Speaker sensitivity vs SQ


My first thread at AG.

Millercarbon continues to bleat on about the benefits of high sensitivity speakers in not requiring big amplifier watts.
After all, it's true big amplifiers cost big money.  If there were no other factors, he would of course be quite right.

So there must be other factors.  Why don't all speaker manufacturers build exclusively high sensitivity speakers?
In a simple world it ought to be a no-brainer for them to maximise their sales revenue by appealing to a wider market.

But many don't.  And in their specs most are prepared to over-estimate the sensitivity of their speakers, by up to 3-4dB in many cases, in order to encourage purchasers.  Why do they do it?

There must be a problem.  The one that comes to mind is sound quality.  It may be that high sensitivity speakers have inherently poorer sound quality than low sensitivity speakers.  It may be they are more difficult to engineer for high SQ.  There may be aspects of SQ they don't do well.

So what is it please?

128x128clearthinker
Sorry Duke, my bad. My point is that dynamics are a matter of volume. A speaker that can hit 110dB without compression is going to be more dynamic than a speaker that can only get to 100 dB even if it is less. efficient. Just a matter of power. Horns are very dynamic because they go very loud. They do it with less power because they tend to be very efficient. As far as sound quality goes, it's a toss up.
Another issue is trying to run 15" woofers up to 700Hz then crossing to a horn. Two very dissimilar drivers crossed right in the meat of the midrange. You really have to push that crossover lower and in order to do it you are stuck adding a tweeter. I'm  all in for limiting crossovers but horns have their limitations in this regard. This is what makes ESLs so special. You can easily get them to go from 100 Hz to 25kHz. Obviously you can make them go lower but IMHO you are much better off going to subwoofer especially if you can use a digital crossover. 
Audio2design, I know for  fact  you do not have to have active loud speakers to take advantage of digital signal processing. Perhaps it will be the way the market goes but I am not so sure and I am an unabashed huge fan of digital signal processing just check out my system page.
Alexberger, in order to have inertia you have to have mass. Thermal distortion (whatever that is ) certainly has no mass. You are more than welcome to your SET amps. They are most definitely not my cup of tea.
I have heard several speakers in an anechoic chamber and they sounded just fine. You can test your own by just moving them outside. Tuning a speaker in an anechoic environment is a fine way of achieving a flat curve or rather the curve you want. In order to tell what your speaker is really doing it is the only way. What happens in a room depends on the speaker's dispersion characteristics and the room itself. A speaker that is more directional will sound very similar in room as in an anechoic chamber. Controlled dispersion is the best room treatment you can get followed by digital room control (speaker control) and a few foam tiles. 

Post removed 
voicing in a chamber would ignore room gain and reality, it’s a great tool for matching to a standard ( impulse, freq response, pair matching thru nulling ( hint ), etc.

For those dealing in the reality of listening in a room, RT 60 is the path to better sound.
So, this is the part I'm not clear on.

You can use DSP to deal with non-linear behavior in a driver which are constant.  You measure the output at say 90 db, and at 70 db, and you create transforms which adjust the output accordingly.

However, what I know of as thermal compression is not constant.  The driver has a thermal history and changes behavior accordingly. That is, as the driver gets hotter it no longer behaves the same way. To compensate for this you would need to have an algorithm which has an accurate thermo-acoustic model for the speaker AND knows the actual voltages applied.  That's a great deal of work vs. getting drivers which are thermally stable in their usable range.

Best,

E