Rerurn to Romy the Cat's Site


In the Forum: Horn-Loaded Speakers
In the Thread: Aporia - Silbatone Acoustics speaker
Post Subject: A thorny issuePosted by Joe Roberts on: 2/9/2009
fiogf49gjkf0d


But we have all heard the difference, with standard stereo installations, where the bass comes from down below and the highs come from up above.


I suspect that much of this effect arises from the perceptual disjunction caused by using sources of radically different type and character, accentuating the differences.

If you use a driver that goes from say, 50 hz to 5k...I'll choose a nice vintage WE 728B for Romy's benefit, do you still get this upper/lower effect? I think not.

How much of this localization effect is from woofer leakage above the crossover point, where signals do become more directional...and the interaction between the two drivers straddling the crossover point? Drivers do not shut off precisely at crossover frequencies.

What happens when drivers have different radiation patterns at the same frequency, even independent of the obvious problem of interference between two discretely-spaced sources for the same signal, further subjected to various phase and level shifts brought on by mechanical and electrical rolloff behaviors?

Hard to generalize, I'd say. An I think it is hard to speculate on theory exactly what causes drivers to stand out in the mix. This is why I always tend to argue for examination of specific cases....what works and why, what doesn't work and why.

Beyond that, in accordance with taste, physiological variation, and learned preferences, what works for one listener might not for another. This is not a matter of pure physics.

Would it then be better to optimize the drivers to be dissimilar from one another: to match the radiation patterns of instruments traditionally used within each frequency range the drivers cover?


In concept, this makes sense, no? But then there is the other question of character or texture or whatever you want to call it. This is not a simple issue of frequency response. To the extent that it is a frequency phenomenon, it is a densely complex one.

All I am saying is that the notion of frequency dependent dispersion as raised by Jean Michel's paper is a relevant and often ignored or forgotten factor in system design. It may be that excessive narrowing is more obvious perceptually that excessive widening, hence more complaints about sharp beaming than the obverse.

This topic brings with it the complicated issue of harmonics. A musical note is composed of a fundamental and various harmonics, with different radiation patters at each harmonic. Do we understand how our ears process the cues we get from this complex stimulus?

I would add this dispersion characteristic to what I am calling "character" as a factor worth consideration in design and evaluation of multi-way systems.

Dispersion is also a vital factor in single driver applications. Not all single-drivers fit the same metric and this was a major engineering point for design of various cone profiles... yet another reason why it is silly to lump all single-driver implementations into one category.

My sense is that the whole picture defies analysis, except in reductionistic, partial, and idealized tactical ways. Listening is not only the best way but probably the only valid way to gauge success in driver system matching, particularly when complex musical signals are in play.

Rerurn to Romy the Cat's Site