Higher sensitivity - more dynamic sound?

I know Doug Button, one of the brilliant engineers of the late 80s and 90s who led us to a series of higher output JBL drivers called "vented gap" which enabled dissipating higher heat which enabled higher SPL (and more reliable live sound systems with fewer boxes producing the same output). These were no home hifi drivers, but Live Sound drivers.

This whole efficiency argument is twisted by marketing. If you ask a designer like Billy Woodman at ATC (who is on Dougs level), he will tell you if you want additional low end from a driver, you can optimize it for more bass but the efficiency will decline. So a lower efficiency driver may have a better performance from a bandwidth perspective and may indeed be desirable. This is what folks like ATC and B+W and Magico do to get superior bandwidth. Its a choice, not a mistake; this choice does not improve or decrease dynamics- the size amp you mate it with does. 

Its the combination of efficiency and power handling of the driver that determine dynamic range.

So your 102dB 1w/1m speaker may not have such good dynamics with a 20W amp if 1W= 102dB SPL then .2W=105, 4W =108, 8W =111, 16W=114dB SPL and we are out of [low distortion] power.  That’s 12dB of dynamic range!  That's not even equal to the dynamics of vinyl.

So now compare a speaker with 86dB 1w/1m:  2W = 89dB, 4W = 92dB, 8W=95dB, 16W = 98dB, 32W =101dB, 64W=104dB, 128W= 107dB, 256W = 110dB SPL!  So the 86dB 1/1m speaker on a 250W/ch amp has 24dB of dynamic range! That's a huge increase when you take into account the log level nature of dB SPL, ideas such as 10dB SPL is considered twice as loud.  12dB dynamics is never better than 24dB dynamics, under any measurement or circumstance.  

Don’t drink the high efficiency kool aid kids!

Brad

@audiokinesis, Bravo!

@lonemountain —

This whole efficiency argument is twisted a bit by marketing. A designer liek Doug will tell you if you want additional low end from a driver, you can optimize t for this but the efficiency will decline.

That’s a matter of size, not that you can’t mate efficiency with extension on principle. My tapped horn subs 97dB sensitive and will do ~127dB’s at the tune, which is 22Hz. 20cf. volume per cab - that’s (one of the reasons) why.

So your 102dB 1w/1m speaker may not have such good dynamics with a 20W amp if 1W= 102dB SPL then .2W=105, 4W =108, 8W =111, 16W=114dB SPL and we are out of [low distortion] power. That’s 12dB of dynamic range! That’s not even equal to the dynamics of vinyl.

Why would you limit the power to 20W for this example? Only suits your argument. I have a 97 to 111dB sensitive subs to main speaker setup powered actively by ~2.5kW total - problem solved.

So now compare a speaker with 86dB 1w/1m:  2W = 89dB, 4W = 92dB, 8W=95dB, 16W = 98dB, 32W =101dB, 64W=104dB, 128W= 107dB, 256W = 110dB SPL!  So the 86dB 1/1m speaker on a 250W/ch amp has 24dB of dynamic range! That's a huge increase when you take into account the log level nature of dB SPL, ideas such as 10dB SPL is considered twice as loud.  12dB dynamics is never better than 24dB dynamics, under any measurement or circumstance.  

That's being creative. Not only will you have to deal with peak power compression effects here but long term power compression as well, which in effect limits the available dynamic bandwidth. May look good on paper, but..

Duke, nice post on thermal/mechanical driver compression. It’s an esoteric topic, but almost every transducer designer has to come to terms with what steps to take in their driver design to cope with heat, bandwidth, efficiency and intended performance. I know the ATC driver designs are remarkable, such as short coil/long gap, in lowering distortion while maintaining a wide dynamic range.  Some of the live sound driver designs these days are insanely good.    

@lonemountain , I don’t follow your reasoning. You said:

"So your 102dB 1w/1m speaker may not have such good dynamics with a 20W amp if 1W= 102dB SPL then .2W=105, 4W =108, 8W =111, 16W=114dB SPL and that’s it! That’s 12dB of dynamic range.

"86dB 1w/1m, 89dB 2w, 92dB 4w. 95dB 8w, 98dB 16W, 101dB 32W, 104dB 64W, 107dB 128W, 110dB SPL at 256W/1m. 86dB 1/1m speaker on a 250W/ch amp = 24dB of dynamic range! 12dB dynamics is better than 24dB dynamics?"

The dynamic range of a system is the difference between the softest and the loudest sounds the system can produce. Assuming both systems are limited by the same noise floor, the 102 dB system is capable of 114 dB peaks while the 86 dB system is capable of 110 dB peaks. So in your example, the 102 dB system has 4 dB more dynamic range.

But that’s not the whole story. Making another comparison and using 110 dB SPL for both, the 102 dB system will have negligible compression effects from the 7 watts it needs (this may not be true if it’s a single fullrange driver), while at 110 dB/256 watts the 86 dB speaker is far more likely to have thermal compression effects. And since musicians use dynamic contrast to convey emotion, compression effects tend to rob the music of emotion and "life". The use of very high-quality drivers (and/or many drivers) in the 86 dB speaker does make a worthwhile improvement in this area.

Let’s revisit noise floor for a moment: To a crude first approximation the in-room reflection field can be thought of as a "noise floor" which can mask low-level sounds. For a given room, the narrower the loudspeaker’s radiation pattern, the higher the direct-to-reverberant sound ratio, and therefore the lower the "noise floor" imposed by the in-room reflection field. So IF our 102 dB speaker is also highly directional (which is likely), it will probably result in a lower effective "noise floor" and correspondingly greater dynamic range and may result in better retrieval of low-level details.

Edit: Looks like several of us were typing at the same time, and I was the slowest.

Duke