The Harbeth phenomenon

Their sensitivity of generally around the 86dB mark makes them rather inefficient and therefore, at least in theory, not a good match for many lower powered tube amps, or any amps below 100wpc. Their frequency range is simply inferior to most high-end speakers since they don't go below 40 Hz.

Disclaimer - I have Spendors, not Harbeths. But they are from the same BBC lineage.

Try this perspective; the vast majority of natural musical sounds we hear are in the midrange, and they are not super loud. The fundamentals of the human voice primarily reside in frequency range of roughly 100 Hz to not much more than 1,000 Hz. Very few instruments have fundamentals above 4,000 or 5,000 Hz.

Everything above that is pretty much harmonics. These are certainly important, but if you don't have the critical midrange right, then all the bass, upper highs and extra volume won't fix the lack of naturalness.

Speakers are like any other product. Think of the old debates between car owners. Some people like the power and acceleration of an old school muscle car and others are more taken by the nimble suspension of a road-tuned sports car. Each group likes what they like and no amount of rationalization from the other side is going to change their minds.

The only thing you can do is listen to a set of Harbeths and see what they do for you. You may be enchanted, and then again, you may not.

James63 wrote: "I don't get how you could go through the effort of making/buying great drivers and using a so so box."

If I can hazard a guess as to the direction you're coming from, there are two broad categories when it comes to speaker cabinets.

One is the Harbeth/Spendor/BBC approach which acknowledges the existence of the back wave of the speaker cone and sets out to do something to help dissipate it. Their practice is to make a very carefully crafted "lossy" cabinet that acts as a "crumple zone" would in a car crash to absorb a lot of that energy. The energy that is absorbed by the cabinet reduces the amount radiated into the room at frequencies where the ear is most sensitive.

The opposite approach is to make an ultra rigid cabinet. The one drawback to this approach is the speaker cone still produces a rear wave. This energy does not magically go away by itself. If the speaker designer doesn't do something with it, you can find that energy re-radiated back into the room through the driver's cone. Energy in the lower voice range is not well absorbed just by speaker stuffing so the designer has to think of other things to do with it.

Each approach - and the variants in between as well as dipole and other boxless designs - has its following, just like car enthusiasts or the subgroups that develop in any other hobby.

However, calling the Harbeth a "so-so box" ignores the extensive effort that went into the speaker as well as the philosophy behind it. It would be just as easy to look at any of the tall, narrow speakers that are popular these days and insulting their designers for complicating the baffle step effect for the sake of furniture fashion.

Again I have never heard them but the problem with the absorption method is that it will only work at certain frequencies.

And the frequencies it does work well at are the frequencies where human hearing is most sensitive.

You forgot to address the point that with rigid wall speaker cabinets, the rear wave simply doesn't magically go away. Your re-radiation problem is still there - it has simply been moved to a different spot.

From a historical standpoint, keep in mind this cabinet type was designed by the BBC for monitoring purposes. There was a lot of reasoned thought and study that went into them by a group of very talented designers whose primary goal was to give the recording engineers an honest representation of their work.

Keep in mind that every choice one makes in engineering any product involves a compromise. When one addresses one problem, there is always another one created. The question is which of the issues is less a problem?

The wide baffle is a good example. A wide baffle can help move the step effect away from an area where hearing is sensitive. Robert E. Greene has discussed the technical aspects of this in great detail in various articles and on his web site. Narrow speakers may solve one issue, but they increase problems in an area where hearing is sensitive.

Same thing with off axis frequency response. This can actually increase problems in a reflective room situation since wide dispersion increases the ratio of reflected energy to direct arrival sound. This can cause smearing problems of its own. The BBC design philosophy is intentionally not interested in a lot of off-axis energy in the higher frequencies.

Again, that is a conscious design decision.

Many people would think such a speaker would work better in your room than a model with wide dispersion at higher frequencies.

The crux of the matter is that different professionals can have legitimate disagreements as to what issue is more important for any given design aspect. You may not agree with their choice but that's fine. Just don't think that wide baffle was some careless and arbitrary choice made by someone who didn't know what they were doing.

I just finished a short listening session of Michael Feinstein with a solo piano singing some Jerry Herman material. I have Spendor SP1/2Es. It would be hard for me to imagine a more natural presentation of this recording than what I heard. So, for whatever design flaws you consider the classic BBC design to have, it works for me. ;-)

I recently auditioned a pair of 7-ES3s for a couple of hours. The distance was more like 7 1/2 or 8 feet, but I didn't notice any driver integration problems at that distance. I found them a very natural sounding speaker, devoid of any audiophile gimmicks.

But then Sam T. is a professional reviewer and by trade has to find something to call out. And different people are sensitive to different things. That said, I've read Tellig uses this model Harbeth as his reference, so it can't be too much of an issue.

Sounds like you're at the point where you simply need to listen for yourself.