How important is the efficiency of a speaker to you?

I think many of the values discussed here and elsewhere assigned to high efficiency speakers are not accurate. First the entire discussion of what is delivering the improved efficiency is not differentiated- is it drivers or how the driver is used that gets you better efficiency? These two ideas.make a huge difference in performance but are distinctly different methods of achieving improved efficiency with their own costs, benefits and penalties..

Example 1: a lower efficient driver can be made more efficient by placing it on a horn, which complete changes other characteristics/parameters of performance and behavior. An acoustic suspension system or a horn system are very different from each other in many ways having a direct impact on performance.

Example 2: You can build a more efficient driver by lowering the mass of its moving parts. Lowering this mass can cause multiple things to happen, typically increasing ring modes and distortion. Here efficency makes the driver sound worse. Another outcome of lowered mass is poorer low frequency performance.

Example 3 Higher efficiency drivers do not always offer lower power compression. If you use titanium instead of aluminum for your former, the thermal resistance is far higher with titanum than aluminum, so heat can’t escape the lower mass system increasing thermal power compression. Higher efficiency increases thermal power compression in this case. DO you know what your VC former material in the speaker system you buy actually is? Thermal power compression is affected by a long list of factors that are far too detailed and complex to discuss here or ina spec sheet: issues such as VC impendance, VC material, front plate/pole piece proximity to coil, airflow and venting of the magnet system. Looking at a efficiency measurement and assessing quality is like looking at frequency response and assessing sound quality- it really isn’t an indicator of performance in a way that is simple or easy to [universally] understand.

I know poeple really want there to be a simple way of looking at things but this is some deep science. A speaker involves practicaly all the sciences and there is no way to simplfiy it and express that in a spec- unless you are willing to bullshit or tell a helluva convincing marketing story. Most do exactly that, but its a diservice to the thinking person’s aims in a playback system.

If you love the idea of a flea watt amp, you are sorta stuck with high efficiency loudspeakers to have any hope of dynamic range. If you are more open minded and are wiling to look at moderate efficiency speakers in the mid 80s, using the wide array of excellent high power amplifiers available, or active speaker configurations, you can have your low distortion, wide dynamic range AND better bass AND wider dispersion, etc, etc. But you cannot have all that AND super high efficiency.

I think we are back to listening to a speaker with a proposed amplifer and assessing if this combination works for you and addresses the things you care about.

Brad

.

@phusis

I normally agree with your posts, but not this one. A transducer engineer like Doug Button (he worked at EV when I did, and then again at JBL when I was there) or Rich at ATC would agree with my post (maybe not exactly the way I say it). The physics of the speaker is a balancing act: if you want more efficiency you give up other advantages. Im not saying this is wrong, it’s simply a choice. A moderate efficiency speaker (86dB 1w/1m) is not a mistake either as it is just a different choice that enables other performance features that high efficiency cannot offer. You cannot have it all. My point is simply that this thread seems to universally promote that high efficiency is the primary hallmark of high performance and that is simply not true. It can be important if you need High SPL, or like horns or big wave-guides (which can image well), or use speakers in a highly reverberant environment where narrow dispersion is an advantage, but these conditions certainly don’t exist in every room or for every listener. It’s a trade off and is not a "good or bad" or "right and wrong" thing. Decide what you want, then figure out what speakers do that.

A good example of deciding what’s important to you is a listener in a small space; a small stand mount speaker with excellent low frequency output, say a LS3/5a type KEF design would be a good choice. Super high efficiency is not gonna happen in this type of design. If a listener wants wide dispersion speaker because he/she wants it to sound the same anywhere in front of the speaker, super high efficiency is not gonna happen. If a listener wants super low distortion because they are in mastering or recording, then super high efficiency is typically not a goal. The hardest part may be understanding what you want in your space vs what other people want in their space, as the reviews say this is good or bad but don’t really discuss the space or the listeners goals much. Whether it’s good or bad is ALL about matching your space and your goals with a speaker. High end speakers are not good or bad on some universal scale. There are too many engineering goals/design features to account for that define good or bad in a given space. Efficiency is just one of these many features.

Brad

@phusis

Nicely written! I enjoy your perspective. When you say that efficiency is not the most important aspect of performance, you and I are on the same page. I disagree that efficiency is the "preferred route" in achieving lower distortion, improved dynamics or increased bass. There are so many other avenues to these features.

I did not say that low eff. designs avoid the limitations of poor efficiency simply by adding more power. There may not be limitations of lower efficiency designs, depending on your goals. Or there could be limitations of a high efficiency design. I think the whole array of solutions are far more complex than high or low efficiency "spec" on a spec sheet. Whether something is 92dB 1w/1m in a loudspeaker gives you zero information about quality when compared to a 86dB 1w/1m spec.. You only know the designer chose to chase efficiency (probably for more than one reason). Intended application is everything..

Some lower efficiency designs can achieve lower distortion in the driver, or can extend low frequency of that driver or both. Or your lower effieciency design may improve cooling and power handling without extending voice coil length. I guess to me its like horsepower tells you zero about the performance of a car. There are so many different options available to a skilled designer that focusing only on a high efficiency design is not wise, again, depending on your goals. I favor lower distortion myself, I want to hear more of the fine details, the reverb tails, the room sound, etc. If I can have lower distortion that reveals more of that fine detail in exchange for a larger power amp, I’m in.  That's the trade off I am talking about.  And I'll stick to my guns on this one, you cannot have it all.  

Brad

.

Again

Speaker sensitivty alone is not an indicator of quality or sonic virtue in loudspeakers.  I can think of other values that might be related to sonic virtue, such as widely varying impedance (in a passive design), high distortion, limited dispersion, inconsistent dispersion across the audio band, etc, etc.  High efficiency is simply an engineering choice made, not an evolution.

Brad       

@daveyf

You are right about the amp issues regarding hi vs lo efficiency designs, well said. Yes, a low efficiency design might require more power.  But these issues are truely complex and trying to make a blanket statement that some performance related feature is always good or always bad doesnt really work in audio. People long for universal truths, but in audio, the truth is conditional.

Brad

Can we say then that efficiency in speakers is not the #1 goal?  That the idea that dynamic range is improved or noise floor is changed due to speaker efficiency alone is simply not true?  While there may be examples that support this idea, there are far too many exceptions to make this speaker efficiency issue a universal need?

Brad

@atmasphere 

Im going to depart with your thinking when you say things like

"There are attempts to get around the thermal compression problem with lower efficiency speakers, such as vented pole pieces and the like. The incentive is high because lower efficiency speakers are a lot cheaper to make."

This is simply not true.  As mentioned before and elsewhere, lower efficiency drivers may sacrifice some efficiency to gain other features that improve performance such as lower distortion, better linearity, better low frequency extension etc.   The idea that lower efficiency is always done as a cost saving excercise is clearly not true.   

Brad

Again: efficiency is just ONE aspect out of many that has to be addressed in high performance loudspeaker design.  High performance drive units are so much more complicated than "how loud they are", with so many more aspects to them than efficiency alone.  Would you judge a car's track performance on miles per gallon?   

For an alternative dicussion about drivers from a man who is highly respected everywhere in the world for driver design, try this.  Billy explains much better than I:

.ATCWhitepaper.pdf (transaudiogroup.com)  .   

Brad

I suspect the $4,000 price claimed is not real- likely seen by atmasphere  somewhere on the resale market because the product is long discontinued.   That price is not representative of what it really sold for back when it was made or what current drivers sell for that replaced it from TAD.  I know the distributor in the US.

 TAD DOES make expensive drivers, they are Japanese (Pioneer) but are used mostly in their own speakers or OEM in horn loaded systems.  Very good drivers for sure.  TAD was a good low distortion driver used regularly by George Ausperger in his large all horn designs which were quite popular in recording studios years ago.  Today, TAD is still used by folks that build these all horn loaded systems that echo those Asperger designs.  I hate the way they sound- high distortion and just loud.  Lower fidelity by today's standards.  They are common for hip hop where SPL matters on client playback.  Mixing still done on lower distortion direct radiators (we sold TImbaland ATC 100s for this exact purpose).

Many studios still have these TAD or JBL loaded horn systems in the wall- they do look cool and are part of the studio "vibe".  They were always used for the band to hear playback at higher SPL in the best possible fidelity (which is very low compared to modern hi fi or studio speakers).  SPL = excitement.  The trend now in studios is direct radiators, such as ATC, that favor low distortion so you can mix faster, worry less about translation and hear more details.  

It is interesting that the TAD 1602 drivers had very little test data or info available.   It does say the 1602"S" was a special version of it with shorter voice coil for lower distortion.  You would have read about that same short coil idea in the white paper from Billy Woodman I posted.   All the ATC studio and hi fi drivers use short coil long gap topologies for reduced harmonic distortion.

This is precisely what I am saying that efficiency is but one part of the picture as today short coil, venting, narrow gaps with precise coil fitment, flat wire coils, the right former material, etc is way more important when amps are relatively cheap.  These types of drivers are impossible to build by machine.  Hand made drivers is the way to get it done and everyone in the industry used to build their own: KEF, B+W, JBL, on and on.  Now, hardly anyone does this anymore as it's way too expensive, difficult to train people and environmentally challenging.  A small company cannot afford to build their own.  ATC is the last of the breed in the UK.  Everyone else has left and gone to China.

Brad      

@rauliruegas 

it’s true - I am ATC USA and do not want to hide that.   Previously I’ve worked for EV (their rep out of Chicago including OEM)and JBL (in charge of cinema and installed sound in the USA)and now ATC ( I’m the importer for both pro and hi fi for the entire US).  I have the privilege of being able to ask an engineer who does driver design for living what’s what.  My goal is as much for me as for my customer: the more you know the better, as most bad decisions in audio are made based on bad info or lack of info.  There is a lot of misleading info posted on forums - this is what motivates me to post, to represent a different perspective, to represent what I know to be true.

Brad

@rauliruegas

Very good point and great example that there is more to know about a speaker than a 1w/1m spec. Could a 86dB 1w/1m speaker with a very flat impenace curve  outperform a 91db 1w/1m speaker with a wildy changing impedance curve? Yes.

Brad

@phusis

My argument is that this efficiency as it relates to speaker performance is a complex issue, much more than a simple number. You forget I am engaged with dealers and end users who ask these questions (and similar ones) quite often. The desire to reduce a complex issue down to a simple one is very attractive when the subject is complex. THAT is my point, not that there aren’t times when facts are facts.

Since you own DH1A, I can use a historical reference: I bet you remember many people arguing the advantage of 1.3 inch throat driver/horns (EV) vs 2 inch driver/horns (JBL) and vice versa. That discussion was often reduced to focus on one attribute (throat diameter) when really it was a far more complex issue than that, depending on what you were trying to accomplish.

Brad

Is the entire "speaker efficiency" conversation motivated by passions and opinions about the free choice of amplifiers -not speakers ? 

Brad 

@daveyf

One of my points is that this very issue of "what amp is enough?" is based around very fuzzy poorly defined/poorly understood info about speaker efficiency, Would you call a 90dB SPL 1w/1m speaker efficient or not?

Brad

Asking the question "What is efficient enough?" brings up mutiple challenges.  the first being hardly anyone listens much over 90dB SPL!   

I was at AXPONA this weekend and in our room we are demong ATC.  I meassuredIf  the SPL when things got "loud" : it was around 92dB SPL.  If the 1w/1m spec of a passive was 90dB, it would require less than 2 watts to get that speaker to play at 92dB 1M.  If it was 86dB 1w/1M it would require 4 watts to get it to 92dB at 1M!  92dB SPL is very loud for many of us, most audiophiles wouldnt even want it at 90-92dB SPL in their living room or listening room. 

It appears in reading this thread that most would agree that above 90dB 1w/1M is efficient and 86dB 1w/1M is "not efficient". We need 2 watts to get our 86dB speaker to 89dB SPL, which is indistinguishable from 90dB to most of us.  This 2W instead of 1W is really a critical issue and one that drives a purchase?   

My point is that very very few of us listen at 92dB SPL. The argument for an "above 90dB 1w/1M speaker" is not a relevant argument as no one even wants to listen that loud.  90% of our listening is WAY under 90dB SPL.  If this is true, why is 90dB+ 1w/1M efficiency such a important spec that we all need to pay attention to it?

Brad