A Broad Discussion of Speakers with Major Audio Luminaries

[Floyd Toole]

It is resonances that are mainly responsible for lack of neutrality in loudspeakers, so the first task for competent speaker designers is to minimize them. After that spectral balance and directivity control need to be attended to. The industry standard measurement method, used by Amir, which had its origins in Canada while I was at the National Research Council, is excellent at revealing resonances. It seems that, historically, this company has studiously ignored the science and gone the route of designs with "personality". It is a choice, and clearly there is a market for "colored" loudspeakers. This may make the loudspeaker sound "distinctive" in a demo, but the trouble is that everything played through them is colored in the same way, which eventually gets tiresome. Neutrality wins, and simple tone controls or equalization can compensate for most problems in recordings or add coloration for personal preferences. The advantage of starting with neutral loudspeakers is that one can go back to listening to what was recorded.

 

[Gringoaudio1]

Do these resonances come from cabinet design primarily?
If from the drivers can notch filters in the crossover solve this?
If from the cabinets I’d like to see what internal bracing would do to help.

Canada has always had a lot of people making a living from stuffing boxes with speaker drivers not knowing what the heck they were doing. I have bought some of these brands when just having a big stack of black components and big party blaster speakers were part of the ego accessories of a 20-something in the 1970s. There has always been a market for affordable speakers. That some brands rise from the dreck to become aspirational items, despite being inferior, is due to marketing and fashion. Some manufacturers learned something from the NRC association like Energy who made well measuring speakers but failed anyway. StudioLab made a decent speaker in its last incarnation after years of big boomy crap. But they too are gone. The hucksters and salesmen prevail. Well not all… thank god Nuance is long gone. Maybe ASR and its brothers who measure will eventually bring an end to the box stuffers and elevate audio speaker manufacturing to a respectable engineering based profession.

I’m enjoying your book. It is taking me a long time to get through with all the responsibilities of life coming before hobbies though!

 

[Floyd Toole]

Do these resonances come from cabinet design primarily?
If from the drivers can notch filters in the crossover solve this?
If from the cabinets I’d like to see what internal bracing would do to help.

Canada has always had a lot of people making a living from stuffing boxes with speaker drivers not knowing what the heck they were doing. I have bought some of these brands when just having a big stack of black components and big party blaster speakers were part of the ego accessories of a 20-something in the 1970s. There has always been a market for affordable speakers. That some brands rise from the dreck to become aspirational items, despite being inferior, is due to marketing and fashion. Some manufacturers learned something from the NRC association like Energy who made well measuring speakers but failed anyway. StudioLab made a decent speaker in its last incarnation after years of big boomy crap. But they too are gone. The hucksters and salesmen prevail. Well not all… thank god Nuance is long gone. Maybe ASR and its brothers who measure will eventually bring an end to the box stuffers and elevate audio speaker manufacturing to a respectable engineering based profession.

I’m enjoying your book. It is taking me a long time to get through with all the responsibilities of life coming before hobbies though!

Most audible resonances in most loudspeakers originate in transducers. Cabinet bracing is part of competent design, using scanning laser vibrometers to identify vibrations that radiate sound - not all do. The knuckle test does not work. Heroic cabinet materials and elaborate bracing schemes are mostly marketing handles.

Parametric equalization based on anechoic data can effectively eliminate audible resonances in loudspeakers. In-room curves cannot provide the evidence. This is why active loudspeakers, incorporating DSP are advantageous.

However, steady-state room curves are the definitive evidence for identifying and attenuating room resonances, which everybody deals with and which account for about 30% of our overall sound quality ratings. Rooms are arguably the weakest link in the sound reproduction process, and I am not talking about reflections.

I think it is fair to say that all countries have done their share of stuffing boxes with drivers and inventing stories to sell them. Several countries have also made some very fine, neutral loudspeakers. Thanks to the Canadian taxpayer supporting the NRC effort and HARMAN International continuing it - both allowing the research to be published for the world to see - there are now clear guidelines for designing timbrally neutral loudspeakers. That some companies do it better than others is a measure of engineering competence. If others choose to ignore the science - it is a free world and business is business. The scientific facts are real and will not go away.

But it is evident from Amir's and Erin's excellent measurements, and others to be found on www.spinorama.com there are many loudspeaker manufacturers the world over who are not only aware of the guidelines but who are following them. The average sound quality of loudspeakers is definitely going up. Unfortunately, we are still saddled with stereo as the default medium - if one is speaking of weak links.

I'm pleased you are enjoying my book. The 4th edition will be quite different, with new insights.

 

[Beave]

While Dr. Toole may very well be correct that "most audible resonances in most loudspeakers originate in transducers," in the case of several Totem designs, the multiple high-Q, high-amplitude resonances come from their ports. The broader variations from neutrality come from the drivers and crossover design along with the box dimensions. It seems the designer designs mostly by ear.

 

[pablolie]

Everything that's been said here obviously is very correct and insightful.

One question though: are we saying that all scientifically designed/engineered speakers should sound the exact same, and the only difference would be the *looks* of the design?

 

[Beave]

Everything that's been said here obviously is very correct and insightful.

One question though: are we saying that all scientifically designed/engineered speakers should sound the exact same, and the only difference would be the *looks* of the design?

No, I don't think so. First, bass extension will always vary. Second, even if speakers are fairly flat on-axis (say, +/- 2dB), that's enough to be audibly different. And third, even if speakers advanced to the point of being so flat on-axis (say, +/- 0.5dB), that would still be slightly audible,. And fourth, speakers can and will still vary significantly in their off-axis dispersion. Two perfectly flat on-axis speakers could each have excellent directivity, but one with narrow dispersion and one with much wider dispersion. Both are "scientifically designed/engineered." Both sound great. But they don't sound the same when put in a room.

 

[Floyd Toole]

Everything that's been said here obviously is very correct and insightful.

One question though: are we saying that all scientifically designed/engineered speakers should sound the exact same, and the only difference would be the *looks* of the design?

Hundreds of double-blind multiple loudspeaker subjective comparisons indicate that we are at or very close to the point of diminishing returns when comparing well engineered loudspeakers. There are subtle differences among them, and the interaction with the differences in recordings causes the results to end up with statistical ties. As mentioned bass extension is very important, but bass quality is determined by the listening room. If the differences are how they interact with the listening room - e.g. directivity - the differences may be heard in monophonic comparisons, but in stereo the differences are less noticed, and in multichannel even less so. The reason is that the acoustic signature of the listening room is subordinate to the "room sound" in the recordings - usually of larger spaces, real or synthesized. More channels deliver more convincing envelopment; the impression of being in a large space.

The most "perfect" loudspeaker cannot sound "perfect" because the rooms and setups dominate bass timbre (they are all different) and they require treatment that few audiophiles bother to do. Recordings are not consistent because they are monitored through unknown loudspeakers in unknown rooms by recording engineers who have a strong probability of suffering from hearing loss. Tone controls are useful accessories.

And, the scientific guidance now existing gives engineers much more flexibility in dealing with distinctive industrial designs. There are limits, of course.

 

[MaxwellsEq]

Everything that's been said here obviously is very correct and insightful.

One question though: are we saying that all scientifically designed/engineered speakers should sound the exact same, and the only difference would be the *looks* of the design?

You've had some excellent answers (i.e. "no"), but it's possible to determine an answer on a basic principle: cabinet and driver size! Some people listen in small rooms, such as home offices where a large floor-standing speaker would not fit. A small "scientifically designed/engineered speaker" in a, well-treated small room is unlikely to have bass extension that plays 32' organ pipes at realistic levels. So this small "scientifically designed/engineered speaker" will inevitably sound different in comparison to a much larger "scientifically designed/engineered speaker", especially if you compare them in a large well-treated room.

 

[pablolie]

You've had some excellent answers (i.e. "no"), but it's possible to determine an answer on a basic principle: cabinet and driver size! Some people listen in small rooms, such as home offices where a large floor-standing speaker would not fit. A small "scientifically designed/engineered speaker" in a, well-treated small room is unlikely to have bass extension that plays 32' organ pipes at realistic levels. So this small "scientifically designed/engineered speaker" will inevitably sound different in comparison to a much larger "scientifically designed/engineered speaker", especially if you compare them in a large well-treated room.

Thanks for all the replies.

A lot of it boils down to bass extension and issues with bass. The answer proposed here is cabinet and driver size... but what is the real benefit of huge speakers in today's subwoofer world? Smaller speakers aided by one or two subs have the advantage of positioning flexibility. Large speakers very seldom are crammed -even by expert reviewers- in environments that don't seem geared to make them shine. I know when I moved from a pretty large listening room to a smaller place (divorce etc), the large speakers I loved sounded downright hellish.

 

[Pieter1267]

Thanks for all the replies.

A lot of it boils down to bass extension and issues with bass. The answer proposed here is cabinet and driver size... but what is the real benefit of huge speakers in today's subwoofer world? Smaller speakers aided by one or two subs have the advantage of positioning flexibility. Large speakers very seldom are crammed -even by expert reviewers= in environments that don't seem geared to make them shine. I know when I moved from a pretty large listening room to a smaller place (divorce etc), the large speakers I loved sounded downright hellish.

Thats why "insert your supernatural here" invented small floorstanders. Put a sub in support and low pass it a bit lower than you would with bookshelf, to me, gives the best compromise. I never forget from the very first moment in the shop, how my small Neat Acoustics sx2 sounded so much better than Totem bookshelfs 4 times the price next to it.

 

[Mart68]

A lot of it boils down to bass extension and issues with bass. The answer proposed here is cabinet and driver size... but what is the real benefit of huge speakers in today's subwoofer world?

Assuming proper sub integration then in reality probably none - but that hasn't stopped me getting some 180 litre full rangers as, probably for psychological reasons only, that's what I prefer.

 

[pablolie]

Assuming proper sub integration then in reality probably none - but that hasn't stopped me getting some 180 litre full rangers as, probably for psychological reasons only, that's what I prefer.

As you are perfectly entitled to! Having read several of your posts, I am sure you use them well. Enjoy!

I have kind of evolved into minimalism that current options can provide, with a very strong preference for tidy and compact.

It's awesome we have so many choices to put competent systems together.

 

[Duke]

The 4th edition will be quite different, with new insights.

That's the book I want. I had the first edition and gave it away after I bought the third edition, which I still have.

Any hints as to when the 4th edition might become available?

 

[Floyd Toole]

That's the book I want. I had the first edition and gave it away after I bought the third edition, which I still have.

Any hints as to when the 4th edition might become available?

The manuscript is at the publisher, as of a couple of weeks ago. Lots of work yet to be done, including a coordinated companion website, so sometime in 2025 is the best I can do. Sorry. I am now relaxing a bit for the holiday season.

Added May 19/2025. The most recent publication date is November 2025.

 

[Duke]

The manuscript is at the publisher, as of a couple of weeks ago. Lots of work yet to be done, including a coordinated companion website, so sometime in 2025 is the best I can do. Sorry. I am now relaxing a bit for the holiday season.

That sounds GREAT, and thank you VERY MUCH for taking the time to reply!

 

[fcracer]

Some manufacturers learned something from the NRC association like Energy who made well measuring speakers but failed anyway.

Gosh, thanks for bringing back some great memories. The first “hifi” I bought with my first real job paycheque was an NAD T750, paired with Energy C2 “Connoisseur” bookshelf speakers. I didn’t know anything at the time and spent 20% of the cost on cables, but I knew I loved the sound of that setup. I recall them having a lot more bass than expected for their size. I had some hardcore house parties with those little beasts. I may still have those speakers in my storage somewhere!

 

[Floyd Toole]

I wondered how accurately the speaker shifts to match a very complex waveform. That's a different ability from producing an even frequency response graph, because those seem to be based on matching tones, and don't test the responsiveness of the speaker as the waveform changes.

I can help make your life a bit easier by introducing some basic science. Loudspeaker transducers are "minimum-phase" devices, meaning that the time-domain behavior is predictable from the frequency response - including "speed", "velocity" etc. of movement necessary to reproduce the combined waveforms of complex sounds. The ideal frequency response to avoid time domain distortions is "flat", as in electronics (also minimum phase). Nobody would build a power amplifier with the frequency response of most loudspeakers. However, the encouraging thing is that some loudspeakers are now approaching the neutrality of electronics - decades late . . . Totem is a company that has not used neutrality as a goal, but we can hope they might change.

 

[MaxwellsEq]

when I said "velocity" I wondered how accurately the speaker shifts to match a very complex waveform. That's a different ability from producing an even frequency response graph, because those seem to be based on matching tones, and don't test the responsiveness of the speaker as the waveform changes

I understand the point you are trying to make, but this is unfortunately not what is happening.

If a loudspeaker has sufficient dynamic range and a perfectly flat frequency response that can go as low as we can hear and as high as we can hear, it will be able to reproduce any waveform audible to humans no matter how complex that waveform.

A lot of Hifi magazines and forums talk about velocity, speed etc. especially in regard to bass - often about playing tunes, or "timing" etc. But there's a bit of underhandedness involved. Quite often "slow" and "overhung" bass is caused by standing waves and room nodes. So if you design a speaker with reduced low frequency extension, you are less likely to energise room nodes and bass sounds "faster". For jazz, rock, pop and a lot of classical music, this is OK, because there is very little deep bass. For electronica and pipe organ music, it's not so good.

 

[Penelinfi]

I guess a small driver is "faster" because it has to move more. You likely can't hear that speed as much as you can hear distortions or lack of, dispersion etc from the speakers.

The small ones will usually have a slimmer baffle. They will radiate omni to a higher frequency.
They might have less sound coming back through the cone. They will have a higher breakup mode or less chance for certain cone resonance. They will also have more distortion in bass usually and more IMD.

Larger drivers tend to be on wider baffles, being more directional earlier, less sound bouncing around the room, can sound cleaner or more congested depending on driver quality, cabinet solidity, etc

 

[Floyd Toole]

I guess a small driver is "faster" because it has to move more.

A small transducer is able to reproduce higher frequencies and these are "faster" than lower frequencies - that is all there is to it. Within a given bandwidth, all loudspeakers are equally "fast". There are no "fast" woofers because the "speed" is determined by the highest frequency it reproduces. If it is bass managed, that is typically 80 Hz. The room resonances are the "slow" part, and unless these are controlled, bass will be "slow", as MaxwellsEq said. Science is different from subjective reviewing - facts are involved.