Can Objective Loudspeaker Measurements Predict Subjective Preferences?

[DukeL]

I hate to come across as an immature fanboy, but wow. Look Ma, I'm talkin' with Floyd Toole!!

Because of the extreme simplicity of stereo (a directionally and spatially deprived format) there are some kinds of recordings that appear to benefit from a bit of added spaciousness. An instrument hard panned L or R is a mono signal, emerging from a tiny point in space - not realistic. We need more channels and loudspeakers. But don't hold your breath. The movie folks figured this out decades ago, but not the music industry.

I agree that there's room for improvement over normal stereo.

Shortly before your book came out, in 2007, I built my first "controlled-pattern bipole" loudspeaker, which used fairly narrow-patterned (Geddes-esque) arrays firing both forward and backward. The idea was to get a relatively reflection-free first arrival sound, followed by a generous helping of spectrally-correct reverberant energy after a decent time-delay from bouncing off the front wall, in a crude first approximation of the acoustic goings-on in a small music venue. Not omni... more like dividing up a wide hemispherical pattern into two identical 90-degree patterns and placing them back-to-back, but bipolar instead of dipolar so we can still have bass + decent efficiency. It wasn't the same as adding more channels, but offered improvement (in both timbre and spaciousness) with acceptable tradeoffs given adequate placement, without the exaggerated image width of omnis, in my opinion. And yes I know what they say about opinions (and I have more than one!), but I'll never have direct access to an anechoic chamber or a true double-blind panel of trained listeners.

So anyway when I read your book, among other things I was struck by the information on what the ideal arrival angles are for reflected energy. The fact that 60 degrees off the centerline (30 degrees forward of side-to-side) was ideal really caught me by surprise. So I switched to a different geometry, twisting my bipole so that the second array of drivers fired at the side wall, to get more optimum arrival angles (ballpark about 60 degrees to the left and right of the centerline for a stereo pair). When room dimensions allowed, this offered a worthwhile improvement in my opinion, but it required a wide room or else the imaging was degraded too much, presumably due to insufficient path-length-induced time delay.

Later a friend of mine, James Romeyn, while experimenting with multiple "bookshelf" speakers in various configurations, tried firing the additional drivers up at the ceiling from down near the floor. This seems to work better in most rooms, presumably because the longer time-delay trumped the more ideal arrival direction. Again this isn't a separate channel (a la Atmos), but it's interesting to switch the "late ceiling splash" arrays on and off (while correcting the loudness) to see what the effects are.

The ideal is probably the "twisted bipole" with a long time delay for that 60-degree bounce energy, and maybe I'll try implementing that someday using dedicated speakers with delays and separate amplification. Have you ever tried that, or something similar?

I realize that none of what I describe above was explicitly encouraged by you or your book, but you gave me that little knowledge which then become a dangerous thing. And I have enjoyed the ensuing ride more than you can imagine. Thus far the market hasn't told me that I've built a better mousetrap, but I've sure had fun trying!

Thank you sir.

Duke

 

[Floyd Toole]

Hi Jon,

There are a few reasons why a small speaker might struggle at high sound levels. We like bass, so very often a small speaker is played louder simply to get some impression of bass. In many cases this can lead to amplifier overload, protection, or clipping, all of which sound pretty bad. Most small speakers have low sensitivity (they need lots of power) and low impedances (so that they draw more current for a given voltage input) making it more likely that power amps, especially those in receivers, will complain.

In the speakers themselves we have:
Power compression - the change of frequency response because of voice-coil heating is an obvious issue. The extent to which this is audible depends obviously on the long term (power) spectrum of the music being listened to, so a standardized measurement is always not completely reliable, but such measurements are routinely done at the design stage.
Distortion is another. In small speakers this most often happens because the small woofer is exceeding its linear displacement range. Add a subwoofer, high-pass filter the small speaker (bass management) and turn up the volume.
Both of these factors depend greatly on the quality of the transducers and overall engineering design - sometimes, not always, this is what one pays for. If you want loud, buy a big speaker, or one or more subs.

 

[Floyd Toole]

Duke,
I'm glad I inspired you to experiment. I've done a fair bit of it over the years and it is a great way to learn. AR, many years ago, had a loudspeaker with side-firing drivers to provide varying amounts of side wall reflections. It never caught on - don't know why - probably expensive. Early in my career I designed a listening room for a serious listener and after discussions we decided to install heavy velour drapes along the front side walls, on tracks that would allow them to be pushed back for reflective side walls or pulled forward for somewhat absorptive side walls. As he lived with it he found himself preferring one or the other "setting" for different kinds of recordings - reflecting for classical, jazz and some pop, and absorbing for "in your face" rock recordings. He had good loudspeakers, wide, spectrally uniform, dispersion so the wall reflections sounded good.

Have fun, Floyd

 

[jonnythan]

Hi Jon,

There are a few reasons why a small speaker might struggle at high sound levels. We like bass, so very often a small speaker is played louder simply to get some impression of bass. In many cases this can lead to amplifier overload, protection, or clipping, all of which sound pretty bad. Most small speakers have low sensitivity (they need lots of power) and low impedances (so that they draw more current for a given voltage input) making it more likely that power amps, especially those in receivers, will complain.

In the speakers themselves we have:
Power compression - the change of frequency response because of voice-coil heating is an obvious issue. The extent to which this is audible depends obviously on the long term (power) spectrum of the music being listened to, so a standardized measurement is always not completely reliable, but such measurements are routinely done at the design stage.
Distortion is another. In small speakers this most often happens because the small woofer is exceeding its linear displacement range. Add a subwoofer, high-pass filter the small speaker (bass management) and turn up the volume.
Both of these factors depend greatly on the quality of the transducers and overall engineering design - sometimes, not always, this is what one pays for. If you want loud, buy a big speaker, or one or more subs.

Thank you for the response. I understand (most of) what you said. It seems that the review industry has never really used measurements to quantify that. We see measurements of distortion at higher volumes, but not compression. I'm obviously not an expert on this, but I'm imagining a speaker that's playing some dialog with a max of say 85 dB, which (in our hypothetical example) is right where it should be with regard to reference on the source material. Now an explosion happens and we should brief bursts of 105 dB. A small sealed 2-way speaker probably won't be able to recreate that, even with distortion. It'll generate say 100 dB instead (I'm clearly just making numbers up). This is one of the important things that I feel is lacking when reviewing speakers.

To give a bit of a real world example, I had a set of NHT SuperZeros in my home theater for a while. They sounded absolutely divine at moderate levels, but when things got heated on screen I could feel them compressing the top end, unable to reproduce the SPL the way my old Polks could.

When talking about home theater, this seems to be what people are talking about when they mention dynamics. Some Klipsch THX Ultra2 horn-loaded speakers will produce those SPLs effortlessly, while my NHTs won't.

Is there not a simple way to measure this? Am I just misunderstanding everything entirely?

 

[DukeL]

...Some Klipsch THX Ultra2 horn-loaded speakers will produce those SPLs effortlessly, while my NHTs won't.

Is there not a simple way to measure this? Am I just misunderstanding everything entirely?

Not that all manufacturers measure with the same yardstick, but the NHT's 87 dB sensitivity, 8 ohm impedance, and 75 watt power handling implies ballpark 106 dB capability.

The smaller of the two Klipsch THX Ultra 2 LRC speakers is rated at 94 dB/2.83 volts, 8 ohms compatible, and 100 watts continuous/400 watts peak. This implies ballpark 114 dB continuous and 120 dB peak.

So based on your observations, a speaker with a calculated 106 dB max output falls short, and a speaker with a 120 dB calculated max output doesn't. Actual measurements would give us more precise information, and would be useful to differentiate between speakers with similar rated performance. All else being equal, I'd put my money on the box with the biggest voice coils because they should have greater thermal capacity.

It is not unusual for a woofer to compress by 3 dB or more at its rated continuous input power, so it's quite likely that neither speaker in this example delivers its calculated long-term max SPL in the real world, but the specs still give us a basis for comparison unless we have reason to believe they are exaggerated.

I've made a few simplifying assumptions here, so the above is not gospel, but I think the trends are valid.

 

[Floyd Toole]

There is more to this than just getting loud. There is getting loud without changing the frequency response - the sound quality - which happens routinely when different drivers heat up at different rates. Small cone and dome speakers will have the most difficulty in this department, but, for a price, there are drivers that do amazingly well. Horns have no problem getting loud, but getting good sound from them has been a historical challenge. New designs, with short throats and sophisticated flares are different animals, and the best of these can hold their own with "high-end" cones and domes. The difference is that they can get very loud without seeming to be loud.

The 105 dB target is a cinema sound level target, and the vast majority of people find it too loud for home theaters - some find it too loud in the cinemas, especially with the current tendency to compress the dynamic range and everything tends to be too loud for too long. When it is for dramatic purpose, like an explosion, that is fine, but some films are loud for no good reason too much of the time. Directors think is creates "excitement". The industry is now grappling with a problem that customers are walking out of movies because they are so loud. Law suits are beginning to appear and legal limits have been imposed in some places. When a movie is turned down one of the first things to suffer is speech intelligibility - without dialog a movie is a waste of time. It will be interesting to see where this goes.

I teach nine hours of classes on this topic at CEDIA and every time I ask the class what they find people of playing their movies at - it is typically much lower than calibrated 0 dB. -3 dB is half power, -10 dB is 1/10 power, so the amps and speakers don't have to break a sweat.

On this same topic, did you know that middle and high-frequency sounds above about 80 dB have the potential of causing irreversible hearing loss? Scary, eh? Fortunately loud bass is not a problem.

Floyd

 

[DukeL]

AR, many years ago, had a loudspeaker with side-firing drivers to provide varying amounts of side wall reflections. It never caught on - don't know why - probably expensive.

The AR "Magic" speaker! Designed by Ken Kantor as I recall, seemed like a brilliant design to me.

ESP uses side-firing dome tweeters in some of their models.

Unless I'm missing something, it seems to me that the Harmon speaker-shuffler room would have handicapped a design like the AR Magic speaker (or dipole, or bipole) which relied in part on a reflective surface.

Early in my career I designed a listening room for a serious listener and after discussions we decided to install heavy velour drapes along the front side walls, on tracks that would allow them to be pushed back for reflective side walls or pulled forward for somewhat absorptive side walls. As he lived with it he found himself preferring one or the other "setting" for different kinds of recordings - reflecting for classical, jazz and some pop, and absorbing for "in your face" rock recordings. He had good loudspeakers, wide, spectrally uniform, dispersion so the wall reflections sounded good.

Brilliant feature, who else would have even thought of that?? Interesting that a hard-core audiophile would prefer significantly different room reflection characteristics for different types of music. Even though your sample size here is small, I think the point is valid, and am mulling over the implications.

There is more to this than just getting loud. There is getting loud without changing the frequency response - the sound quality - which happens routinely when different drivers heat up at different rates.

I've seen power compression specs for prosound woofers, like the JBL 2226, based on the driver being run at the indicated power continuously for some length of time, such that presumably magnet heating effects are in play.

Can rapid voice coil heating (before the magnet has a chance to heat up) cause audibly significant compression on short-term peaks, thereby possibly changing the frequency response of the system on loud peaks even if the average input power remains fairly low? I recall reading a Stereophile article where the idea of voice coil heating being of audible significance in and of itself was tested and discredited, but presumably more rigorous testing has been done on the prosound side.

New [horn] designs, with short throats and sophisticated flares are different animals, and the best of these can hold their own with "high-end" cones and domes.

In your opinion, are there still areas where the JBL M2 may be audibly and significantly inferior to its cone-and-dome competition? My understanding is that most of your work has been on the home audio side with cone-and-dome designs, yet (by the yardsticks your research has validated) the prosound M2 seems to beat the best of the cone-and-domes at their own game... well, unless wife acceptance factor is considered part of that game.

Thanks,

Duke

 

[Floyd Toole]

Hi Duke,

I agree the AR Magic speaker was a smart idea. I would have thought it would have had follow-ons, especially now that the necessary electronics are inexpensive and more capable. The flexibility of the Harman room is that it can be programmed to shuffle speakers where we want. If a speaker is specifically designed as the "magic" speaker was, it would be shuffled into a stereo L or R location closer to a wall.

The reproduction of stereo has been seriously handicapped by a segment of the industry that believes that "sound stage and imaging" are dominated by the loudspeaker, or the wire, or the spikes, or the power cord . . . They assume that all the information is in the recordings, waiting to be liberated by the right tweak. Just as the timbre of recordings is determined by the highly variable monitor speakers being used, and the room they are in, so is presentation of localizable images and spaciousness. These are determined by mics, mic positioning, and greatly by the sometimes massive signal processing done while mixing - listing to loudspeakers in a control room, or nowadays, possibly a converted bedroom or garage.

My late friend John Eargle - to whom I dedicated my book - was a remarkable combination of musician, recording engineer, acoustical engineer, loudspeaker designer, writer . . . He once told me that in some of his classical recordings, when a solo instrument needed a bit of highlighting, he would put a mic on it and in the mix, add in the appropriate electronically generated reflections and reverb to make it fit into the context of the rest of the orchestra. It was so tastefully done that even knowing it did not allow it to be detected. The harpsichord, or whatever, just was more clearly audible. Of course, this was done in the recording control room, and was not at all a "natural" acoustic phenomenon. He would chuckle, knowing that if some purist audiophiles knew this they would be revolted.

Just as tone controls are truly needed to correct for spectral imbalances in many recordings, adjustable speakers or room acoustics can be useful to create the sound stage and imaging in the playback of recordings that listeners favor. To assume that all recordings can be treated equally is naive.

The audibility of power compression in its many variations probably could use some more research to define what is audible and what is tolerable. The magnet heating that you describe is important in pro audio sound reinforcement systems where the loudspeakers are required to work at or close to their design limits for long periods. Such heating and cooling has a very long time constant. This is not the case in most home systems. Although the modification of motor strength through magnet heating is a factor, most of the audible effects are from voice coil heating, which has a much shorter time constant. I just saw a test of a high-end audiophile speaker that in going from an average level of 70 dB (loud conversation, background music) to 90 dB (a moderate crescendo, or foreground rock listening) lost about 4 dB in output over about 3 octaves in the mid-high-frequency range. It became a different loudspeaker at different listening levels.

There appears to be nothing about the sound of the M2 that is attributable to it having a horn. It sounds as it measures - very neutral. The best cone and dome loudspeakers are likewise very neutral. Neutral is good, because it means that coloring resonances are not present. However, such excellence does not mean that all recordings will sound good - you will hear the problems as well as the virtues. If a particular recording does not sound right, try the tone controls - I do, and it often works. The notion that tone controls are abhorrent is another "head in the sand" idea.

So, where does this leave us? For sound stage and imaging, maybe we want adjustable loudspeakers or room acoustics. For accurate sound quality, maybe we want tone controls. If we are fussy.

Floyd

 

[gene]

There appears to be nothing about the sound of the M2 that is attributable to it having a horn. It sounds as it measures - very neutral. The best cone and dome loudspeakers are likewise very neutral. Neutral is good, because it means that coloring resonances are not present. However, such excellence does not mean that all recordings will sound good - you will hear the problems as well as the virtues. If a particular recording does not sound right, try the tone controls - I do, and it often works. The notion that tone controls are abhorrent is another "head in the sand" idea.

Floyd, I'm hoping Harman demos the M2s at CEDIA via two-channel so I can finally take a close listen.

Why you still messing with antiquated tone controls? We have Automatic room EQ and I hear it's wonderful

 

[DukeL]

Although the modification of motor strength through magnet heating is a factor, most of the audible effects are from voice coil heating, which has a much shorter time constant.

Thank you once again for taking the time to reply. I have suspected what you are saying here, but hadn't seen it spelled out in so many words, so I didn't know for sure that voice coil heating effects were audibly significant apart from their effect on magnet strength. But it seemed to me that voice coils would heat up virtually instantaneously (a 100 watt peak would be even more efficient at heating it up than touching it with a 100-watt soldering iron). Would it make sense to call these short-time-constant effects "thermal modulation" to distinguish them from magnet-strength-related "thermal compression"?

I just saw a test of a high-end audiophile speaker that in going from an average level of 70 dB (loud conversation, background music) to 90 dB (a moderate crescendo, or foreground rock listening) lost about 4 dB in output over about 3 octaves in the mid-high-frequency range.

Fascinating!! How long at 90 dB before the compression showed up? Was it fast enough to blunt what should have been a 20 dB peak when listening at an average of level of 70 dB?

There appears to be nothing about the sound of the M2 that is attributable to it having a horn. It sounds as it measures - very neutral. The best cone and dome loudspeakers are likewise very neutral.

If both are similarly neutral, seems to me the M2 is significantly less likely to compress peaks or change timbre as the level changes, so I'd have to be convinced that a cone and dome system is as good as the M2, rather than the other way around.

I have a question about your experience in correlating subjective preference with measured performance. Did you ever run into a situation where either a) the speaker measures quite well by your proven criteria, but scores significantly lower than expected subjectively; or b) the speaker scores very high subjectively, but has much poorer measured performance than you'd have expected from a speaker that scored high? In other words, were there outliers that clearly had either measured or subjective merit, and if so, any speculation as to what was going on?

Edit: I have another question, if perchance you are in the mood for such. In the section of your book entitled "Standing Waves in Real Rooms", beginning on page 208, you describe a very well-built room at Harmon whose measured modal behavior was significantly different from its calculated modal behavior. Does this imply that Todd Welti's investigation of symmetrical multi-sub configurations made some simplifying assumptions, and that visual symmetry does not necessarily equate to acoustic symmetry at low frequencies (especially since few rooms have symmetrical "soft spots" like windows, doors, closets, and HVAC vents)?

Thanks!

Duke

 

[3db]

I voted who cares because there should be another option... along the lines of a speakers measurements that fit your subjective wants or needs... ie best match for your hearing.

Measurements and tests from reputable 3rdparty sources don't lie about a speaker's performance. It is what it is and physics cannot be argued with. Where the issue lies is the brain's interpretation of what we hear. Hearing is very subjective while measurements are objective and absolute. However audiophiles claim the following,

If it sounds good and measures bad you're measuring the wrong thing. - old audiophile expression
If it sounds bad but measures good you're also measuring the wrong thing. - old audiophile expression

The above old audiophile statements that I have quoted come out of ignorance by the audiophile community as a whole and shows a complete misunderstanding in the difference between objective measurements and subjective interpretation. Subjective interpretation does not mean listening only. It also infers other biases such as our moods, visual stimulation, etc. How many of us sat down and listened to our systems listening to the same source of music and on different days and have reached different opinions of our system? Its not the system that performed differently. Its our subjective interpretation of what we heard at that time that has influenced what he heard.

One other point I would like to make is be careful what you read into subjective experiences of professional reviewers. If two different speaker makes have very similar curves and are reviewed in the same room by the same reviewer but the reviewer reaches two totally different conclusions, then I would flag that review as suspect and question the reviewer's state of mind during the review and write up process. I would expect the conclusions to be very similar and consistent based on those curves.

 

[Floyd Toole]

Hi 3db,
Sorry for the delayed response - I don't follow forums "religiously". You said in your note to me: "My instincts tell me that tests peformed by publications such as Stereophile and Soundstage will accurately model a speakers behaviour. Whether or not we like the sound of the speakers is a subjective preference that goes beyond what the tests are designed to do. Is it safe to say that if you have complete test results of speakers one prefers, you can use these results to purchase another pair of speaker with very similiar test results without auditioning them and like their sound."

The Soundstage measurements come from the facility I created around 1981 in my lab at the National Research Council in Ottawa Canada. Those are very good data. In fact I understand that the facility is still used by PSB Loudspeakers, the ones you own, and a few other manufacturers. Soundstage does not buy or publish the full set of NRCC measurements, but there are enough to identify good loudspeakers without ever listening to a note of music. They are the basis for the system that was further developed after I moved to Harman, and that has been incorporated into the ANSI-CEA 2034 recommendation for loudspeaker measurements, so it has gone well beyond a personal opinion at this point.

The John Atkinson measurements in Stereophile are basically correct and useful, but they suffer slightly from not being measured in an anechoic space, and there is no post processing to optimize the visual interpretations. That said, it is not difficult to recognize loudspeakers that have the potential of sounding good, or the reverse. Most interesting is that the conclusions drawn from inspecting the measurements do not always agree with the subjective reviews that have been done in uncontrolled, take-it-home-and-listen-to-it exercises. This is in distinct contrast to my experiences, and those of my colleagues over the decades, where in double-blind subjective evaluations the subjective opinions monotonously confirm the conclusions drawn from a visual inspection of competent and complete technical measurements.

So, your instincts are quite correct. All that is missing are accurate and comprehensive technical measurements on the loudspeakers you may be interested in. The "subjective preference" you allude to is something that has not been strongly revealed in double-blind listening - most people, most of the time, like and dislike the same loudspeakers. In fact it was this (then surprising) observation in my very first (blind) tests almost 50 years ago - described in Section 17.1, p. 339 in my book - that got my research started. If people agree on what is good, the next step is to find the right measurements to describe what they are liking. We have now done that.

The missing component in audio right now is consistency in the timbral quality of recordings. Recording facilities are not standardized, and not necessarily very good. So, the most "perfect" loudspeaker in the world cannot always sound good. In fact, we still need tone controls, especially a bass control, to compensate for easily audible spectral balance variations in recordings. Yet, many audiophiles seem to go through life thinking that recordings are flawless and that all problems are at the playback end. Wrong!

 

[3db]

Hi 3db,
Sorry for the delayed response - I don't follow forums "religiously". You said in your note to me: "My instincts tell me that tests peformed by publications such as Stereophile and Soundstage will accurately model a speakers behaviour. Whether or not we like the sound of the speakers is a subjective preference that goes beyond what the tests are designed to do. Is it safe to say that if you have complete test results of speakers one prefers, you can use these results to purchase another pair of speaker with very similiar test results without auditioning them and like their sound."

The Soundstage measurements come from the facility I created around 1981 in my lab at the National Research Council in Ottawa Canada. Those are very good data. In fact I understand that the facility is still used by PSB Loudspeakers, the ones you own, and a few other manufacturers. Soundstage does not buy or publish the full set of NRCC measurements, but there are enough to identify good loudspeakers without ever listening to a note of music. They are the basis for the system that was further developed after I moved to Harman, and that has been incorporated into the ANSI-CEA 2034 recommendation for loudspeaker measurements, so it has gone well beyond a personal opinion at this point.

The John Atkinson measurements in Stereophile are basically correct and useful, but they suffer slightly from not being measured in an anechoic space, and there is no post processing to optimize the visual interpretations. That said, it is not difficult to recognize loudspeakers that have the potential of sounding good, or the reverse. Most interesting is that the conclusions drawn from inspecting the measurements do not always agree with the subjective reviews that have been done in uncontrolled, take-it-home-and-listen-to-it exercises. This is in distinct contrast to my experiences, and those of my colleagues over the decades, where in double-blind subjective evaluations the subjective opinions monotonously confirm the conclusions drawn from a visual inspection of competent and complete technical measurements.

So, your instincts are quite correct. All that is missing are accurate and comprehensive technical measurements on the loudspeakers you may be interested in. The "subjective preference" you allude to is something that has not been strongly revealed in double-blind listening - most people, most of the time, like and dislike the same loudspeakers. In fact it was this (then surprising) observation in my very first (blind) tests almost 50 years ago - described in Section 17.1, p. 339 in my book - that got my research started. If people agree on what is good, the next step is to find the right measurements to describe what they are liking. We have now done that.

The missing component in audio right now is consistency in the timbral quality of recordings. Recording facilities are not standardized, and not necessarily very good. So, the most "perfect" loudspeaker in the world cannot always sound good. In fact, we still need tone controls, especially a bass control, to compensate for easily audible spectral balance variations in recordings. Yet, many audiophiles seem to go through life thinking that recordings are flawless and that all problems are at the playback end. Wrong!

Thank you for taking the time to respond.

 

[PENG]

That's why I am always in the science camp.

 

[3db]

That's why I am always in the science camp.

You and me both!!

 

[Mikado463]

Kind of explains why I like Revel speakers ………...

 

[Winkleswizard]

While science is great, grab a calibrated USB mic and a copy of REW software and take a few measurements around your room.

Even if your speakers are well designed and your room is more dead than most, you will see frequency response variations that are clearly audible.

If you care about correlating audible issues with what measurements, try checking the distortion levels on your speakers. You will quickly find out why few speaker manufacturers spec distortion figures.

On the other hand, if you are really happy with your system, may not want to start measuring it ...

Ww

 

[Bucknekked]

That's why I am always in the science camp.

me too.or is that, me three?

 

[Art Vandelay]

While science is great, grab a calibrated USB mic and a copy of REW software and take a few measurements around your room.

Even if your speakers are well designed and your room is more dead than most, you will see frequency response variations that are clearly audible.

If you care about correlating audible issues with what measurements, try checking the distortion levels on your speakers. You will quickly find out why few speaker manufacturers spec distortion figures.

On the other hand, if you are really happy with your system, may not want to start measuring it ...

Ww

Keeping in mind that you're measuring the combined distortion of the microphone and the speaker. Fwiw, my speakers generate less distortion than I was expecting - at normal listening levels, but only above 300Hz. It's in the region of 0.1 - 0.2% up to about 10kHz.

Perhaps my room is unusual but with the speakers toed in towards the listening position there's no point of first reflection from the mids or tweeters at the listening position (from the side walls) - because the speakers are far enough from a sidewall to throw first reflection points to the region behind the listening position. For said reason I found that the ceiling was the most critical point of reflection, and since virtually all tower type cone / dome floorstanders exhibit a messy vertical dispersion characteristic this would also hold for most brands of tower speakers with vertical driver arrays. The solution for me was to mount a couple of dispersers of dimensions appropriate for frequencies above about 1kHz at the 2 ceiling reflection points, and doing so resulted in a noticeable improvement to the overall tonal balance. Fwiw, the speakers are B&W 800 diamonds.

 

[Winkleswizard]

Keeping in mind that you're measuring the combined distortion of the microphone and the speaker. Fwiw, my speakers generate less distortion than I was expecting - at normal listening levels, but only above 300Hz. It's in the region of 0.1 - 0.2% up to about 10kHz.

Perhaps my room is unusual but with the speakers toed in towards the listening position there's no point of first reflection from the mids or tweeters at the listening position (from the side walls) - because the speakers are far enough from a sidewall to throw first reflection points to the region behind the listening position. For said reason I found that the ceiling was the most critical point of reflection, and since virtually all tower type cone / dome floorstanders exhibit a messy vertical dispersion characteristic this would also hold for most brands of tower speakers with vertical driver arrays. The solution for me was to mount a couple of dispersers of dimensions appropriate for frequencies above about 1kHz at the 2 ceiling reflection points, and doing so resulted in a noticeable improvement to the overall tonal balance. Fwiw, the speakers are B&W 800 diamonds.

While I did not know how many speaker manuafacturers were specing distortion, I did know B&W does. In any case, my main point in this a case was more about understanding the system. Not much point in pursuing amplifiers that have really low distortion, when your speakers are order of magnitude worse. Anyway, good choice on your part.

As for measuring speakers in a room, distortion is more of a relative measurement. Frequency response is clearly more of a direct measure and while the real challlenge is at lower frequencies, you make a interesting point about the impact of ceiling reflections. While they do impact FR measurement, there are major questions over their audibilty. If you have not already done so, suggest you read this https://www.audioholics.com/room-acoustics/room-reflections-human-adaptation or better yet, get Dr. Toole’s book.

In a more simple way, consider a performer playing a violin in your room. If your room treatment really improved the sound quality of the live performer, it may be worthwhile. But part of my prior posting was meant to warn against our tendency to overanalyze and overcompensate. From my automation systems experience, we used to say “just because you can, does not mean you should”. There are different factors involved, but I find this often holds true in many sound system situations as well.

Happy Listening!

Ww