
> Per Fig 14, trained listeners prefer a steady-state house curve which slopes down ~9dB from about 50Hz to 13kHz and then falls a little faster to 20kHz.
> The X-Curve is flat from 50-2k Hz and then drops about 9dB from 2kHz to ~13kHz
> The X-Curve therefore differs significantly from the preferred curve.
> There is no need for adjustments to a house curve due to venue size.
> More work is needed before any specific changes to the X-curve or EQ practices can be recommended.

I was not able to pullout much of interest specifically regarding home theater and music rooms except a hint that it may be possible to establish a unified standard that works for all. Wouldn't that be nice!

This doesn't help much to those of us who are looking for a deeper understanding to improve our personal home listening experience. It does reinforce the directional advice that is often given and cites significant data and detail to help with the understanding of it.
Not much to disagree with there. Toole is simply pointing out the complete lack of standards in the "circle" if you will, beginning at the recording stage. He also points out the limitations of some of the so called "room correction" products being used by (Home) consumers, who are often trying to "fix" erratic 3d acoustic sources with 2d electronic "correction".
No amount of speaker positioning for spatial reproduction preferences, is going to correct anomalies in the onset or direct response of the system, simply based on the way human hearing works.

cheers
 

·
Banned
Joined
·
4,838 Posts
Discussion Starter #18
AJ,

I appreciate your initial questions because they forced me to think through my remarks and make sure they were presented as clearly as I could present them. I enjoy a good open discussion, but unfortunately we have somehow reached a point of pointless arguing. All due respect, your latest round of comments has pretty much missed my meaning altogether. I enjoy a good discussion, but I think we're at the point of pointless arguing, & I choose not to engage. Thanks again for your initial remarks and questions.
 

·
Moderator
Joined
·
1,669 Posts
No problem Wayne. When either you or Leonard get a chance, read starting from here to see how much we actually diverge on your point of contention, rather than this Toole EQ/measurements paper.

cheers
 

·
Banned
Joined
·
4,838 Posts
Discussion Starter #20
jtalden: Thank you for an excellent summary. I thought I read a little bit more into the part about room EQ:

  • If desired objectives are not achieved because of poorly controlled speaker directivity, get better speakers. "Equalization cannot change loudspeaker directivity; the remedy is a better loudspeaker."(p516)
  • Room EQ is of limited value, is generally useless in correcting for speakers with poorly controlled directivity. "Equalization is very limited in what it can “correct,” yet the notion that changing the signal supplied to a sound system consisting of an unknown loudspeaker in an unknown room can “equalize” or “calibrate” a system is widespread."(p517)
Of course there is no argument with either of these points in theory, they are well proven and completely logical. In practical terms, many a listener with imperfect speakers in his home listening environment and on a limited budget has resorted to room correction to try to get some level of improvement in performance, sometimes successfully, sometimes not. I would suggest that a better way to put these points for such a listener would be:
  • Do not correct too much. Less is usually better.
  • Do not expect too much in terms of improvement. There is only so much that can be accomplished and negative side effects are quite possible. Room correction might actually give a flatter measured frequency response that sounds worse.
  • Consider room treatment, and possibly better speakers, rather than sinking a lot of money into room correction.
  • If you must perform room correction, do some room treatment first. If done correctly, negative effects the speaker's directivity characteristics will be minimized and room correction results will be more effective.
I have no disagreement with Toole on these points as he has stated them. My only reason for the "softened" stance is that well-meaning givers of advice can be extreme in stating that "room correction is a complete waste, Toole says don't waste your time," (paraphrasing from memory, of course). The world is seldom so black and white, it seems appropriate to add a few gray tones to that particular picture.
 
1 - 20 of 41 Posts
Top